CN107862699B - Bone edge extraction method, device, equipment and storage medium from bone CT images - Google Patents

Abstract

The present invention is applicable to the field of computer technology, and provides a method, device, equipment and storage medium for bone edge extraction from bone CT images. The method includes: receiving a bone CT image bone edge extraction request input by a user, and the bone edge extraction request includes: For the corresponding bone CT image, according to the bone edge extraction request, use the preset data fitting function to fit the grayscale data of the bone CT image to obtain the data fitting curve of the bone CT image, and fit the data according to the bone CT image. Curve, calculate the gray mean value of the bone edge area in the bone CT image, set the high threshold parameter and low threshold parameter of the preset bone edge detection algorithm according to the calculated gray mean value, and use the bone edge detection algorithm to detect the bone CT. The image is subjected to edge extraction, thereby improving the accuracy and speed of bone edge extraction in bone CT images, thereby improving the efficiency of bone edge extraction.

Description

Bone edge extraction method, device, equipment and storage medium from bone CT images

technical field

The invention belongs to the field of computer technology, and in particular relates to a bone edge extraction method, device, equipment and storage medium for bone CT images.

Background technique

The spine is an important support for the human body and consists of vertebrae, sacrum, coccyx and related connections. The key to minimally invasive treatment of intervertebral disc herniation is to safely, thoroughly and effectively eliminate the leakage of the nucleus pulposus of the spine. To ensure the accuracy and safety of surgery, it is necessary to take X-ray tomography (CT) images of patients with disc herniation before surgery. The edge extraction of 2D bone CT images is the first step in 3D reconstruction, and it is also the key to intraoperative 2D to 3D image registration. Therefore, accurate and rapid acquisition of edge information of CT images is of great significance in computer-aided spine diagnosis and minimally invasive surgical treatment.

The edge in the image is the local extreme point of gray level or a collection of points where the gray level changes sharply. The conventional idea of edge detection is to find the boundary according to the first and second derivatives of the image. The more classic edge detection operators are: Sobel, Prewitt, Roberts, Laplacian, Canny operators, etc. Among them, Canny operator is widely used because of its excellent performance such as high signal-to-noise ratio, high positioning accuracy and single edge response. The basic idea of the Canny operator is to use the first-order directional derivative of the two-dimensional Gaussian function in any direction as the noise filter, filter it by convolution with the image, and then find the local gradient maximum value for the filtered image to determine image edges. Subsequently, some new edge detection algorithms appeared, such as nonlinear wavelet transform method, neural network method, fuzzy technology, mathematical morphological transformation method and so on.

Bone CT images mainly include bone, bone marrow, muscle and body fluid noise. Due to the different components of each part, they appear as different grayscale values on the imaging, and the edges mainly exist at the connection of each component. If a traditional edge detection algorithm such as the Canny operator is used, the edges of various other tissues other than bone edges will be detected. However, the new edge detection operators often require a lot of computation, and the bone edge extraction of the human body with multiple complex components, such as spine bones, often has certain limitations.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a bone edge extraction method, device, equipment and storage medium for bone CT images, aiming to solve the problem of bone edge extraction due to the large amount of calculation and low accuracy of bone edge extraction in the existing bone edge extraction method. Extraction inefficiencies.

In one aspect, the present invention provides a method for extracting a bone edge from a bone CT image, the method comprising the following steps:

receiving a skeleton CT image skeleton edge extraction request input by a user, where the skeleton edge extraction request includes a corresponding skeleton CT image;

According to the bone edge extraction request, use a preset data fitting function to fit the grayscale data of the bone CT image to obtain a data fitting curve of the bone CT image;

According to the data fitting curve of the bone CT image, calculate the gray mean value of the bone edge region in the bone CT image;

According to the gray mean value obtained by the calculation, the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm are set, and the bone edge detection algorithm is used to perform edge extraction on the bone CT image.

In another aspect, the present invention provides a bone edge extraction device from a bone CT image, the device comprising:

a request receiving unit, configured to receive a skeleton CT image skeleton edge extraction request input by a user, where the skeleton edge extraction request includes a corresponding skeleton CT image;

a data fitting unit, configured to use a preset data fitting function to fit the grayscale data of the bone CT image according to the bone edge extraction request to obtain a data fitting curve of the bone CT image;

a mean value calculation unit, configured to calculate the gray mean value of the bone edge region in the bone CT image according to the data fitting curve of the bone CT image; and

The edge acquisition unit is configured to set the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm according to the gray mean value obtained by the calculation, and use the bone edge detection algorithm to perform edge detection on the bone CT image. extract.

