[go: up one dir, main page]

WO1999066841A1 - Systemes d'imagerie ultrasoniques et procedes associes - Google Patents

Systemes d'imagerie ultrasoniques et procedes associes Download PDF

Info

Publication number
WO1999066841A1
WO1999066841A1 PCT/US1999/013912 US9913912W WO9966841A1 WO 1999066841 A1 WO1999066841 A1 WO 1999066841A1 US 9913912 W US9913912 W US 9913912W WO 9966841 A1 WO9966841 A1 WO 9966841A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
frame
cycle
frames
time
Prior art date
Application number
PCT/US1999/013912
Other languages
English (en)
Inventor
Joseph A. Urbano
Christopher B. Knell
Kevin S. Randall
Andrew J. Wood
Anthony P. Lannutti
Original Assignee
Ecton, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US09/103,878 external-priority patent/US6056691A/en
Priority claimed from US09/103,973 external-priority patent/US6004270A/en
Application filed by Ecton, Inc. filed Critical Ecton, Inc.
Priority to AU46999/99A priority Critical patent/AU4699999A/en
Publication of WO1999066841A1 publication Critical patent/WO1999066841A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
    • G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
    • G01S7/52053—Display arrangements
    • G01S7/52057—Cathode ray tube displays
    • G01S7/5206—Two-dimensional coordinated display of distance and direction; B-scan display
    • G01S7/52066—Time-position or time-motion displays
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/88—Sonar systems specially adapted for specific applications
    • G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
    • G01S15/8906—Short-range imaging systems; Acoustic microscope systems using pulse-echo techniques
    • G01S15/8979—Combined Doppler and pulse-echo imaging systems
    • G01S15/8988—Colour Doppler imaging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
    • G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
    • G01S7/52023—Details of receivers
    • G01S7/52034—Data rate converters
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
    • G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
    • G01S7/52053—Display arrangements
    • G01S7/52057—Cathode ray tube displays
    • G01S7/5206—Two-dimensional coordinated display of distance and direction; B-scan display
    • G01S7/52063—Sector scan display
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
    • G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
    • G01S7/52085—Details related to the ultrasound signal acquisition, e.g. scan sequences

Definitions

  • Ultrasound imaging involves the display of information obtained from reflections of (echoes of) pulses of ultrasonic waves directed into the body. These echoes contain information about the underlying structure of the tissue and blood flow in the region exposed to ultrasound waves.
  • the sophistication and capabilities of ultrasound imaging systems have increased dramatically in recent years, resulting in bulkier and costlier systems which collect an ever increasing volume of data for each imaging session.
  • the explosion in system cost and amount of data collected creates significant problems for the health care industry which must control equipment costs, data storage costs, and labor costs associated with diagnostic testing.
  • Today, many rural and remotely located health care facilities cannot afford to purchase, install or maintain the bulky and costly sophisticated ultrasound imaging systems presently on the market.
  • Fig. 31 illustrates the steps associated with a conventional contrast agent imaging technique for a heart study.
  • the tissue 310 is insonated (i.e., exposed to ultrasound waves) via an ultrasound transducer probe 312 in the absence of a contrast agent, and a single image frame is acquired and stored in a first location within an image frame memory 314.
  • the pre-injection image is acquired at a known point in the heart cycle by using a QRS trigger, such as described in U.S. Patent No. 5,255,683 (Monaghan), incorporated in its entirety herein by reference.
  • the contrast agent is injected into the patient and another single image frame is acquired after the contrast agent has perfused into the heart vessels. This post-injection image is stored in a second location in the image frame memory 314.
  • the two images are then processed by an offline computer 316.
  • the computer aligns the two images as best as possible and performs a subtraction, thereby obtaining a "contrast-only difference image" (not shown).
  • the contrast-only difference image is used for rendering a diagnosis.
  • the subtraction of two ultrasound image frames taken at different points in time, but at the same times in the physiological cycle is referred to generically as a "difference image.”
  • a difference image is typically the absolute value of the image differences.
  • Difference images obtained in contrast agent imaging are slightly different than generic difference images.
  • pixel brightness differences which are less than zero are typically zeroed instead of being assigned an absolute value, since it is presumed that the post-injection image should always be brighter than the pre-injection image. That is, any pixels in the pre-contrast image which are brighter than the corresponding pixels in the post-contrast image are presumed to represent noise.
  • the pre-injection and post-injection images may be further improved via filtering.
  • Filtering is useful in reducing acoustic "speckle.”
  • Acoustic speckle is caused by interference patterns from wavefronts. The interference patterns cause constructive or destructive interference.
  • Acoustic speckle shows up as bright spots and black holes on the image.
  • a conventional speckle reduction technique a plurality of pre- contrast and post-contrast images are obtained at the same point in successive cardiac cycles. Again, the QRS trigger is used as a reference point for identifying when the images should be captured.
  • the plurality of pre-contrast and post-contrast images are then separately filtered using a conventional filtering technique, such as averaging a plurality of such images.
  • the result is a speckle-reduced pre-contrast image and a speckle-reduced post-contrast image obtained at a single point in the cardiac cycle.
  • the resultant filtered images are then subtracted from each other using the offline computer 316.
  • the conventional contrast agent technique there are significant deficiencies in the conventional contrast agent technique. For example, there is a significant time period between acquisition of the pre-injection and post-injection image frames. In this time period, the operator may have moved the probe 312, or the patient may have moved, so that the probe 312 is not in the same exact location during the acquisition of the two images. Furthermore, it is not possible to determine that relative movement occurred during the image acquisition.
  • the offline computer 316 can move the pre-injection image up or down, or left or right, to match up with the post-injection image. Thus, while some movement can be corrected, it is frequently impossible to exactly align the images.
  • Figs. 33A-33C show one conventional alignment scheme which uses a region-of-interest approach to automatically align the images. While the region of interest may be alignable, this does not mean that the rest of the image is necessarily aligned. See Fig. 33C which shows a perfectly overlapping region of interest, but poor overlap in other image regions. If the images cannot be accurately aligned, the resultant subtraction image will be inaccurate.
  • Retesting is sometimes necessary if alignment problems occur during offline processing which cannot be corrected by the computer 316. Retesting is expensive and often impractical. When the patient is suffering from an acute condition, the time lost in the process may cause a significant problem in patient management. Furthermore, the offline alignment process may introduce errors into the heart study, particularly if the operator makes a mistake in aligning the images.
  • the present invention provides systems and methods for obtaining ultrasound imaging data at an adjustable collection image frame rate, as well as at an adjustable acquisition rate.
  • the present invention also provides systems and methods to use a frame correlation process acting on ultrasound imaging data to locate the occurrence of a predetermined event in a physiologic cycle.
  • the present invention further provides systems and methods to reduce speckle in full motion ultrasound image data by filtering image frames of each physiologic cycle across the cycles, instead of filtering from frame to frame.
  • the present invention also provides a scheme for packaging ultrasound imaging data obtained from anatomic structures which have periodic physiologic motion that define successive physiologic cycles.
  • the imaging data is packaged into image loops.
  • Each image loop includes frame data representing a plurality of image frames acquired at spaced time intervals within a physiologic cycle.
  • a communication scheme is provided for transmitting the image loops to a remote location for playback in real-time or pseudo real-time.
  • the present invention also provides a scheme for increasing the effective acquisition frame rate in a medical ultrasound imaging system which acquires imaging data of an anatomic structure having periodic physiologic motion that defines successive physiologic cycles.
  • Systems and methods are also provided for imaging anatomic structures or vessels which have periodic physiological motion that define physiologic cycles.
  • the systems and methods process echoes produced from transmission of ultrasound energy into a subject's body by an image frame acquisition device.
  • a template image is captured at a first time period.
  • the template image comprises image data of a portion of the subject's body acquired at a selected time period during the physiologic cycle.
  • the captured template image is stored.
  • a real-time image is taken at a second time period subsequent to the first time period.
  • the real-time image includes image data of the portion of the subject's body acquired at the same selected time period during the physiologic cycle as the template image.
  • the stored template image and the real-time image are simultaneously displayed on an image display.
  • the simultaneously displayed images have a visually perceptible effect when the real-time image becomes closely aligned with the template image at the same selected time period during the physiologic cycle.
  • a difference image is calculated, stored and displayed.
  • the difference image may be a contrast-only image if a contrast agent has been injected into the imaged region during the second time period.
  • Fig. 1 is a sample ultrasound sector scan image of an anatomic structure and shows parameters of an ultrasound scan
  • Fig. 2 is an electrocardiogram of the human heart
  • Figs. 3A-3C show the structure of image loops in accordance with the present invention
  • Fig. 4 is an image frame memory for storing ultrasound image frame data as a series of image loops shown in Figs. 3A-3C;
  • Fig. 5 is a schematic block diagram of a communication environment for transmitting the image loops from an image acquisition location to a remote location
  • Fig. 6 is a schematic block diagram of a medical ultrasound imaging system in accordance with the present invention which has an adjustable collection image frame rate
  • Figs. 7 A and 7B show two types of image frames which may be captured using the system of Fig. 6;
  • FIG. 8 illustrates a sample data acquisition session when using the system of Fig. 6
  • Fig. 9 is a flowchart of the process for acquiring data when using the system of Fig. 6;
  • Figs. 10A and 10B illustrate imaginary brightness data for a four pixel portion of an image frame
  • Fig. 11 is a table of mean difference values and associated frame rates for use with the system of Fig. 6;
  • Fig. 12 shows a scheme for collecting data representing time between adjacent collection image frames for use with the system of Fig. 6;
  • Fig. 13 is a schematic block diagram of a system for locating the occurrence of a predetermined event in a physiologic cycle in accordance with the present invention
  • Figs. 14A is a graph of a portion of a heart cycle shortly before, during, and shortly after systole which indicates relative motion of the heart being imaged, and also shows a table of corresponding frame correlation coefficients obtained in this region of the heart cycle;
  • Fig. 14B is a table showing corresponding mean first difference values obtained from experimental heart image data for one complete heart cycle;
  • Fig. 15 is a schematic block diagram of a medical ultrasound imaging system in accordance with the present invention which has an adjustable acquisition image frame rate;
  • Fig. 16 is a graph of heart motion vs. time throughout the heart cycle;
  • Fig. 17 is a table of frame rates for different heart cycle intervals for use with the system of Fig. 15;
  • Fig. 18 is another table that may be used in conjunction with the table of Fig. 17 to determine the frame rate of the system in Fig. 15 based on the patient's heart rate;
  • Fig. 19 illustrates a conventional temporal filtering scheme
  • Fig. 20 illustrates a scheme for filtering ultrasound imaging data across physiologic cycles (i.e., cycle to cycle), in accordance with the present invention
  • Fig. 21 is a schematic block diagram of a filter suitable for implementing the cycle to cycle filtering scheme of the present invention.
  • Fig. 22 shows how a four pixel region is filtered in accordance with the cycle to cycle filtering scheme
  • Fig. 23 is a velocity wheel for illustrating how to implement cycle to cycle filtering of Doppler ultrasound image data
  • Fig. 24 shows a simplified example of a conventional temporal filtering scheme for implementing frequency-type image compounding
  • Fig. 25 shows an example of how to implement frequency-type image compounding across a physiologic cycle
  • Fig. 26 is a schematic block diagram of a system for increasing the effective acquisition frame rate in a medical ultrasound imaging system in accordance with the present invention
  • Fig. 27 graphically illustrates the operation of the system of Fig. 26;
  • Fig. 28 is a schematic block diagram of a first system for implementing a pseudo real-time transmission scheme in accordance with the present invention.
  • Fig. 29 is a schematic block diagram of a second system for implementing a pseudo real-time transmission scheme in accordance with the present invention.
  • Figs. 30 A and 3 OB show sample contents of memory buffers in the systems of Figs. 28 and 29 for a plurality of successive physiologic cycles.
  • Fig. 31 illustrates the steps associated with a conventional contrast agent imaging technique for a heart study
  • Fig. 32 illustrates a conventional offline processing scheme for use in a heart study
  • Figs. 33A-33C illustrate display screens used for alignment processing in the conventional offline processing scheme of Fig. 32;
  • Figs. 34A-34B illustrate an alignment processing scheme in accordance with the present invention
  • Fig. 35 is a schematic block diagram of a medical ultrasound imaging system for contrast agent imaging and difference image processing of anatomic structures which have periodic physiological motion that define physiologic cycles in accordance with the present invention
  • Fig. 36 is a schematic block diagram of a difference image processing system in accordance with another embodiment of the present invention.
  • Figs. 37A-37C show the structure of image loops for use with the present invention
  • Fig. 38 is an image frame memory for storing ultrasound image frame data as a series of image loops shown in Figs. 37A-37C.
  • prior art Fig. 1 shows a sample ultrasound sector scan image of an anatomic structure 10, as well as the parameters of the scan.
  • the structure 10 is a heart
  • feature 12 is an outline of a heart chamber.
  • the image is obtained using a conventional transducer 14 which forms a plurality of scan lines or beams 16 that traverse an angular sector 18 in a plane.
  • the scan line data are processed by ultrasound image processor 20 and viewed on display 22 as a succession of image frames, in a conventional manner.
  • the image formed by each successive pass through the angular sector 18 (e.g., 128 scan lines in the example of Fig. 1) defines one image frame.
  • the "image frame rate" of an ultrasound device is the number of image frames obtained per unit time. A typical frame rate may range from 15 to 45 frames per second (FPS).
  • the "image frame period" is the amount of time between adjacent frames.
  • Apex Distance 24 This is the distance from the transducer's transmitter emission point to the region of the image where data collection begins.
  • Scan Line data points (dp) - These are data points along each scan line 16. Each data point ultimately represents a pixel value on the display 22. For illustration purposes, fourteen data points dp r dp are shown along the 128th scan line. A typical scan line 16 actually may have 300-400 data points. Distance between data points (d) - This is the distance between adjacent data points in each scan line 16. ⁇ - This is the angle between adjacent scan lines 16.
  • ⁇ is related to scan line density (i.e., number of scan lines in a frame), also referred to herein as the image frame "resolution.” ⁇ decreases as scan line density increases.
  • the present invention is particularly useful for imaging anatomic structures which exhibit periodic physiological motion wherein the motion defines successive periodic cycles.
  • the heart and lungs are examples of anatomic structures which have periodic cycles. Since the heart cycle (cardiac cycle) is used to illustrate certain features of the present invention, some background on the heart cycle is provided to further understand the invention.
  • Fig. 2 shows a normal electrocardiogram (ECG or EKG) of the human heart.
  • EKG is a scalar representation that shows deflections resulting from atrial and ventricular activity as changes in the magnitude of voltage and polarity with time. The deflections are referred to as "waves.” For example, the first deflection is the P wave.
  • the heart cycle is defined as the period from the beginning of one heart beat to the beginning of the next heart beat.
  • the heart cycle has two important time intervals during each cycle, namely the systole and diastole.
  • diastole the left ventricle fills with blood.
  • systole the left ventricle contracts to pump the blood out of the heart.
  • the so-called “systolic time interval” is the interval from the onset of the QRS interval to the aortic component of the second heart sound. Most of the remaining time of the heart cycle comprises the diastole.
  • the QRS trigger provides a convenient way to detect the onset of systole and is often used to control ultrasound equipment for capturing image data.
  • All ultrasound imaging systems have memory for collecting and storing image data.
  • the image data may be stored as raw data representing the pixels obtained from data points along each scan line 16 shown in Fig. 1 , or the image data may be stored as processed frame data, in the same manner as a frame of video.
  • Each ultrasound system has its own data storage structure and memory structure.
  • the present invention addresses these deficiencies in the prior art by providing a packaging/formatting structure for ultrasound imaging data, and particularly, for ultrasound imaging data obtained from an anatomic structure having periodic physiological motion.
  • the imaging data is stored and packaged or formatted as image loops.
  • Each image loop includes frame data representing a plurality of image frames acquired at spaced time intervals within a physiologic cycle.
  • Fig. 3A shows an image loop 100.
  • the image loop 100 includes a loop header 102 and the frame data 104 for one physiologic cycle.
  • the frame data comprises four image frames acquired during each cycle.
  • the image frame data 104 is packaged as a plurality of frame packets 106j-106 4 , described in more detail below with respect to Fig. 3B.
  • the loop header 102 may be used to identify the loop by number, for example, cycle number 482 in a stream of collected data.
  • the anatomic structure is the heart
  • the physiologic cycle is the heart cycle
  • the image loop 100 includes a plurality of image frames acquired for one physiologic cycle at a predefined time relative to a QRS trigger.
  • the image frames may be acquired from one QRS trigger to the next, or from a few milliseconds after the QRS trigger to a few milliseconds after the next QRS trigger.
  • the frame rate of the acquisition equipment is 30 FPS and the heart cycle is about one second, then there will be about 30 frame packets in each image loop 100 (as opposed to the four shown in Fig. 3A).
  • Fig. 3B shows a frame packet 106.
  • Each frame packet 106 includes a frame header 108 and the scan line data associated with the frame.
  • the scan line data 110 is packaged as a plurality of scan line packets 112,-112 4 , described in more detail below with respect to Fig. 3C.
  • the frame header 108 identifies the frame by number, for example, frame number 3 in the sequence of four frame packets of Fig. 3 A, or frame number 3 in a sequence of 30 frame packets in the heart example described above. Alternatively, the frame numbers may begin with zero. In the heart cycle example, frame zero would be acquired at the QRS trigger or a predetermined time period displaced therefrom.
  • Fig. 3C shows a scan line packet 112.
  • Each scan line packet 112 includes a scan line header 114 and the individual line data 116 associated with the scan line.
  • the scan line header 114 comprising a line number representing a position within a scan sector, such as the last scan line shown in Fig. 1 (scan line 128).
  • the individual line data represents the data points along the scan line, such as dp r dp shown in Fig. 1.
  • the loop header 102, the frame header 108 and the scan line header 114 thus function much like the identification portion or ID tag of a data packet in a digital communication scheme.
  • These headers may also be used to store additional information to assist in processing the image data.
  • the loop header 102 may also include any of the following types of information:
  • Type of physiologic cycle e.g., heart cycle, respiratory cycle
  • Geometric parameters of the ultrasound scan e.g., apex distance 24, distance between data points (d), ⁇ ).
  • the frame header 108 may also include any of the following types of information: (1) Time between adjacent, successively acquired or collected frames (i.e., current and previous frame). This information is particularly important when image frames are acquired or collected at different frame rates. During playback, this information is used to determine how long the frame should persist on the display screen before being replaced by the next frame.
  • Geometric parameters of the ultrasound scan especially if one or more parameters vary on a frame-to-frame basis. If the parameters are the same for all frames, then they should be placed in the loop header to avoid redundancy of information.
  • the image frame is a collection image frame or an interrogation image frame (see discussion below).
  • the scan line header 114 may also include any of the following types of information:
  • One preferred embodiment of the invention for use in collecting image data of a heart cycle synchronizes frame acquisition using the QRS trigger point, wherein each image loop goes from one QRS trigger point (or a predetermined time with respect to the QRS trigger point) to the next QRS trigger point.
  • an image loop may be acquired every predetermined number of seconds. If no triggering event is used, sufficient frames should be collected with each image loop to ensure that each loop includes at least one full physiologic cycle of data.
  • an image loop as defined herein starts at a first predefined time with respect to a predetermined event in a physiologic cycle and ends at a second predefined time with respect to the predetermined event in the physiologic cycle or in one or more successive physiologic cycles.
  • the image loop has a length of one physiologic cycle.
  • the image loop may have a length greater or less than one physiologic cycle, such as one-half cycle, 1 Vz cycles, or two cycles.
  • an image loop may consist of only the systole portion of a heart cycle, or may consist of two full heart cycles.
  • the image loop may start at the beginning of a first physiologic cycle and end at the end of the same physiologic cycle, right before the start of the next physiologic cycle.
  • the image loop is not necessarily acquired from the same physiologic cycle.
  • the image loop for the single physiologic cycle may, for example, be made up of image frames from the second half of a first physiologic cycle and the first half of the subsequent physiologic cycle.
  • the frames in the image loop are still ordered from the start of the physiologic cycle to the end of the physiologic cycle, determined by the collection time of the image frame with respect to the triggering event.
  • the first three image loops may be packaged by the single cycles, as follows:
  • the first three image loops may be packaged as follows:
  • Fig. 4 is an image frame memory 118 for storing ultrasound image frame data as a series of image loops 100.
  • Each image loop 100 includes an image loop header 102 and a plurality of frame packet 106.
  • Image loops By packaging the image data in image loops, the image data becomes very easy to manipulate and to communicate to a remote site. Disparate equipment can easily interpret the image data for display. Image loops simplify memory management of the image data because selected portions of data can be easily identified for storage, retrieval or display.
  • Fig. 5 shows a patient location 120 and a remote location 122.
  • Ultrasound image data is acquired from a transducer 124, and the raw image data is fed into an ultrasound image processor 126 which formats the data into image loops 100, as described above. All of the information associated with an image loop 100 is conventional, and thus no additional explanation is provided on how to capture such information. However, heretofore, such information was not captured and used for packaging or formatting image data in the image loop format.
  • the image loops 100 may be stored locally in the image frame memory 118 and/or viewed on display 128.
  • the image loops 100 may also be communicated to the remote location 122 via communication or transmission medium 130 either directly from the processor 126, or from the image frame memory 118.
  • the image loops 100 Upon receipt at the remote location 122, the image loops 100 are processed by ultrasound image processor 126' and stored locally in the image frame memory 118' and/or viewed on display 128'.
  • image loops 100 One beneficial application of image loops 100 is in telemedicine, wherein the patient location 120 in Fig. 5 is rural or remote diagnostic site, and highly skilled specialists receive and analyze the data at the remote location 122.
  • the communication process can be significantly simplified compared to the prior art.
  • the image loops 100 can be easily transmitted using various communication media, such as the Internet. If the communication medium does not consistently support the full required bandwidth of the real-time data stream of image data, one preferred transmission scheme is to transmit the image loops 100 in pseudo real-time. Two suitable processes for implementing a pseudo real-time transmission scheme are shown in Figs. 28 and 29, described below.
  • the image frame data stored in the image loops 100 may represent raw image data before it is processed by a scan converter, or the image frame data may represent display-type pixel data (i.e., raw image data which was processed by a scan converter).
  • the image loops 100 are also useful in implementing the remaining portions of the present invention, as discussed below. However, their use is optional in the other portions of the present invention.
  • a B-scan technique amplifies and processes the echoes according to their strength and creates a visual display ranging from white for the strongest echoes to varying shades of gray.
  • the display may be referred to as a B-Mode display.
  • the B-mode display presents a two-dimensional display that is updated periodically as determined by the frame rate. Data is typically acquired at the highest image frame rate necessary to fully capture the motion of the fastest moving structures presumed to be present. For a fast-moving structure such as the heart, an acquisition frame rate of more than 30 FPS must be used to accurately represent the motion during the fastest portion of the heart cycle. For a slow-moving structure such as the liver, a lower frame rate may be used, such as less than 15 FPS.
  • the acquisition frame rate is limited by the number of scan lines being obtained, the frequency of the ultrasound transmission signal, and the depth of the scan.
  • the scan depth is an important factor in limiting the frame rate. A deeper depth will result in slower receipt of the echo signal, since the speed of sound introduces an inherent limit on the speed of the transmission and echo signals.
  • the motion of tissue structure is normally periodic. Each period is known as a heart cycle and its length of time is determined by the heart rate.
  • the rate of motion of the structure exposed to ultrasound waves varies at different times within the heart cycle. For example, the rate of motion during systole (i.e., the systolic portion of the cycle) is greater than the rate of motion during diastole (i.e., the diastolic portion of the cycle).
  • a slower sampling rate may be used for the diastolic portion of the cycle.
  • the fixed frame rate may be selected automatically or manually. If the fixed frame rate is selected to be sufficient to accurately represent the motion during the fastest portion of the heart cycle, then the image will be oversampled during the slower portions of the cycle (e.g., the diastolic portion). If the fixed frame rate is selected to be somewhat less than the frame rate necessary to accurately represent the motion during the fastest portion of the heart cycle, then the image will sometimes be oversampled, and will sometimes be undersampled. Oversampling results in the acquisition of redundant (unnecessary) data, which must be stored and processed or transmitted. Undersampling results in the loss of potentially important image information.
  • the present invention provides two different schemes for matching the frame rate more closely with the rate of motion of the anatomic structure.
  • the first scheme matches the collection frame rate more closely with the rate of motion of the anatomic structure, and is described with respect to Figs. 6-12.
  • the second scheme matches the acquisition and collection frame rate more closely with the rate of motion of the anatomic structure, and is described with respect to Figs. 15-18.
  • the first scheme obtains and processes two types of ultrasound images, (1) collection image frames which are "collected” (i.e., captured and stored for immediate or subsequent display or review, and alternatively, for transmission to a remote site), and (2) interrogation image frames which are used to monitor activity in the time between acquisition of collection image frames, and which are not “collected.”
  • the interrogation image frames may be temporarily stored in a memory buffer, as described below.
  • the first scheme uses a frame correlation procedure to determine if and/or when the anatomic structure is exhibiting movement, and adjusts the type of images which are acquired at subsequent time intervals in response to the amount of movement.
  • the interrogation frames effectively sample the anatomic motion, and provide an early warning that the rate of motion is changing. If the rate of motion has changed, then an adjustment should be made to the number of collection image frames that are acquired and collected in a given time period.
  • Fig. 6 is a schematic block diagram of a medical ultrasound imaging system 132 for implementing the first scheme.
  • the system 132 includes an image frame acquisition device 134, an image frame controller 136 (which may part of the acquisition device 134 or which may be separate from the acquisition device 134), an image processor 178, an image frame memory 138, a frame correlation processor 140, and an image display 142.
  • the image frame acquisition device 134 includes a transducer probe 144 and a processor 146.
  • the transducer probe 144 may be any suitable conventional probe containing a transmitter (not shown) for transmission of ultrasound energy into the subject's body, and a receiver (not shown) for receiving the resultant echoes.
  • the output of the image frame acquisition device 134 is connected to the input of the image processor 178.
  • the image processor 178 has a first output connected to a. first input of the image frame memory 138 and to the image display 142 for passing unfiltered or temporally filtered collection image frames to the image frame memory 138 and to the image display 142.
  • the image processor 178 also has a second output connected to a first input of the frame correlation processor 140 for passing collection image frames and interrogation image frames to the frame correlation processor 140.
  • the image processor 178 has an optional third output connected to a second input of the image frame memory 138 for passing interrogation frames to the image frame memory 138.
  • the image frame memory 138 may optionally be connected at its output to an input of the frame correlation processor 140 for passing selected interrogation frames to the frame correlation processor 140.
  • the output of the frame correlation processor 140 is connected to the input of the image frame controller 136.
  • the output of the image frame controller 136 is connected to an input of the processor 146.
  • the acquisition device 134 produces a series of image frames.
  • Each image frame comprises data representing an image of a portion of the subject's body at a given time.
  • the device 134 produces image frames of at least two different resolutions. The image frames at lower resolutions are acquired in less time than the image frames acquired at higher resolutions.
  • the first embodiment of this scheme is described in the context of a dual resolution scheme wherein the collection image frames have a first resolution, and the interrogation image frames have a second resolution.
  • the collection image frames have a higher resolution than the interrogation frames.
  • Image frames produced at the first (higher) resolution are interchangeably referred to herein as "full-resolution image frames" or "full-resolution frames” and preferably represent the maximum resolution of the acquisition device
  • Image frames produced at the second (lower) resolution are interchangeably referred to herein as "low resolution interrogation image frames" or "low resolution interrogation frames.”
  • the processor 146 of the acquisition device 132 thus outputs two different types of image frames, namely, full-resolution frames and low resolution interrogation frames.
  • the processor 146 controls the transducer probe 144 to produce one or the other types of image frames, depending upon the signal that it receives from the image frame controller 136, as discussed in more detail below.
  • the image frame memory 138 receives the full-resolution frames and stores them in a storage portion 148 of the frame memory 138.
  • the low resolution interrogation frames are not stored in the image frame memory 138, since only the current low resolution interrogation frame is needed for the preferred mode of operation.
  • an alternative embodiment of the invention discussed below may use recently acquired low resolution interrogation frames, and thus a temporary storage portion 150 may be provided in the frame memory 138 for this purpose.
  • the full-resolution image frames obtained from the output of the acquisition device 134 are communicated to the image display 142 for real-time monitoring of the image acquisition process.
  • the image processor 178 includes a filter or switch 180 which blocks all interrogation image frames (regardless of their resolution) from being passed to the image display 142 or to the storage 138, and allows all collection image frames (e.g., full-resolution image frames) to pass therethrough.
  • a filter or switch 180 which blocks all interrogation image frames (regardless of their resolution) from being passed to the image display 142 or to the storage 138, and allows all collection image frames (e.g., full-resolution image frames) to pass therethrough.
  • One suitable scheme for implementing the switch 180 is to use an image frame tagger in the processor 146 of the acquisition device 134 to tag or identify each acquired image frame as being either a collection image frame or an interrogation frame. If image data is acquired as image loops 100, the frame header 108 may be used to tag the type of image frame, as discussed above.
  • Fig. 7 A shows a full-resolution frame 152 which contains the maximum possible number of ultrasound scan lines achievable by the particular equipment (128 in this example), and a low resolution interrogation frame 154 which samples only selected scan lines, such as every 4th scan line.
  • the full-resolution frame 152 thus has a greater line density than the low resolution interrogation frame 154.
  • the general operation of the Fig. 6 imaging system 132 is as follows: Successively acquired image frames 152 or 154 are continuously compared using a conventional scheme (e.g., a pixel difference scheme) to determine the amount of difference between the successive images.
  • the amount of difference is expressed as a "frame correlation coefficient.” If there are a large number of differences (i.e., a low frame correlation coefficient), then the "effective frame rate" is set to a fast rate, or is maintained at a fast rate if already set to a fast rate. If there are a small number of differences (a high frame correlation coefficient), then the effective frame rate is reduced, or maintained at a reduced rate if already set to a reduced rate.
  • the goal of the system 132 is to obtain and store as many full-resolution image frames 152 in a given time as is necessary to sufficiently capture important anatomic motion, while minimizing the amount of full-resolution image frames 152 which merely provide redundant information.
  • the "effective frame rate" is determined by how many full-resolution image frames are collected in a given period of time.
  • maximum resolution In this case, no low resolution interrogation frames are obtained, and the entire imaging period is used to collect full-resolution images.
  • the frame rate would thus be the maximum frame rate permissible by the equipment, such as 45 FPS.
  • anatomic motion is very slow, and thus very little change from frame to frame. In this case, it may be acceptable to obtain only 15 FPS.
  • dead time there is so-called "dead time" between capturing of successive full-resolution images 152.
  • low resolution interrogation frames 154 are obtained.
  • the number of low resolution interrogation frames 154 taken between full resolution frames 152 depends upon the length of the dead time and the line density of the low resolution interrogation frame 154. Since a low resolution interrogation frame 154 has a significantly lower line density than a full- resolution frame 152, a plurality of low resolution interrogation frames 154 may be taken in the same time period that it would take to obtain one full-resolution frame 152. In the example of Figs. 7A and 7B, four low resolution interrogation frames 154 can be obtained in the time period required to obtain one full-resolution frame 152, since the capture time is generally proportional to the number of scan lines.
  • the acquisition system 132 is currently collecting full-resolution frames 152 at less than the maximum rate, and low resolution interrogation frames 154 are being collected in the "dead time" between collection of the full-resolution frames 152.
  • the low resolution interrogation frames 154 are processed in the following manner:
  • the frame correlation processor 140 determines the amount of difference between the current low resolution interrogation frame 154 and the last full-resolution frame 152.
  • the frame correlation processor 140 uses all of the scan lines of the low resolution interrogation frame 154 and every 4th corresponding scan line of the full-resolution frame 152. If there are a large number of differences between the current low resolution interrogation frame 154 and the sampled scan lines of the last full-resolution image 152 (i.e., a low frame correlation coefficient), then the image frame controller 136 increases the effective frame rate by causing more full-resolution images to be acquired in a given time period. That is, the FPS is increased, wherein "F" refers to a full-resolution frame 154. If there are no significant differences between the latest low resolution interrogation frame 154 and the last full resolution image 152
  • the frame rate is reduced, or is not changed if it is already at the lowest rate.
  • the acquisition system 132 is currently collecting full-resolution frames 152 at the maximum rate, and no low resolution interrogation frames 154 are being collected, since there is no "dead time" between collection of the full-resolution frames 152.
  • This condition is preferably used upon start-up of the equipment, since no information is available about the motion of the anatomic structure exposed to ultrasound waves. This condition may also occur at any time in the. acquisition process if fast anatomic motion is detected.
  • the full-resolution frames 152 are processed in the following manner:
  • the frame correlation processor 140 determines the amount of difference between the current full-resolution frame 152 and the last full-resolution frame 152. If there are a large number of differences between the two frames 152 (i.e., a low frame correlation coefficient), then the image frame controller 136 maintains the frame rate at the current rate, and thus continues to collect only full-resolution frames 152. If there are no significant differences between the two frames 152 (i.e., a high frame rate correlation coefficient), then the full-resolution image frame rate is reduced, and the system 132 starts obtaining low resolution interrogation frames 154 in the dead time between collection of full- resolution frames 152.
  • the image frame controller 136 adjusts the collection image frame rate in inverse proportion to the most recent frame correlation coefficient.
  • the resolution of the ultrasound image is determined by the number of scan lines per frame.
  • the acquisition device 134 produces the desired image resolution (as determined from the output signal of the image frame controller 136) by causing the transmitter of the transducer probe 144 to transmit the appropriate number of scan lines to obtain the desired resolution.
  • the transducer probe 144 would transmit 128 scan lines to obtain the full- resolution frame 152, and would transmit 128/4 or 32 scan lines to obtain the low resolution interrogation frame 154.
  • the image frame controller 136 outputs a frame trigger signal to cause the acquisition device 134 to obtain a full-resolution image whenever it is deemed necessary based on the frame correlation results.
  • each successively acquired low resolution interrogation frame 154 is compared to the last acquired full- resolution frame 152.
  • each low resolution interrogation frame 154 is compared to the previously acquired image frame, regardless of whether that image frame is a full-resolution image frame 152 or a low resolution interrogation frame 154.
  • a succession of currently acquired low resolution interrogation frames 154 may be compared to previously acquired low resolution interrogation frames 154.
  • This scheme is not preferred, because small changes between successive low resolution interrogation frames 154 may not trigger any change to the frame rate, when, in fact, the current low resolution interrogation frame 154 has become very different from the last acquired full-resolution image frame 152.
  • the temporary storage portion 150 of the image frame memory 138 is used to store and retrieve a previously acquired low resolution interrogation frame 154, when needed (i.e., whenever the last two acquired image frames are low resolution interrogation frames 154).
  • the system 132 may be programmed to collect a full-resolution frame 152 at periodic time intervals, even if no instruction signal or trigger signal is received from the image frame controller 136 or other controlling device to do so. This may occur if the frame correlation processor 140 detects no change from image to image, and the image frame controller 136 sets the effective frame rate to a very low rate. Periodic collection of full-resolution image frames 152 may be desirable for system integrity, even though the additional full-resolution image frames 152 may be redundant.
  • Fig. 8 illustrates a sample data acquisition session. At start-up, full- resolution images are obtained.
  • the samples F2 and F3 taken at times tl and t2 show very high correlation, indicating that the frame rate can be significantly lowered, while still capturing sufficient information.
  • the frame rate is lowered to obtain one full resolution sample every 4th time period.
  • Interrogation frames are taken during the dead time (t3-t5).
  • the interrogation frames 11-14 are actually taken during the dead time between t2 and t3, the interrogation frames 15-19 are actually taken during the dead time between t3 and t4, and so on.
  • the interrogation frames are bunched together to simplify the illustration.
  • a comparison of the interrogation frames 11-112 obtained during the dead time (t3-t5) with full-resolution frame F3 indicate that no change should be made to the frame rate of once every 4th time period. Thus, this frame rate continues unchanged.
  • a comparison of the interrogation frames 113-125 obtained during the dead time (t7-t9) with full- resolution frame F4 indicate that no change should be made to the frame rate of once every 4th time period. Thus, this frame rate continues unchanged.
  • the last one or two interrogation frames show very low correlation with the last acquired full-resolution image F5, meaning that the anatomic structure is now moving faster. As a result, the frame rate is increased to the original maximum rate.
  • Fig. 9 shows a self-explanatory function-type flowchart of the process described above.
  • the scheme illustrated in Fig. 8 has two different frame rates, namely, a first (fast) frame rate of one full-resolution frame per time period, and a second (slow) frame rate of one full-resolution frame per four time periods.
  • One suitable two-speed embodiment may have a fast frame rate of about 40 FPS (one full-resolution frame 152 is collected about every 25 ms), and a slow frame rate from about 10 FPS to a frame rate of about 30 FPS (one full-resolution frame 152 is collected about every 100 ms to about every 33 ms).
  • the scope of the invention also includes schemes having more than two different frame rates.
  • Figs. 10A and 10B illustrate the concept of frame correlation and show, in very simplified terms, the function of the frame correlation processor 140.
  • mean difference avg ⁇ abs ⁇ f(n) - f(n-l) ⁇ or ⁇
  • f(n) is the current image frame and f(n-l) is the previous image frame.
  • a low level threshold may be applied to the image frames to remove all pixels below a chosen intensity level.
  • Fig. 11 shows one method of using the mean differences for setting the collection image frame rate.
  • the image frame controller 136 may include a table of mean differences and collection image frame rates associated with each mean difference.
  • a threshold mean difference is used to set the frame rate. If more than two frame rates are used, then different ranges of mean differences may correlate to different respective frame rates.