In another aspect, the present invention also provides a medical device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor is implemented when the processor executes the computer program Such as the steps of the bone edge extraction method of the bone CT image.

In another aspect, the present invention also provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the bone edge extraction method as described in the bone CT image A step of.

The present invention receives a bone CT image bone edge extraction request input by a user, the bone edge extraction request includes a corresponding bone CT image, and uses a preset data fitting function to perform grayscale data on the bone CT image according to the bone edge extraction request. Fitting, obtains the data fitting curve of the bone CT image, calculates the average gray value of the bone edge area in the bone CT image according to the data fitting curve of the bone CT image, and calculates the preset bone edge according to the calculated gray average value. The high threshold parameter and low threshold parameter of the detection algorithm are set, and the bone edge detection algorithm is used to extract the edge of the bone CT image, thereby improving the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the efficiency of bone edge extraction.

Description of drawings

Fig. 1 is the realization flow chart of the bone edge extraction method of the bone CT image provided by Embodiment 1 of the present invention;

2 is a schematic diagram of a bone CT image edge detection process according to Embodiment 1 of the present invention;

3 is a comparison diagram of the edge detection result without a threshold and the edge detection result with a threshold provided by Embodiment 1 of the present invention;

4 is a schematic structural diagram of a device for extracting bone edges from a bone CT image according to Embodiment 2 of the present invention;

5 is a schematic structural diagram of a device for extracting bone edges from a bone CT image according to Embodiment 3 of the present invention; and

FIG. 6 is a schematic structural diagram of a medical device provided in Embodiment 4 of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

The specific implementation of the present invention is described in detail below in conjunction with specific embodiments:

Example 1:

FIG. 1 shows the implementation process of the bone edge extraction method from the bone CT image provided by the first embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown, and the details are as follows:

In step S101, a request for extracting a bone edge of a bone CT image input by a user is received, and the bone edge extraction request includes a corresponding bone CT image.

The embodiments of the present invention are suitable for medical equipment, especially for bone edge extraction of bone CT images in medical equipment. In this embodiment of the present invention, the received bone edge extraction request may include a corresponding bone CT image of the bone edge to be extracted, and the bone CT image may also be received separately.

In step S102, according to the bone edge extraction request, a preset data fitting function is used to fit the grayscale data of the bone CT image to obtain a data fitting curve of the bone CT image.

In the embodiment of the present invention, since the bone CT image mainly includes many different parts of the human body or animal body, such as bones, bone marrow, muscle, and body fluid noise, and these different parts have different components, these different parts are imaged as bones. CT images show different grayscale values, and the grayscale values of these different parts show different distribution states, so it is necessary to use different fitting functions to fit the grayscale data of these different parts in bone CT images. Therefore, preferably, the preset data fitting function is a mixed data fitting function composed of multiple Gaussian functions, so as to improve the fitting effect of the data fitting curve of the bone CT image.

对骨骼CT图像中的骨骼边缘区域灰度数据、骨髓区域灰度数据、肌肉区域灰度数据及背景噪声区域灰度数据进行拟合。其中，w_Gl为混合数据拟合函数中各个函数所占的比例，l＝1,2,3,4，且w_Gl满足

f_Gl(x)表示该混合数据拟合函数中的各个函数，优选地，各个函数可以为

这样，通过该混合数据拟合函数可有效提高骨骼CT图像数据拟合的精确性，拟合效果可参考图2。图2(a)示出了骨骼CT图像，图2(b)示出了使用本发明实施例的混合数据拟合函数拟合得到的效果图。Because the distribution states of the bone marrow region data, the muscle region data and the background noise data in the bones are different, preferably, when using the preset data fitting function to fit the grayscale data of the bone CT image, the mixed data is used. fit function

The grayscale data of bone edge region, bone marrow region grayscale data, muscle region grayscale data and background noise region grayscale data in bone CT images were fitted. Among them, w _Gl is the proportion of each function in the mixed data fitting function, l=1, 2, 3, 4, and w _Gl satisfies

f _G1 (x) represents each function in the mixed data fitting function, preferably, each function can be

In this way, the hybrid data fitting function can effectively improve the accuracy of bone CT image data fitting, and the fitting effect can be referred to FIG. 2 . Fig. 2(a) shows a CT image of the bone, and Fig. 2(b) shows an effect diagram obtained by fitting the mixed data fitting function according to the embodiment of the present invention.

Preferably, before using the preset data fitting function to fit the grayscale data of the bone CT image, the parameters in the data fitting function are initially estimated first, and then the parameters obtained by the initial estimation are optimized, so as to improve the The availability of the data fitting function is improved, thereby improving the accuracy of the data fitting. The parameters include parameters of each Gaussian function in the data fitting function and weights of each Gaussian function in the data fitting function.

(l＝1,2,3,4)，从而进一步提高了数据拟合函数的可用性。Further preferably, in the initial estimation of the parameters in the data fitting function, the k-means clustering algorithm is used for the initial estimation to obtain the initial values of these parameters in the data fitting function.

(l=1, 2, 3, 4), thereby further improving the usability of the data fitting function.

(l＝1,2,3,4)进行迭代优化操作，得到优化后的参数

从而进一步提高了数据拟合函数的可用性，进而提高数据拟合的精确性。Further preferably, when optimizing the parameters obtained by the initial estimation, the initial values of these parameters are first used as the initial values of the iteration of the Expectation Maximization Algorithm (EM algorithm), and then the initial values of these parameters are calculated by the EM algorithm.

(l=1, 2, 3, 4) perform iterative optimization operation to obtain the optimized parameters

Thus, the availability of the data fitting function is further improved, thereby improving the accuracy of the data fitting.

In step S103, according to the data fitting curve of the bone CT image, the average gray value of the bone edge region in the bone CT image is calculated.

In the embodiment of the present invention, after the data fitting curve of the bone CT image is obtained by fitting, first, according to the size of the gray value, the curve segment corresponding to the bone edge region is obtained from the data fitting curve of the bone CT image, and then according to the obtained The data corresponding to the curve segment of , calculate the average gray value of the bone edge area in the bone CT image.

In step S104, the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm are set according to the calculated gray mean value, and the bone edge detection algorithm is used to extract the edge of the bone CT image.

In the embodiment of the present invention, firstly, the high threshold parameter of the preset bone edge detection algorithm is set according to the calculated average gray value, and then the preset bone edge detection algorithm is set according to the relationship between the low threshold and the high threshold. The low threshold parameter of the algorithm is set, so that by limiting the high threshold parameter and low threshold parameter of the bone edge detection algorithm, the influence of other edges other than the bone edge is eliminated, thereby improving the effect of bone edge extraction. Please refer to FIG. 2 and FIG. 3 , wherein, FIG. 2( a ) and FIG. 3 ( a ) are CT images of bones, and FIG. 2 ( c ) and FIG. 3 ( c ) are schematic diagrams of edge detection results with a threshold provided by an embodiment of the present invention , Figure 3(b) is a schematic diagram of the edge detection result without threshold. From Figure 3(b) and Figure 3(c), it can be seen that by limiting the high threshold parameter and low threshold parameter of the bone edge detection algorithm, the edge of the bone is eliminated. The influence of other edges outside the skeleton improves the effect of bone edge extraction. After that, the bone edge detection algorithm is used to extract the edge of the bone CT image, so as to improve the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the efficiency of bone edge extraction.

Preferably, the bone edge detection algorithm is a Canny edge detection algorithm, so as to improve the efficiency of bone edge extraction.

Since the edges of different parts mainly exist at the connection of each part, preferably, when using the bone edge detection algorithm to extract the edge of the bone CT image, use a preset filtering algorithm to perform smoothing and denoising processing on the bone CT image, Calculate the magnitude and direction of the gray gradient of each pixel in the bone CT image after denoising, and perform a non-maximum suppression operation on the gray gradient amplitude. According to the gray gradient amplitude after the non-maximum suppression operation, The pixel points of the bone CT image are divided into strong edge points, weak edge points and non-edge points. After that, from the weak edge points, obtain the weak edge points connected with the strong edge points, set the acquired weak edge points and strong edge points connected with the strong edge points as the edge points of the bone CT image, and execute the operation on the edge points. The connection operation is used to obtain the edge of the bone CT image, thereby further improving the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the efficiency of bone edge extraction.

Specifically, when dividing the pixel points of the skeletal CT image into strong edge points, weak edge points and non-edge points according to the gray gradient amplitude after the non-maximum suppression operation, first obtain the preset high threshold and low Then compare the gray gradient amplitude after the non-maximum suppression operation with the high threshold and low threshold, and divide the pixels whose gray gradient amplitude is greater than the high threshold into strong edge points. The pixels whose gray gradient amplitude is less than the low threshold are divided into weak edge points, and the pixels whose gray gradient amplitude is between the high threshold and the low threshold are divided into non-edge points. edge point.

Embodiment 2:

FIG. 4 shows the structure of the device for extracting the bone edge of the bone CT image provided by the second embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown, including:

The request receiving unit 41 is configured to receive a bone CT image bone edge extraction request input by a user, where the bone edge extraction request includes a corresponding bone CT image.