  • the values shown in Fig. 11 are for illustration purposes, and do not represent values associated with an embodiment of the present invention.
  • Suitable normalized mean difference thresholds for a three frame rate embodiment, based on experimental data wherein relative brightness levels of pixels range from 0 to 127, are as follows:
  • the SLOW frame rate may be used for normalized values less than 9 or less than 10
  • the FAST frame rate may be used for values greater than or equal to 9, or greater than or equal to 10.
  • Fig. 12 shows one preferred scheme for storing the time interval between adjacent (successively acquired) full-resolution image frames 152.
  • the time interval thus effectively represents the instantaneous collection image frame rate. During playback, this information is used to determine how long the image frame 152 should persist on the display screen before being replaced by the next image frame 152.
  • the time intervals shown in Fig. 12 are represented in number of time periods, and match the image frames 152 of Fig. 8.
  • the time between adjacent frames may be stored as part of the frame header 108, as discussed above.
  • the acquisition device 134 preferably operates by setting the effective collection image frame rate based on the input from the image frame controller 136, and then automatically acquiring interrogation frames 154 during dead times.
  • the interrogation frames 154 are preferably acquired at the fastest rate possible to maximize usage of the dead time.
  • the number of interrogation frames 154 that can be acquired in the dead time depends upon its resolution (i.e., scan line density).
  • the 4: 1 ratio example of Figs. 7A and 7B provides an acceptable compromise between resolution and number of interrogation frame samples 154 obtainable during the dead times.
  • the resolution of the collection image frame is either the same or lower than the resolution of the interrogation image frame.
  • the resolutions of the two images are the same.
  • the implementation of this embodiment is generally similar to the first embodiment, except for the following differences:
  • each image frame must be tagged or otherwise identified as being either a collection image frame or an interrogation image frame, so that the succession of image frames output from the acquisition device 134 may be processed, filtered and collected in the same manner as in the first embodiment.
  • the interrogation image frames must be identified so that they are not "collected.” Since the resolutions of the two types of image frames are the same, the acquisition image frame rate remains constant, whereas the collection image frame rate changes depending upon the latest output of the frame correlation processor 140.
  • the second embodiment provides the same advantages as the first embodiment in reducing the volume of image data that must be collected while providing an adjustable collection image frame rate scheme which improves performance over conventional acquisition systems which acquire and collect image frames at a fixed frame rate.
  • the frame correlation processor 140 may use more than one previous image frame to determine the correlation coefficient, and different weightings may be applied to the image frames. These techniques are well known and thus are not described in further detail herein.
  • the system 132 may be designed to always collect interrogation frames 154 between full-resolution frames 152, even at the fastest frame rate setting.
  • the number of interrogation frames 154 taken between full-resolution frames 152 would vary according to the frame rate.
  • the resolution of the interrogation frames 154 may be varied for selected dead times, or within each dead time.
  • An interrogation frame could have a scan line density similar to a full resolution image. However, this would degrade system performance because the number of interrogation frames that can be collected during the "dead time” would be reduced, and thus less sampling would occur during the dead time. This would cause a delay in the frame rate change, resulting in either loss of important image data or collection of unnecessary image data.
  • Fig. 13 is a schematic block diagram of an imaging system 156 which may be used to replace a physiological monitor. Fig. 13 is described in the context of a QRS trigger detection device, although the system 156 may be used for detecting points in other types of physiologic cycles which are characterized by motion changes.
  • the system 156 includes the same image frame acquisition device 134, image frame memory 138 and frame correlation processor 140 described above.
  • the frame correlation processor 140 receives a stream of successively acquired image frames, consisting of collection image frames, such as full-resolution frames 152, or combinations of collection image frames and interrogation frames.
  • the frame correlation processor 140 determines the difference between the current image frame and at least one preceding image frame and outputs a frame correlation coefficient based upon the difference, in the same manner as described above.
  • the correlation coefficients output from the frame correlation processor 140 are input to a QRS trigger detector 158 which determines the QRS trigger therefrom.
  • the time interval associated with the QRS trigger is identified as the instance of time when the frame correlation coefficient falls below a predeterrnined value, or instance of time when the rate of change of the frame correlation coefficient becomes greater than a predeterrnined value.
  • Fig. 14A is a graph of a portion of a heart cycle shortly before, during, and shortly after systole which indicates relative motion of the heart being imaged.
  • Fig. 14A also shows corresponding frame correlation coefficients obtained in this region of the heart cycle. As relative motion increases, the frame correlation coefficient decreases, and vice- versa. The frame correlation coefficients are relatively high and stable, until the onset of systole when they begin to decrease. Likewise, the frame correlation coefficients are relatively low during systole, and begin to rise again as systole ends.
  • the QRS trigger may be located by detecting when the frame correlation coefficients begin to decline, since the QRS trigger is associated with the onset of systole.
  • the QRS trigger may be used to calculate the heart rate, or to control conventional acquisition equipment which currently requires a QRS trigger obtained from a conventional physiological monitor.
  • the QRS trigger detector 158 may use the mean difference values instead of the frame correlation coefficients.
  • Fig. 14B shows corresponding mean first difference values obtained from experimental heart image data for one complete heart cycle. These values correspond to calculations based on pixels having relative brightness levels ranging from 0 to 127. These experimental values also illustrate how the mean difference values discussed above for the experimental data of actual measured pixels may be used to select the appropriate frame rate. Referring to Fig. 14B, the mean difference values at the start of systole reach a peak of 10, thereby providing a readily identifiable indicator of this stage in the heart cycle.
  • the present invention also provides a second scheme for matching the acquisition and collection image frame rates more closely with the resolution needs of the underlying structure.
  • the acquisition and collection image frame rates are the same in this scheme, and are referred to hereafter as "the frame rate.”
  • the frame rate is set based on the expected rate of tissue motion.
  • the second scheme also uses the acquisition device 134 and the image frame controller 136 of the first scheme. However, the second scheme does not perform frame correlation and does not collect interrogation frames. Instead, a physiological monitor monitors a cyclical physiological parameter associated with a physiologic cycle, and detects a predetermined event in the physiologic cycle.
  • the image frame controller 136 functions as a frame rate controller to select the frame rate based upon the current point in time of the physiologic cycle with respect to the detected event.
  • the frame rate may also be varied based upon the latest measurement of the total time of the physiologic cycle. This scheme thus estimates or predicts the expected rate of tissue motion during each phase of the cycle and sets the frame rate based on the expected rate of tissue motion.
  • the physiologic cycle is the heart cycle
  • the cyclical physiological parameter is an electrocardiogram
  • the detected event is the QRS trigger in the heart cycle.
  • image frame controller 136 receives a signal from an EKG monitor, determines the predetermined points in time of the heart cycle relative to the QRS trigger, and sets the frame rate with respect to the current point in time.
  • the frame rate may also be adjusted if the heart rate changes during the image acquisition session. For example, if the heart rate becomes very fast, such as during a treadmill stress test, the frame rate should be increased for all points in time in the heart cycle, to ensure that sufficient image data is collected.
  • Fig. 15 is a schematic block diagram of a sample system 158 for implementing the second scheme.
  • the system 158 is similar to the system 132 of Fig.
  • image frame controller 136 receives a signal from EKG monitor 160, instead of from the frame correlation processor 140.
  • Fig. 16 shows a graph of heart motion vs. time throughout the heart cycle, as well as a two-frame rate scheme implemented by the system 158.
  • the heart rate in Fig. 16 is about 55 beats per minute (BPM).
  • a fast frame rate is used during systole where a large amount of motion occurs.
  • a slow frame rate is used during the remaining portions of the heart cycle, including during diastole where less motion occurs than during systole.
  • Normative patient data may be used to predict the approximate start and end points of systole and diastole relative to the QRS trigger for the current patient.
  • EKG data may be used to determine the exact location of such points for the current patient.
  • Fig. 17 shows a more sophisticated scheme wherein the heart cycle is divided into a plurality of intervals or segments, such as 32 intervals, and a frame rate lookup table is used to select the desired frame rate for the current portion of the cycle.
  • the QRS trigger may be used to track the location in the cycle. Again, assumptions are made for a particular individual based on population data of motion rates at different times between QRS triggers.
  • QRS triggers from a plurality of heart cycles are obtained.
  • the heart cycles are averaged to obtain an average cycle time.
  • the average cycle time is then divided into evenly spaced time intervals.
  • the time intervals are dynamically adjusted if the heart rate changes.
  • the time intervals may also be used in the two frame rate embodiment. For example, intervals 1-7 may be the fast rate, and intervals 8-32 may be the slow rate.
  • Fig. 18 shows another table that may be used to determine the frame rate based on the heart rate, in conjunction with the table of Fig. 17.
  • An average heart rate may be assigned a multiplier factor of 1.0.
  • the multiplier factor would otherwise be proportional to heart rates which are above or below an average heart rate.
  • the values shown in Fig. 18 merely illustrate the concept of adjusting the frame rate, in part, based on the heart rate, and do not represent preferred values. Actual values depend upon the amount of image data that the operator wishes to obtain.
  • the time interval between adjacent image frames 152 is stored, in the same manner as the first scheme.
  • the time information may be stored in the frame loop header when packaging the image frames 152 in image loops 100.
  • Fig. 19 illustrates a simple frame-to-frame temporal filtering scheme, as applied to a four-pixel region of an ultrasound image wherein each pixel is represented as a brightness number from 1-9. In this scheme, the pixels of the current frame are modified to be the average of the current pixel values and the pixel values of the previous image frame.
  • this technique may be adequate in reducing speckle, without affecting the quality of the acquired data.
  • the image is moving slowly between frames 1 and 2 in the region of interest, since only one pixel changed, and only by one brightness level.
  • this technique reduces speckle, but also degrades the image, resulting in an image that does not accurately represent the actual state.
  • the image is moving fast between frames 2 and 3, since all of the pixels changed, and by large amounts.
  • the resultant pixel numbers do not accurately represent the actual image and the resultant image becomes blurred.
  • the large change in pixel values does not indicate, per se, that the image is moving fast. Rather, it is an indication that the image at the time of capture of frame 3 is very different from the image at the time of capture of frame 2, and thus it can be inferred that this resulted from a quick movement in the anatomic structure.
  • Sophisticated temporal filtering techniques may use three or more frames and/or variable weightings and may filter recursively, but still suffer from the problem illustrated in Fig. 19.
  • the present invention applies the temporal filter across a sequence of physiologic cycles, instead of frame-to-frame.
  • a physiologic cycle is divided into discrete time periods, and corresponding time periods from cycle to cycle are filtered. If the scheme of the present invention is applied to the heart, the QRS trigger may be used as the starting point of each heart cycle.
  • the QRS trigger is used as a reference point for obtaining a single pre-contrast image and a single post-contrast image at a predetermined point of interest in the cardiac cycle. The two images are then compared or subtracted from each other. To reduce any speckle associated with the pre-contrast and post-contrast images, some conventional systems obtain a plurality of pre-contrast and post-contrast images at the exact same point in successive cardiac cycles. Again, the QRS trigger is used as a reference point for identifying when the image for each cycle should be captured.
  • the plurality of pre- contrast and post-contrast images are then separately filtered using a conventional filtering technique, such as averaging a plurality of such images.
  • a conventional filtering technique such as averaging a plurality of such images.
  • the result is a speckle-reduced pre-contrast image and a speckle-reduced post-contrast image obtained at a single point in the cardiac cycle.
  • the resultant filtered images are then compared or subtracted from each other.
  • This conventional technique is referred to herein as "single image cycle-to-cycle filtering.” This technique does not provide image frames of the entire current physiologic cycle for full motion analysis of the cardiac cycle. That is, this conventional technique provides a filtered static image at only one point in the cardiac cycle. Thus, full motion analysis of the cardiac cycle cannot be performed using this conventional technique.
  • the scheme of the present invention provides full- motion speckle-reduced image data.
  • Fig. 20 shows a simplified scheme wherein adjacent time periods for adjacent cycles are filtered in the same manner as in Fig. 19 (that is, by averaging pixel brightness values for adjacent frames).
  • four image frames are taken during each periodic physiologic cycle, such as the heart cycle.
  • frames 1 and 2, 2 and 3, 3 and 4, and so on would be filtered.
  • filtering is performed at the same point in each cycle, on a cycle-to-cycle basis or "across" the physiologic cycle.
  • Image (frame) acquisition is continuously resynchronized to the QRS trigger event. Any frame that had been in progress when the trigger occurs is discarded and a new frame is begun.
  • filtering is performed on the frames obtained at the following time periods:
  • FRAMES 1 and 5 (see example of four pixel region in Fig. 20)
  • FRAMES 2 and 6 FRAMES 3 and 7
  • FRAMES 4 and 8
  • the preferred filter is an nth-order infinite impulse response (IIR) filter.
  • IIR infinite impulse response
  • y(n) b 0 x(n) + b, x(n-l) + ... t ⁇ x(n-N) + a, y(n-l) + a 2 y(n-2) + ... a M y(n-M)
  • Fig. 21 is a schematic block diagram of a first order IIR filter 162 suitable for implementing the cycle to cycle filtering scheme of the present invention.
  • the filter 162 includes a filter function block 164 and an image frame memory 166.
  • the output of the filter function block 164 is connected to the input of the image frame memory 166 and to an image frame display (not shown).
  • the filter function block 164 has a first input for receiving image frame data x(n) from an image frame acquisition device (not shown), and a second input for receiving image frame data y(n-l) stored in the image frame memory 166.
  • the filter 162 has the following function:
  • the constant "c" is set to zero whenever it is desired to pass through the current frame, such as when the filter must be reset, as discussed in more detail below.
  • the value of "c” is set by an image frame acquisition device (not shown).
  • the top line in Fig. 22 shows the new, incoming image data x(n) at times n+1 and n+2.
  • FIG. 22 shows the contents of the image frame memory buffer of memory 166 at times n, n+1 and n+2.
  • the image data represents the data at the same point in the physiologic cycle.
  • n+1 is not the next acquired image frame, as would be the case in a temporal filter, but is the image frame of the next cycle at the same point in the cycle as the image frame n.
  • the lower right pixel of the frame in the memory buffer is 2.
  • the corresponding pixel of the next new frame is 0.5.
  • the new pixel value is averaged with the latest pixel value in the memory buffer (now 1.5), and the resultant new value in the memory buffer becomes 1.0.
  • a number of different conditions may require that the constant "c" be set to zero so as to reset the frame memory 166.
  • the frame memory 166 When the frame memory 166 is reset, the current frame becomes the new frame in the memory buffer, and the frame history information is lost. After resetting, the filtering process is restarted by returning the constant "c" to the previously set value.
  • One condition that may require resetting is if the QRS trigger becomes lost while collecting heart images. Once the trigger is relocated, new values should be loaded into the filter's memory buffer.
  • Another condition that may require resetting is if a mismatch of image frames occurs, as a result of either an extra frame or a shortage of frames in the current cycle. This may be caused by a sudden change in the heart rate or by the QRS trigger occurring close to the start of a new frame.
  • the system may be programmed to initially ignore the extra or missing frame, passing through the frames in the memory buffer unfiltered by temporarily setting the constant "c" to be 1, thereby making y(n) equal to y(n-l).
  • Another alternative is to match up the times of new frames as best as possible with the frames in the memory buffer, ignoring the extra or missing frame.
  • One preferred technique for deciding when to reset the filter is to perform a frame correlation procedure using the current image frame and the image frame in the memory buffer. If the current image frame is below a predetermined threshold, then the filter would be reset for the particular frame period, or perhaps for all frame periods. This technique is based on the principle that frames acquired at the same point in each cycle should not vary significantly from cycle to cycle, and thus should be highly correlated (i.e., high correlation coefficient). If the current frame is very different from the history of past image frames, then it is likely that a significant change has occurred in the anatomic structure, such as when the operator moves the probe, and past image data should be deleted.
  • nth-order IIR filter is the preferred filter for the present invention
  • other filters may be used, and are within the scope of the invention.
  • Other possible filters include linear filters such as finite impulse response filters, and nonlinear filters which take the median, minimum or maximum of a plurality of data values.
  • the cycle to cycle filtering scheme may be easily implemented using the image loops 100, since the frame header 108 already identifies the frame by number.
  • the frames of cycle 1 would be packaged into image loop 1 , and the four frames would be identified in the frame headers of image loop 1.
  • the frames of cycle 2 would be packaged into image loop 2, and the four frames would be identified in the frame headers of image loop 2.
  • the corresponding frames of successive physiologic cycles which must be matched up for filtering are thus readily identifiable, even if the image loops 100 are stored out of sequence.
  • the frames within each loop are given successive frame numbers beginning with 0 or 1.
  • the frame numbers for cycles 1-3 have repeating sequence numbers of 1-4, instead of numbers 1-12. It is not necessary to use image loops 100 to implement the cycle to cycle filtering scheme.
  • the raw image data is preferably stored in an image frame memory 138 for playback at other times, or for transmission to another site for playback at the other site.
  • the frame memory 138 is not shown in the figures related to this feature.
  • the cycle to cycle filtering scheme may also be used to process Doppler ultrasound image data.
  • Doppler velocity-mode ultrasound shifts in frequency between emitted ultrasonic waves and their echoes are used to measure the velocity of moving objects, based on the principle of the Doppler effect. This technique is typically used to examine cardiovascular blood flow.
  • Another Doppler ultrasound technique is called "color flow imaging" which typically uses a “velocity mode,” but may also use a "color power mode.”
  • the pixel values represent direction and velocity of the anatomic structure, or more typically, blood, at the location represented by the pixel, instead of brightness or intensity as in grayscale ultrasound. Each velocity value or ranges of velocity values (and direction) are color- coded to a different color and shade.
  • the velocity values and their direction are displayed on a screen as a multicolored image.
  • the power i.e., magnitude
  • the power is an indication of the strength of blood flow or fluid flow at a particular location.
  • Each power value or ranges of power values are color-coded to a different color and shade.
  • the power values are displayed on a screen as a multicolored image.