The data fitting unit 42 is configured to use a preset data fitting function to fit the grayscale data of the bone CT image according to the bone edge extraction request to obtain a data fitting curve of the bone CT image.

The mean value calculation unit 43 is configured to calculate the gray mean value of the bone edge region in the bone CT image according to the data fitting curve of the bone CT image.

The edge acquisition unit 44 is configured to set the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm according to the calculated gray mean value, and use the bone edge detection algorithm to perform edge extraction on the bone CT image.

In this embodiment of the present invention, the request receiving unit 41 receives a bone CT image bone edge extraction request input by a user, the bone edge extraction request includes a corresponding bone CT image, and the data fitting unit 42 uses a preset according to the bone edge extraction request. The data fitting function fits the grayscale data of the bone CT image to obtain the data fitting curve of the bone CT image, and the mean value calculation unit 43 calculates the bone edge region in the bone CT image according to the data fitting curve of the bone CT image. Gray mean value, the edge acquisition unit 44 sets the high threshold parameter and low threshold parameter of the preset bone edge detection algorithm according to the calculated gray mean value, and uses the bone edge detection algorithm to perform edge extraction on the bone CT image, thereby The accuracy and speed of bone edge extraction in bone CT images are improved, thereby improving the efficiency of bone edge extraction.

In this embodiment of the present invention, each unit of the device for extracting bone edges from a bone CT image can be implemented by corresponding hardware or software units. Each unit can be an independent software or hardware unit, or can be integrated into a software and hardware unit. Here It is not intended to limit the present invention. For the specific implementation of each unit, reference may be made to the description of Embodiment 1, which will not be repeated here.

Embodiment three:

FIG. 5 shows the structure of the device for extracting the bone edge of the bone CT image provided by the third embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown, including:

The request receiving unit 51 is configured to receive a bone edge extraction request of a bone CT image input by a user, where the bone edge extraction request includes a corresponding bone CT image.

In this embodiment of the present invention, the bone edge extraction request received by the request receiving unit 51 may include the corresponding bone CT image of the bone edge to be extracted. Of course, the bone edge extraction request does not include the corresponding bone CT image of the bone edge to be extracted. When imaging, bone CT images can also be received separately.

The parameter optimization unit 52 is configured to initially estimate the parameters in the data fitting function, and optimize the parameters obtained by the initial estimation.

In the embodiment of the present invention, the parameter optimization unit 52 firstly estimates the parameters in the data fitting function, and then optimizes the parameters obtained by the initial estimation, thereby improving the usability of the data fitting function and further improving the data fitting accuracy. The parameters include parameters of each Gaussian function in the data fitting function and weights of each Gaussian function in the data fitting function.

(l＝1,2,3,4)，从而进一步提高了数据拟合函数的可用性。Preferably, in the initial estimation of the parameters in the data fitting function, the k-means clustering algorithm is used for the initial estimation to obtain the initial values of these parameters in the data fitting function

(l=1, 2, 3, 4), thereby further improving the usability of the data fitting function.

(l＝1,2,3,4)进行迭代优化操作，得到优化后的参数

从而进一步提高了数据拟合函数的可用性，进而提高数据拟合的精确性。Preferably, when optimizing the parameters obtained by the initial estimation, the initial values of these parameters are first used as the initial values of the iteration of the EM algorithm of the maximum expectation algorithm, and then the initial values of these parameters are calculated by the EM algorithm.

(l=1, 2, 3, 4) perform iterative optimization operation to obtain the optimized parameters

Thus, the availability of the data fitting function is further improved, thereby improving the accuracy of the data fitting.

The data fitting unit 53 is configured to use a preset data fitting function to fit the grayscale data of the bone CT image according to the bone edge extraction request to obtain a data fitting curve of the bone CT image.

In the embodiment of the present invention, since the bone CT image mainly includes many different parts of the human body or animal body, such as bones, bone marrow, muscle, and body fluid noise, and these different parts have different components, these different parts are imaged as bones. CT images show different grayscale values, and the grayscale values of these different parts show different distribution states, so that the data fitting unit 53 needs to use different fitting functions to match the grayscale values of these different parts in the bone CT image. Therefore, preferably, the preset data fitting function is a mixed data fitting function composed of multiple Gaussian functions, so as to improve the fitting effect of the data fitting curve of the bone CT image.