  • Temporal filtering is typically performed in Doppler color flow imaging in both velocity and power modes as it is done for normal gray-scale structure imaging modes (e.g., B-mode). The effect is to produce a less noisy, more uniform and aesthetically pleasing image.
  • the temporal filtering algorithm is typically modified to persist velocities as they decrease in magnitude while passing through (not filtering) velocities that have a higher magnitude (or a different direction) than the present state of the filter. This is the so- called "fast attack" persistence method.
  • Cycle to cycle filtering may be applied to both of these Doppler imaging modalities as well.
  • the filtering scheme In the Doppler power mode, the filtering scheme is exactly the same as described above for B-mode imaging.
  • Doppler color velocity mode velocity direction must be considered, primarily to account for "aliasing" of the flow velocity. Aliasing results in an erroneous representation of the true velocity of the blood flow due to sampling at a rate which is less than twice the highest Doppler shift frequency caused by echoes from moving blood.
  • the mean Doppler shift frequency just exceeds the sampling rate, the velocity will appear to have changed direction. That is, the velocity will wrap around the other side of the velocity wheel (from Positive Vmax to Negative Vmax), as shown in Fig. 23.
  • the aliasing effect can be addressed in several ways.
  • One way is to use the fast attack persistence method described above. This is not recommended, however, because it is expected that the velocities within each image frame taken from the same point in the physiologic cycle will be very close in value. Therefore, a standard filtering approach is better.
  • the fast attack persistence method assumes that flow is pulsatile, not constant.
  • a preferred technique to address the aliasing effect is to "wrap" the data values the shortest distance around the velocity wheel. This is easiest to illustrate in the case of a two-sample average.
  • the present invention may be used in the prior art area of "image compounding.”
  • image compounding multiple frames of image data are acquired with different imaging parameters, such as "transmit frequency,” and then filtering or otherwise combining this set of images to create a compound image frame.
  • the transmit frequency and/or receiver bandpass filter is varied alternately from frame to frame and then temporal filtering is applied across frames.
  • Fig. 24 shows a simplified example of a conventional temporal filtering scheme for implementing frequency-type image compounding.
  • the cycles and frames of Fig. 24 correlate with those of Fig. 20.
  • the frequency is toggled between X and Y in successive frames. Since different frequencies cause a variation in the speckle pattern, this technique enhances the speckle reduction performance of the standard temporal filter.
  • this scheme also suffers from an inability to accurately represent fast moving anatomic structures.
  • the frame obtained by temporally filtering frame 2 and frame 3 would not accurately represent the structure, since significant motion has occurred between the acquisition times of the two frames.
  • Fig. 25 shows an example of how to implement frequency-type image compounding across a physiologic cycle.
  • the frames in Fig. 25 correspond to the frames in 520.
  • the frequency is toggled between X and Y for each physiologic cycle.
  • the pixel processing in Fig. 25 is the same as in Fig. 20.
  • the cycle to cycle filtering scheme may be used to increase the effective frame rate in a medical ultrasound imaging system, and particularly, in a system which acquires imaging data of an anatomic structure having periodic physiological motion, wherein the motion defines successive physiologic cycles.
  • Fig. 26 shows a system 168 for increasing the effective frame rate.
  • the system 168 includes a physiological monitor 170, an image frame controller 172, an image frame acquisition device 174 (which is generally similar to the acquisition device 134 of Figs. 6 and 15, except as distinguished below), and a filter 176.
  • the output of the monitor 170 is connected to the input of the image frame controller 172.
  • the output of the image frame controller 172 is connected to a control input of the acquisition device 174.
  • the outputs of the acquisition device 174 are connected to the filter 176.
  • the output of the filter 176 may be connected to an image display (not shown) for viewing the final processed image frames.
  • the outputs of the acquisition device 174 and/or the output of the filter 176 may also be connected to an image frame memory (not shown) for storing either the raw, or processed image data.
  • the monitor 170 monitors a cyclical physiological parameter associated with the physiologic cycle, and outputs a signal upon the occurrence of a predetermined event in the physiologic cycle.
  • the physiological monitor 170 would be an EKG monitor 160, such as shown in Fig. 15.
  • the image frame acquisition device 174 acquires imaging data at a fixed frame rate.
  • the device 174 has a processor (not shown, but generally similar to processor 146 of Figs. 6 and 15) for processing echoes produced from transmission of ultrasound energy into a subject's body.
  • the acquisition device 174 produces at least a first and a second series of image frames for each physiologic cycle.
  • Each image frame comprises data representing an image of a portion of the subject's body taken at a known time with respect to the predetermined event detected by the monitor 170.
  • Fig. 27 graphically illustrates the function of the image frame controller
  • Fig. 27 shows a scheme which acquires five image frames during each periodic physiologic cycle.
  • the frame numbers in this example correspond to the frame number scheme used in image loops.
  • the heart cycle is used in this example, although the illustrated waveform is a generic cyclical pattern.
  • the top graph in Fig. 27 shows a first series of timing intervals for acquiring image frames
  • the bottom graph in Fig. 27 shows a second series of timing intervals for acquiring image frames.
  • the frame numbers correspond to the timing intervals.
  • the second series of timing intervals are spaced between the first series of timing intervals.
  • both the first series and second series have the same frame rate of five frames per cycle.
  • the image frame controller 172 causes the second series of image frames to be acquired at times between the acquisition of the first series of image frames in an alternating manner.
  • the alternating scheme functions as follows: CYCLE 1 : Collect frames 0-5 from first series. CYCLE 2: Collect frames 0-5 from second series.
  • CYCLE 3 Collect frames 0-5 from first series.
  • CYCLE 4 (not shown in Fig. 27): Collect frames 0-5 from second series. The data collection continues in this manner, alternating between the two series.
  • the circled frames in Fig. 27 show the frames that were acquired.
  • the timing for frame acquisition may be synchronized to the QRS trigger. For example, acquisition of the first series of image frames may begin exactly at the QRS trigger, and acquisition of the second series of image frames may begin after the QRS trigger, such as at a time halfway between the acquisition time of frames 0 and 1 in the first series of images.
  • U.S. Patent No. 5,099,847 discloses a frame multiplying scheme which acquires image frames in a manner conceptually similar to Fig. 27.
  • the image frames are displayed as an interleaved sequence of frames, thereby providing an effective display frame rate which is double that of the acquisition device frame rate, in the example given in Figs. 1 and 2 of this patent.
  • Other multiples of the initial frame rate are possible by collecting additional series of frames, offset from the first and second series by other non-overlapping periods.
  • the scheme of the present invention does not use the image frames to form an interleaved display. Instead, the image frames for successive cycles are subjected to filtering across physiologic cycles, resulting in a new set of image frames which have the same frame rate as the original acquisition device 174, but which incorporate information between the first series of frames that would have been missed without this scheme.
  • the scheme of the present invention provides an improved, filtered version of the extra information.
  • the following groupings of image frames for the first three cycles are filtered:
  • Frame 0 of cycle 1 Frame 0 of cycle 2
  • Frame 0 of cycle 3 Frame 1 of cycle 1
  • Frame 1 of cycle 2 Frame 1 of cycle 3
  • Frame 2 of cycle 3 Frame 2 of cycle 3
  • the filter 176 thus filters successive image frames obtained at the same point in time within each physiologic cycle.
  • the filter is applied to each image frame within the entire current physiologic cycle, thereby deriving full motion image data.
  • the filter treats the second series of image frames as being located at the same points in time as their respective adjacent first series of image frames.
  • the resultant full motion image data has a higher effective acquisition frame rate than the fixed frame rate of the acquisition device 174.
  • the filter 176 is an nth-order IIR filter function, such as the first order IIR filter 162 shown in Fig. 21.
  • the concept shown in Fig. 27 may be expanded to collect additional image frame series by using additional physiological event trigger delay times, such as
  • QRS trigger delay times.
  • the frame rate boosting scheme described herein may also be used in conjunction with the adjustable acquisition frame rate scheme of Figs. 6-18 to further boost the effective frame rate, while minimizing the collection of redundant information.
  • one advantage of packaging imaging data in image loops 100 is that the image data may be easily transmitted using various communication media, such as the Internet. If the communication medium does not consistently support the full required bandwidth of the real-time data stream of image data, one preferred transmission scheme is to transmit the image loops 100 in pseudo real-time. In this manner, the viewer is presented with temporally accurate image sequences.
  • "Real-time transmission,” in the context of the pseudo real-time transmission scheme thus refers to the ability to display the image data remotely as fast as it is acquired at the patient location site, even though there may be a finite and consistent delay between the viewing time and the acquisition time due to the time required to package and output the image data or due to delays in the transmission medium.
  • Fig. 28 shows a schematic block diagram of a preferred embodiment of a system 200 for implementing a pseudo real-time transmission scheme.
  • the system 200 is described in the context of heart images acquired in synchronization with an
  • the system 200 may be used for transmitting any type of image data of an anatomic structure having periodic physiological motion, wherein the motion defines successive physiologic cycles and the motion is synchronized to a periodic physiologic event in the cycle. To ensure that the ultimately displayed image data is accurate, the system 200 should have the ability to transmit the image data in at least near real time.
  • image frames are tagged with actual frame-to-frame timing information, as well as a frame number corresponding to a particular phase of the cardiac cycle. That is, the frame number is resynchronized at each new cardiac cycle.
  • the system 200 includes an image frame acquisition device 202 at a patient location, similar to the acquisition devices disclosed in Figs. 5 or 6.
  • the image frames are preferably packaged as image loops 100, described above with respect to Figs. 3A-3C and Fig. 4.
  • the output of the acquisition device 202 is a series of image loops 100.
  • the image loops 100 may be packaged using image frames acquired from only a single physiological cycle, or from portions of adjacent physiologic cycles.
  • the output of the acquisition device 204 is communicated to a first ping-pong memory 204 at the patient location.
  • the first ping-pong memory 204 has two buffers 206 and 208 for temporarily storing consecutively acquired image loops.
  • the buffers are referred to herein as loop A and loop B, respectively.
  • the first ping-pong memory 204 also includes an input multiplexer or input switch 210 connected at its input to the output of the acquisition device 204, and connected at a first and second output to respective inputs of the buffers 206 and 208.
  • the first ping-pong memory 204 also includes an output demultiplexer or output switch
  • the system 200 includes a second ping-pong memory 214, similar to the first ping-pong memory 204.
  • the second ping-pong memory 214 has two buffers 216 and 218 for temporarily storing consecutively acquired image loops.
  • the buffers are referred to herein as loop C and loop D, respectively.
  • the second ping-pong memory 214 also includes an input multiplexer or input switch 220 connected at its input to the transmission medium 212, and connected at a first and second output to respective inputs of the buffers 216 and 218.
  • the second ping-pong memory 214 also includes an output demultiplexer or output switch 222 connected at its input to the respective outputs of the buffers 216 and 218, and connected at its output to a remote viewing station or remote image display 224 for providing the real-time or pseudo real-time image of the subject's body.
  • the ping-ping memories 204 and 214 are conventional, and thus are not described in further detail.
  • the switch 210 Upon initiation of image data acquisition, the switch 210 is connected to the loop A buffer for filling the loop A buffer with image data.
  • the switch 220 is switched to the loop C buffer, and the loop C buffer receives the image loop transmitted from the loop A buffer in step (3) above.
  • the switch 220 switches to the loop D buffer to allow the loop D buffer to receive the next transmission of image data, and the switch 222 is switched to the loop C buffer, thereby allowing the contents of the loop C buffer to be displayed on the image display 224.
  • the image loop in the loop C buffer is continuously played (i.e., looped) until the loop D buffer is filled with a new image loop.
  • the switch 220 switches to the loop C buffer to allow the loop C buffer to receive the next transmission of image data, and the switch 222 is switched to the loop D buffer, thereby allowing the contents of the loop D buffer to be displayed on the image display 224.
  • the image loop in the loop D buffer is continuously played (i.e., looped) until the loop C buffer is filled with a new image loop. Steps (2) and (3) are continuously repeated until data transmission is stopped.
  • the loop A and loop B buffers alternately function as “capture buffers” and “transmit buffers,” whereas the loop C and loop D buffers alternately function as “receive buffers” and "display buffers.” If the transmission medium bandwidth fully supports the image data rate, then the display will be a true real-time representation of the image data and no "looping" or image data overwriting would occur. If the bandwidth is inadequate, even for brief periods, the viewed image display will not be interrupted. Since the physiologic cycles are assumed to be periodic and very similar from cycle to cycle, the repetition of a cycle from time to time or periodically (e.g., two to four loops for each cycle on a consistent basis) still provides an accurate representation of the temporal nature of the image.
  • the sender's outgoing data buffer (first ping-pong memory 204) should be large enough to store up to two cycles of image data so that while one cycle is being transmitted, new image data may be written into the buffer simultaneously. However, if it takes longer than one cycle to empty the transmit buffer, the new data from the next cycle will be overwritten with the following cycle of data so that the data in the capture buffer does not continue to grow. This, in effect, adapts the amount of transmitted data to the actual bandwidth being provided by the transmission medium without having to know the actual bandwidth beforehand. Also, this ensures that the receiving station (remote location) is always updated with the most recent image data.
  • a cyclical memory allows image loops to be packaged by physiologic cycles, but not necessarily using image data from the same cycle. This idea is discussed above in the example of four physiologic cycles, A-D, wherein four image frames 1-4 are taken per cycle, and the image loops include portions of adjacent cycles (e.g., Image loop 1- Bl B2 A3 A4; Image loop 2 - Cl C2 B3 B4; Image loop 3 - DI D2 C3 C4). In this manner, when the data transmission is completed, the new transmit buffer can immediately send its image loop, even if the current physiologic cycle has not been completed.
  • Fig. 29 shows a schematic block diagram of an alternative system 184 for implementing a pseudo real-time transmission scheme.
  • the output of the acquisition device 204 is communicated to a first buffer or memory 186 at the patient location.
  • the first memory 186 has two buffers 188 and 190.
  • the buffers are referred to herein as loop A and loop B, respectively.
  • the loop A buffer receives the output of the acquisition device 204, and thus is connected to the input of the first memory 186.
  • the loop A buffer thus acts as the "capture buffer.”
  • the output of the first memory 186 is the output of the loop B buffer and is communicated via communication or transmission medium 192 to an input of a second buffer or memory 194.
  • the loop B buffer thus acts as the "transmit buffer.”
  • the second memory 194 also has two buffers 196 and 198.
  • the buffers are referred to herein as loop C and loop D, respectively.
  • the loop C buffer receives the output of the loop B buffer, and thus acts as the "receive buffer.”
  • the output of the second memory 194 is the output of the loop D buffer, and is communicated to a remote viewing station or remote image display 200 for providing the real-time or pseudo real-time image of subject's body.
  • the loop D buffer thus acts as the "display buffer.”
  • the scheme functions as follows:
  • the scheme functions as follows: (1) The loop C buffer receives the loop B image data and transfers it to the loop D buffer after a full image loop is received in the loop C buffer. (2) The image data in loop D is continuously played (i.e., looped) until the loop C buffer receives another full image loop.
  • Figs. 30A and 30B show sample contents of the loop A-D buffers in Figs. 28 and 29 for the first fourteen successive physiologic cycles obtained by the acquisition device 204 in a session.
  • the bandwidth of the transmission medium 192 is sufficient to support real-time image display. Therefore, the image data in the loop D buffer, which is output to the image display 200, is identical in sequence to the image data in the loop A buffer.
  • the loop B buffer to the loop C buffer in the time that it takes to collect exactly three physiologic cycles of image data.
  • the loop D buffer loops three times through each physiologic cycle of image data, skipping two cycles of image data between each displayed cycle of image data.
  • the loop B buffer transmits only the latest available image data. During the transmission time, any image data which cannot be transferred from the loop A buffer to the loop B buffer is discarded.
  • Means for performing the data transferring and transmitting functions of the memories 204 and 214 in the first embodiment, and the memories 186 and 194 in the second embodiment, should be well understood by an artisan familiar with such memories, in view of the description provided herein of its functions.
  • the preferred embodiment of the pseudo real-time transmission scheme uses image loops 100 having the structure shown in Figs. 3A-3C and Fig. 4.
  • the image data may be collected, organized or packaged in conventional ways, as long as the beginning and end points, or a triggering event, of the physiologic cycles are identifiable, and the scope of the pseudo real-time transmission scheme includes the transmission of image data collected, organized or packaged in conventional ways.
  • the preferred embodiment of the pseudo real-time transmission scheme uses two ping-pong memories.
  • other types of memory structures which provide the highlighted functions of the ping-pong memories may be used and are within the scope of the invention.
  • the pseudo real-time transmission scheme is equally effective in transmitting and remotely viewing other types of non-image-related periodic physiological data, such as EKG waves.
  • the inventions set forth above may also be applied to medical imaging systems other than ultrasound which acquire full motion image data of moving anatomic structures.
  • medical imaging systems other than ultrasound which acquire full motion image data of moving anatomic structures.
  • nuclear magnetic resonance (NMR) or magnetic resonance imaging (MRI), computerized axial tomography (CAT), positron emission tomography (PET), and the like acquire discrete scan lines of imaging data for reconstruction into a full-frame image of the body parts. These scan lines may be packaged into image loops.
  • the frame rates of the image frames may be adjusted using any of the techniques described above to reduce the amount of acquired data, without compromising image quality.
  • the image frames acquired by such systems may be filtered across physiologic cycles, instead of temporally, to reduce image artefacts.
  • a template image is acquired prior to contrast agent injection.
  • the template image is superimposed on the real-time image and displayed with a color map to distinguish it from the real-time image.
  • the operator aligns the real-time image with the template image by probe manipulation just prior to and during the contrast agent injection.
  • a contrast-only difference image is computed and captured.
  • the contrast-only difference image may be viewed in real time or immediately after acquisition, since no postprocessing or pre-contrast/post-contrast image alignment steps need to be performed. The operator can easily toggle between the contrast-only difference image and the superimposed template image and real-time image during the image acquisition session.
  • the image capture process may occur in a number of different ways.
  • the image acquisition device obtains a static template image at one point in the physiologic cycle.
  • the static template image is frozen on the image display during acquisition of the real-time image.
  • the real-time image may be taken during the entire physiologic cycle, or during a portion thereof.
  • the image display has a visually perceptible effect when the real-time, post-injection image frame display becomes closely aligned with the template image frame at the appropriate point in the physiologic cycle.
  • the image acquisition device obtains and displays a plurality of template image frames during a portion of, or during the entire physiologic cycle.
  • the multiple frame mode thus obtains full-motion images.
  • the sequence of template image frames are continuously replayed simultaneously with the real-time images in coordination with the physiologic cycle.
  • the image display has a visually perceptible effect when the real-time, post-injection image frame display becomes closely aligned with the template image frames at the same points in time in the physiologic cycle.
  • a portion of the heart cycle such as the systole portion, may be obtained for the template image, and the same portion of the heart cycle may be captured for the real-time image.
  • the multiple frame mode requires more image processing and memory resources than the trigger mode, since the template image is composed of a plurality of image frames, instead of a single image frame.
  • the multiple frame mode has an advantage over the trigger mode in that it is not necessary to wait a full physiologic cycle to see a comparison between the next real-time image and the template image. It is also easier to visually align the real-time image with the template image if the template image has a plurality of frames.
  • a single contrast-only difference image may be obtained at a predesignated point in the physiologic cycle, or a plurality of contrast-only difference images may be obtained at the same portion of the physiologic cycle as the template image.
  • the patient's heart rate changes after injection of the contrast agent. A heart rate change will not likely cause a problem when using the trigger mode, since the image comparison occurs at only one point in the physiologic cycle.
  • a heart rate change may cause a mismatch in the image frames, particularly if the multiple frame mode uses the entire physiologic cycle as opposed to only a portion thereof (e.g., only the systole portion). That is, there will be a different number of template image frames in a single physiologic cycle than the number of real-time image frames in a single physiologic cycle.
  • One solution is to realign respective images using conventional image processing techniques, such as by using schemes described in U.S. Patent No. 4,729,379 (Ohe) or U.S. Patent No. 4,878,115 (Elion), both of which are incorporated by reference in their entirety herein.
  • Figs. 34A and 34B show the steps of one prefened embodiment of the present invention wherein pre-injection and post-injection images are aligned. The description of these figures include both the trigger mode and the multiple frame mode.
  • the image(s) may be acquired at a known point in the heart cycle by using the QRS trigger as discussed above. See Fig. 34A. This image frame (or frames) becomes the "template image.”
  • the single image frame obtained in step (1) may be further improved via filtering during acquisition by obtaining a plurality of images at the exact same point in the cardiac cycle (using the QRS trigger as a reference point) and averaging the images, in the same manner as discussed above with respect to the prior art. Furthermore, when obtaining a plurality of template image frames and/or a plurality of real-time image frames, it is prefened to extend the prior art filtering concept to perform such filtering across the entire physiologic cycle (or the portion of the cycle being analyzed) by filtering successive image frames obtained at the same point in time within each physiologic cycle.
  • the template image or real-time image is displayed in a first color, and the other image is displayed in a second color. See Fig. 34B.
  • one "color” may be a grayscale (gray-scale) display wherein the strength of the ultrasound echo signal creates a visual display ranging from white for the strongest echoes to varying shades of gray.
  • one suitable display scheme is to display the template image in grayscale, and to display the real-time image using a standard color spectrum.
  • the scope of the invention includes any color display scheme which allows the user to visually distinguish between the template image and the real-time image.
  • a low level threshold may be applied to the template image and to the real-time image to remove all pixels below a chosen intensity level.
  • Template image pixels are displayed over real-time pixels when their intensity level is greater than the real-time image pixel intensity (i.e., when the template image shines through the real-time image). If perfect alignment is achieved, the template image would actually disappear from the screen, however, this may be difficult to achieve in practice.
  • the goal of the alignment step (3) above is to align tissue outlines and blood vessel outlines of the real-time image frames with the conesponding outlines in the template image frame(s).
  • the operator sees a visually perceptible effect when alignment has occuned.
  • Ultrasound images are characteristically made up of dark areas, typically representing blood pools or vessels, as well as bright areas representing highly reflective tissue.
  • the arrangement of bright and dark image details describes the underlying structure of the tissue under study. Such details are especially evident when imaging the heart, which is made up of large chambers and vessels that fill with blood periodically and are sunounded by highly reflective muscle tissue. When the blood chambers are filled, they show up as dark areas of the image and the sunounding muscle tissue or "structure" is very bright.
  • Tissue and blood vessel outlines are thus readily distinguishable as a result of the dark and bright image details.
  • the bright and dark areas of the template image overlap the bright and dark areas of the real-time image.
  • the bright areas of one image cross over or cover the dark areas of the other image, and vice-versa.
  • a visually perceptible effect occurs when the images are in alignment because the bright dark details of the template image are closely matched to those of the real-time image and bright areas of one image do not cross over or cover dark areas of the other image.
  • Fig. 35 is a schematic block diagram of a medical ultrasound imaging system 318 for contrast agent imaging of anatomic structures which have periodic physiological motion that define physiologic cycles.
  • the system 318 comprises an image frame acquisition device 320 including a transducer probe 322 and a processor
  • a template image memory 326 connected to a first output of the acquisition device 320, an image display processor 327 for receiving the output of the template image memory 326 and a second output of the processor 324, an image display 328 for receiving the output of the display processor 327, a difference image memory 330 connected at a first input to a second output of the processor 324 and connected at a second input to the output of the template image memory 326, a physiological monitor 332, and an image frame controller 334 for sending an acquisition trigger signal to the acquisition device 320 when the time of interest within the physiologic cycle is detected.
  • the two outputs of the processor 324 schematically represent outputs at two different points in time, not necessarily two output terminals.
  • the display processor 327 is part of the image display 328. However, these elements may be separate components of the system 318.
  • the physiological monitor 332 is an EKG monitor which detects the QRS trigger, and the image frame controller 334 produces the trigger signal at a predetermined time relative to the QRS trigger.
  • the processor 324 processes echoes produced from transmission of ultrasound energy into a subject's body.
  • the acquisition device 320 produces at least one template image frame prior to contrast agent injection.
  • Each template image frame comprises data representing an image of a portion of the subject's body acquired at a given time during the physiologic cycle.
  • the acquisition device 320 also produces real-time image frames during or after contrast agent injection. In another embodiment of the present invention, described below, the acquisition device 320 produces the real- time image frames before contrast agent injection.
  • Each real-time image frame comprises data representing a real-time image of a portion of the subject's body acquired at the same given time during the physiologic cycle as the template image frame or frames (hereafter, "template image").
  • the template image memory 326 stores the template image for a given physiologic cycle.
  • the image display 328 simultaneously displays the stored template image and the real-time image. To assist the operator in the alignment process, the image display 328 has a visually perceptible effect when the real-time image becomes closely aligned with the template image at the same points in time of the physiologic cycle.
  • the difference image memory 330 stores difference image data obtained by subtraction processing of the template image and the real-time image at the same times in the physiologic cycle. If the real-time image is obtained after the contrast agent has entered the imaged region, and alignment is proper, then the difference image is a contrast-only difference image. If the real-time image is obtained without the presence of a contrast agent, and alignment is proper, and no significant anatomic or blood flow change has occuned in the imaged region since the capture time of the template image, then the difference image would theoretically be blank (i.e., all pixel values should be zero). To obtain the contrast-only image, the operator must inject the contrast agent, wait for the contrast agent to reach the region of interest, capture the real-time image, and subtract the template image therefrom.
  • the display processor 327 causes the template image or the real-time image to be displayed in a first color, and the other image to be displayed in a second color.
  • the display processor 327 may apply a low level threshold to the template image, or to the template image and the real-time image, to remove all image pixels below a preset intensity level, thereby enhancing the visually perceptible effect when the real-time image becomes closely aligned with the template image at the same points in time of the physiologic cycle.
  • the difference image memory 330 receives the template image from the output of the display processor 327 as shown in Fig. 35, instead of directly from the template image memory 326.
  • a contrast-only difference image is calculated from two aligned images, pixel brightness differences which are less than zero are zeroed instead of being assigned an absolute value.
  • the real-time image is obtained without the presence of, or effect of, the contrast agent.
  • the real-time image is subtracted from the template image to obtain a "difference image."
  • the real-time image may be obtained prior to contrast agent injection, preferably immediately before the contrast agent enters the vessels of interest.
  • the real-time image may be obtained after contrast agent injection, but in a manner which minimizes or eliminates the effect of the contrast agent on the real-time image.
  • the microbubbles may be intentionally burst after they are detected as being in the imaged region. After being burst, the operator aligns the real-time image (now appearing without the effects of the contrast agent) with the template image. When the contrast agent returns, the contrast-only difference image is captured.
  • This embodiment of the present invention is generally similar to the four-step process described above, except that steps (2) and (3) are reversed.
  • the operator has two possible visually perceptible techniques for aligning the images.
  • the operator may try to overlay the two images as best as possible by using tissue and blood vessel outlines formed by bright and dark regions, as discussed above.
  • the operator may switch to a difference image mode which continuously performs the mean difference calculation above on the appropriately coordinated image frames and displays the difference image, and the operator manipulates the probe until as much of the displayed image as possible is eliminated from the image display.
  • an audio signal or an indicator lamp may be used in conjunction with, or as a substitute for, the image display in the alignment process.
  • an audio signal or lamp may be activated when alignment has occuned.
  • the audio signal may have different volumes or pitches which vary depending upon how close the operator gets to the proper alignment to assist the operator in "honing in” on the conect position.
  • the audio signal or other type of indicator may be controlled by circuitry which is responsive to the instantaneous mean difference value.
  • Fig. 36 is a schematic block diagram of the elements of the system 318 for implementing such schemes.
  • the elements include the template image memory 326 and image frame acquisition device 320 of Fig. 35.
  • the outputs of the template image memory 326 and the image frame acquisition device 320 may be processed by low level threshold processors 336 which process the image frame(s) in the same manner as the display processor 327 of Fig. 35.
  • the raw, or processed template image and the real-time image are connected to respective inputs of a conventional frame conelation processor 334, and alternatively, to a difference image processor 335 for display on the image display 328.
  • the display 328 may be toggled by the operator between a difference image display mode and a template image/realtime image display mode in assisting the operator during probe placement.
  • the frame conelation processor 334 outputs a frame conelation coefficient. In a simple scheme, the frame conelation coefficient is inversely proportional to the output of the mean difference calculation discussed above. As further background to the frame conelation process, see U.S. Patent No. 5,060,515 (Kanda et al.), incorporated in its entirety herein by reference.
  • the output of the frame conelation processor 334 is connected to the input of a coefficient processor 340 which controls an audio device 342 and/or indicator lamp 344.
  • the coefficient processor 340 compares the coefficients for the image frames from the same point or points in successive physiologic cycles to locate a maximum possible value (mean difference as close as possible to zero) as the operator moves the probe 322 around the area of interest. When a maximum achievable coefficient is detected, the operator holds the probe at that location, and captures the real-time image frames at that location after the contrast agent has entered the region of interest.
  • an operator input device (not shown) is provided to cause the systems of Figs. 35 and 36 to calculate, store and/or display the difference image when the operator achieves alignment.
  • All ultrasound imaging systems have memory for collecting and storing image data.
  • the image data may be stored as raw data representing the pixels obtained from data points along each scan line, or the image data may be stored as processed frame data, in the same manner as a frame of video.
  • Each ultrasound system has its own data storage structure and memory structure. Accordingly, it is sometimes difficult or impossible to view image data on one ultrasound system which is acquired on a different system. Likewise, stored image data often cannot be easily shared among systems, unless the data storage structure is well understood and any necessary data conversion steps are performed.
  • the present invention addresses these deficiencies in the prior art by providing a packaging/formatting structure for ultrasound imaging data obtained from an anatomic structure having periodic physiological motion, and which must be manipulated to perform difference image processing.
  • the packaging/formatting structure described herein is a prefened structure for manipulating the imaging data.
  • the scope of the invention also includes conventional techniques for storing ultrasound imaging data used for difference image processing.
  • the imaging data in accordance with the prefened embodiment of the present invention, is stored and packaged or formatted as image loops.
  • Each image loop includes frame data representing a plurality of image frames acquired at spaced time intervals within a single physiologic cycle.
  • the image frames refened to herein may be template image frames acquired in the multiple frame mode, or real-time image frames.
  • Fig. 37 A shows an image loop 400.
  • the image loop 400 includes a loop header 402 and the frame data 404 for one physiologic cycle.
  • the frame data comprises four image frames acquired during each cycle.
  • the image frame data 404 is packaged as a plurality of frame packets 406,-406 4 , described in more detail below with respect to Fig. 37B.
  • the loop header 402 may be used to identify the loop by number, for example, cycle number 482 in a stream of collected data.
  • the anatomic structure is the heart
  • the physiologic cycle is the heart cycle
  • the image loop 400 includes a plurality of image frames acquired for one physiologic cycle at a predefined time relative to a QRS trigger.
  • the image frames may be acquired from one QRS trigger to the next, or from a few milliseconds after the QRS trigger to a few milliseconds after the next QRS trigger. If the frame rate of the acquisition equipment is 30 FPS and the heart cycle is about one second, then there will be about 30 frame packets in each image loop 400 (as opposed to the four shown in Fig. 33A).
  • Fig. 37B shows a frame packet 406.
  • Each frame packet 406 includes a frame header 408 and the scan line data associated with the frame.
  • the scan line data 410 is packaged as a plurality of scan line packets 412 r 412 4 , described in more detail below with respect to Fig. 37C.
  • the frame header 408 identifies the frame by number, for example, frame number 3 in the sequence of four frame packets of Fig. 37A, or frame number 3 in a sequence of 30 frame packets in the heart example described above.
  • the frame numbers may begin with zero. In the heart cycle example, frame zero would be acquired at the QRS trigger or a predetermined time period displaced therefrom.
  • Fig. 37C shows a scan line packet 412.
  • Each scan line packet 412 includes a scan line header 428 and the individual line data 416 associated with the scan line.
  • the scan line header 428 comprising a line number representing a position within a scan sector, such as the last scan line shown in Fig. 31 (scan line 128).
  • the individual line data represents the data points along a scan line.
  • the loop header 402 the frame header 408 and the scan line header 428 thus function much like the identification portion or ID tag of a data packet in a digital communication scheme. These headers may also be used to store additional information to assist in processing the image data.
  • the loop header 402 may also include any of the following types of information:
  • Timing of the image loop with respect to a known point of the physiological cycle such as the timing with respect to the QRS trigger.
  • Time interval of the image loop This may be used for playback control.
  • Type of physiologic cycle e.g., heart cycle, respiratory cycle
  • Geometric parameters of the ultrasound scan e.g., apex distance, distance between the scan line data points.
  • Type of scan line data e.g., B-mode data, color flow, harmonic mode
  • the frame header 408 may also include any of the following types of information: (1) Time between adjacent, successively acquired or collected frames (i.e., cunent and previous frame). This information is particularly important when image frames are acquired or collected at different frame rates. During playback, this information is used to determine how long the frame should persist on the display screen before being replaced by the next frame.
  • Geometric parameters of the ultrasound scan especially if one or more parameters vary on a frame-to-frame basis. If the parameters are the same for all frames, then they should be placed in the loop header to avoid redundancy of information.
  • the image frame is a template image frame, a real-time image frame, or a difference image or contrast-only image frame.
  • the scan line header 414 may also include any of the following types of information:
  • One prefened embodiment of the invention for use in collecting image data of a heart cycle synchronizes frame acquisition using the QRS trigger point, wherein each image loop goes from one QRS trigger point (or a predeterrnined time with respect to the QRS trigger point) to the next QRS trigger point.
  • each image loop goes from one QRS trigger point (or a predeterrnined time with respect to the QRS trigger point) to the next QRS trigger point.
  • an image loop as defined herein starts at a first predefined time with respect to a predetermined event in a physiologic cycle and ends at a second predefined time with respect to the predetermined event in the physiologic cycle or in one or more successive physiologic cycles.
  • the image loop has a length of one physiologic cycle.
  • the image loop may have a length greater or less than one physiologic cycle, such as one-half cycle, 1 Vi cycles, or two cycles.
  • an image loop may consist of only the systole portion of a heart cycle, or may consist of two full heart cycles.
  • the image loop may start at the begirining of a first physiologic cycle and end at the end of the same physiologic cycle, right before the start of the next physiologic cycle.
  • the image loop is not necessarily acquired from the same physiologic cycle.
  • the image loop for the single physiologic cycle may, for example, be made up of image frames from the second half of a first physiologic cycle and the first half of the subsequent physiologic cycle. The frames in the image loop are still ordered from the start of the physiologic cycle to the end of the physiologic cycle, as determined by the collection time of the image frame with respect to the triggering event.
  • the first three image loops may be packaged by the single cycles, as follows:
  • Image loop 1 Al A2 A3 A4 Image loop 2 - Bl B2 B3 B4 Image loop 3 - Cl C2 C3 C4
  • the image loops may be packaged as follows:
  • Fig. 38 is an image frame memory 418 for storing ultrasound image frame data as a series of image loops 400.
  • Each image loop 400 includes an image loop header 402 and a plurality of frame packet 406.
  • the image data becomes very easy to manipulate by the memory and processing elements of Figs. 35 and 36.
  • Image loops simplify memory management of the image data because selected portions of data can be easily identified for storage, retrieval or display.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

L'invention concerne des données d'imagerie ultrasonique obtenues à une fréquence de trames d'image réglable (Fig. 6) ainsi qu'à une fréquence d'acquisition réglable (Fig. 15). Dans les données d'imagerie ultrasonique plein mouvement, on réduit l'effet de chatoiement en filtrant les trames d'image d'un cycle physiologique à travers les cycles, au lieu de les filtrer de trame à trame. L'invention concerne également des données d'imagerie ultrasonique qui sont groupées en boucles d'images (100). Ces données sont obtenues à partir de structures anatomiques présentant un mouvement physiologique périodique qui définit des cycles physiologiques successifs. L'invention concerne également un mécanisme permettant d'augmenter la fréquence d'acquisition de trames dans un système d'imagerie ultrasonique médical (168) qui acquiert des données d'imagerie d'une structure anatomique présentant un mouvement physiologique périodique qui définit des cycles physiologiques successifs (Fig. 26 et 27).
PCT/US1999/013912 1998-06-24 1999-06-21 Systemes d'imagerie ultrasoniques et procedes associes WO1999066841A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU46999/99A AU4699999A (en) 1998-06-24 1999-06-21 Ultrasonic imaging systems and methods

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US09/103,973 1998-06-24
US09/103,878 US6056691A (en) 1998-06-24 1998-06-24 System for collecting ultrasound imaging data at an adjustable collection image frame rate
US09/103,973 US6004270A (en) 1998-06-24 1998-06-24 Ultrasound system for contrast agent imaging and quantification in echocardiography using template image for image alignment
US09/103,878 1998-06-24

Publications (1)

Publication Number Publication Date
WO1999066841A1 true WO1999066841A1 (fr) 1999-12-29

Family

ID=26800957

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/013912 WO1999066841A1 (fr) 1998-06-24 1999-06-21 Systemes d'imagerie ultrasoniques et procedes associes

Country Status (2)

Country Link
AU (1) AU4699999A (fr)
WO (1) WO1999066841A1 (fr)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1367942A4 (fr) * 2000-07-21 2009-11-18 Verathon Inc Systeme de teleevaluation d'informations sous forme d'ultrasons obtenues par un dispositif programmable de recueil de donnees propres a des applications
CN106308848A (zh) * 2015-07-10 2017-01-11 通用电气公司 一种用于测量超声影像的方法及装置
CN111417347A (zh) * 2017-11-28 2020-07-14 北京深迈瑞医疗电子技术研究院有限公司 一种造影成像方法以及超声成像设备
CN113116388A (zh) * 2019-12-31 2021-07-16 深圳开立生物医疗科技股份有限公司 一种动脉腔内超声影像处理方法及相关装置
CN116019481A (zh) * 2021-10-26 2023-04-28 深圳迈瑞生物医疗电子股份有限公司 超声造影电影文件播放方法、超声电影文件播放方法和超声设备
CN116687441A (zh) * 2023-08-07 2023-09-05 深圳英美达医疗技术有限公司 超声成像方法、装置、超声成像设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5181513A (en) * 1990-05-29 1993-01-26 Pierre-Jean Touboul Method of acquiring ultrasound images
US5325859A (en) * 1992-01-14 1994-07-05 Ken Ishihara Ultrasonic diagnosis apparatus
US5467770A (en) * 1994-11-25 1995-11-21 General Electric Company Color adaptive frame averaging
US5797846A (en) * 1996-12-30 1998-08-25 General Electric Company Method to control frame rate in ultrasound imaging

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5181513A (en) * 1990-05-29 1993-01-26 Pierre-Jean Touboul Method of acquiring ultrasound images
US5325859A (en) * 1992-01-14 1994-07-05 Ken Ishihara Ultrasonic diagnosis apparatus
US5467770A (en) * 1994-11-25 1995-11-21 General Electric Company Color adaptive frame averaging
US5797846A (en) * 1996-12-30 1998-08-25 General Electric Company Method to control frame rate in ultrasound imaging

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1367942A4 (fr) * 2000-07-21 2009-11-18 Verathon Inc Systeme de teleevaluation d'informations sous forme d'ultrasons obtenues par un dispositif programmable de recueil de donnees propres a des applications
CN106308848A (zh) * 2015-07-10 2017-01-11 通用电气公司 一种用于测量超声影像的方法及装置
CN106308848B (zh) * 2015-07-10 2024-01-23 通用电气公司 一种用于测量超声影像的方法及装置
CN111417347A (zh) * 2017-11-28 2020-07-14 北京深迈瑞医疗电子技术研究院有限公司 一种造影成像方法以及超声成像设备
CN111417347B (zh) * 2017-11-28 2023-10-20 北京深迈瑞医疗电子技术研究院有限公司 一种造影成像方法以及超声成像设备
CN113116388A (zh) * 2019-12-31 2021-07-16 深圳开立生物医疗科技股份有限公司 一种动脉腔内超声影像处理方法及相关装置
CN116019481A (zh) * 2021-10-26 2023-04-28 深圳迈瑞生物医疗电子股份有限公司 超声造影电影文件播放方法、超声电影文件播放方法和超声设备
CN116687441A (zh) * 2023-08-07 2023-09-05 深圳英美达医疗技术有限公司 超声成像方法、装置、超声成像设备及存储介质
CN116687441B (zh) * 2023-08-07 2024-01-30 深圳英美达医疗技术有限公司 超声成像方法、装置、超声成像设备及存储介质

Also Published As

Publication number Publication date
AU4699999A (en) 2000-01-10

Similar Documents

Publication Publication Date Title
US6086537A (en) System for reducing speckle in full motion ultrasound image data by filtering across physiologic cycles
US6004270A (en) Ultrasound system for contrast agent imaging and quantification in echocardiography using template image for image alignment
US6350238B1 (en) Real-time display of ultrasound in slow motion
US5860927A (en) Diagnostic ultrasound Doppler imaging system and Doppler imaging method for ultrasonic diagnosis
US6488629B1 (en) Ultrasound image acquisition with synchronized reference image
CN100403989C (zh) 图像处理装置及超声波诊断装置
EP1637082B1 (fr) Dispositif de diagnostic par ultrasons
US6210333B1 (en) Medical diagnostic ultrasound system and method for automated triggered intervals
US6447450B1 (en) ECG gated ultrasonic image compounding
US5690111A (en) Ultrasound diagnostic apparatus
JP4676334B2 (ja) 生体信号モニタ装置
US5709210A (en) Ultrasound system for imaging
US20040225219A1 (en) Volumetric ultrasonic image segment acquisition with ECG display
WO1992014400A1 (fr) Procede pour le diagnostic des maladies cardiaques et coronariennes
JP2003111759A (ja) 超音波撮像の空間分解能及び時間分解能を改善する方法及び装置
JPH07178086A (ja) 超音波診断装置及び超音波診断方法
US20160015367A1 (en) Ultrasound imaging of fast-moving structures
US5845004A (en) Method and apparatus for performing frame interpolation in an ultrasound imaging system
KR20070074027A (ko) 대상체의 운동주기에 대응시켜 초음파 영상을디스플레이하는 초음파 시스템 및 방법
US5976086A (en) Ultrasonic diagnostic apparatus and filter to obtain sequential frames of image data
WO1999066841A1 (fr) Systemes d'imagerie ultrasoniques et procedes associes
EP0585070B1 (fr) Amélioration pour la discrimination de mouvement de la paroi d'organe
JP2002224114A (ja) 超音波診断装置及び超音波診断方法
EP1669031A1 (fr) Procédé de sélection d'une partie d'une série d'images d'échocardiographies
JPH09140711A (ja) 時相自動判定方法及び超音波診断装置

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AL AM AT AU AZ BA BB BG BR BY CA CH CN CU CZ DE DK EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT UA UG UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW SD SL SZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

122 Ep: pct application non-entry in european phase