对骨骼CT图像中的骨骼边缘区域灰度数据、骨髓区域灰度数据、肌肉区域灰度数据及背景噪声区域灰度数据进行拟合。其中，w_Gl为混合数据拟合函数中各个函数所占的比例，l＝1,2,3,4，且w_Gl满足

f_Gl(x)表示该混合数据拟合函数中的各个函数，优选地，各个函数可以为

这样，通过该混合数据拟合函数可有效提高骨骼CT图像数据拟合的精确性。Because the distribution states of the bone marrow region data, the muscle region data and the background noise data in the bones are different, preferably, when using the preset data fitting function to fit the grayscale data of the bone CT image, the mixed data is used. fit function

The grayscale data of bone edge region, bone marrow region grayscale data, muscle region grayscale data and background noise region grayscale data in bone CT images were fitted. Among them, w _Gl is the proportion of each function in the mixed data fitting function, l=1, 2, 3, 4, and w _Gl satisfies

f _G1 (x) represents each function in the mixed data fitting function, preferably, each function can be

In this way, the accuracy of bone CT image data fitting can be effectively improved through the mixed data fitting function.

The mean value calculation unit 54 is configured to calculate the gray mean value of the bone edge region in the bone CT image according to the data fitting curve of the bone CT image.

In the embodiment of the present invention, after the data fitting curve of the bone CT image is obtained by fitting, the mean value calculation unit 54 first obtains the curve segment corresponding to the bone edge region from the data fitting curve of the bone CT image according to the size of the gray value , and then calculate the average gray value of the bone edge region in the bone CT image according to the data corresponding to the obtained curve segment.

The edge acquisition unit 55 is configured to set the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm according to the calculated gray mean value, and use the bone edge detection algorithm to perform edge extraction on the bone CT image.

In the embodiment of the present invention, the edge acquisition unit 55 firstly sets the preset high threshold parameter of the bone edge detection algorithm according to the calculated gray mean value, and then sets the preset high threshold parameter according to the relationship between the low threshold and the high threshold. The low threshold parameter of the bone edge detection algorithm is set, so that by limiting the high threshold parameter and low threshold parameter of the bone edge detection algorithm, the influence of other edges other than the bone edge is eliminated, thereby improving the effect of bone edge extraction. After that, the bone edge detection algorithm is used to extract the edge of the bone CT image, so as to improve the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the efficiency of bone edge extraction.

Preferably, the bone edge detection algorithm is a Canny edge detection algorithm, so as to improve the efficiency of bone edge extraction.

Since the edges of different parts mainly exist at the connection of each part, preferably, when using the bone edge detection algorithm to extract the edge of the bone CT image, use a preset filtering algorithm to perform smoothing and denoising processing on the bone CT image, Calculate the magnitude and direction of the gray gradient of each pixel in the bone CT image after denoising, and perform a non-maximum suppression operation on the gray gradient amplitude. According to the gray gradient amplitude after the non-maximum suppression operation, The pixel points of the bone CT image are divided into strong edge points, weak edge points and non-edge points. After that, from the weak edge points, obtain the weak edge points connected with the strong edge points, set the acquired weak edge points and strong edge points connected with the strong edge points as the edge points of the bone CT image, and execute the operation on the edge points. The connection operation is used to obtain the edge of the bone CT image, thereby further improving the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the efficiency of bone edge extraction.

Specifically, when dividing the pixel points of the skeletal CT image into strong edge points, weak edge points and non-edge points according to the gray gradient amplitude after the non-maximum suppression operation, first obtain the preset high threshold and low Then compare the gray gradient amplitude after the non-maximum suppression operation with the high threshold and low threshold, and divide the pixels whose gray gradient amplitude is greater than the high threshold into strong edge points. The pixels whose gray gradient amplitude is less than the low threshold are divided into weak edge points, and the pixels whose gray gradient amplitude is between the high threshold and the low threshold are divided into non-edge points. edge point.

Therefore, preferably, the data fitting unit 53 includes:

对骨骼CT图像中的骨骼边缘区域灰度数据、骨髓区域灰度数据、肌肉区域灰度数据及背景噪声区域灰度数据进行拟合；Data fitting subunit 531 for fitting functions using mixed data

Fit the grayscale data of bone edge region, bone marrow region grayscale data, muscle region grayscale data and background noise region grayscale data in bone CT images;

f_Gl(x)表示混合数据拟合函数中的各个函数；Among them, w _Gl is the proportion of each function in the mixed data fitting function, l=1, 2, 3, 4, and w _Gl satisfies

f _Gl (x) represents each function in the mixed data fitting function;

Preferably, the edge acquisition unit 55 includes:

The image denoising unit 551 is used for smoothing and denoising the bone CT image using a preset filtering algorithm;

The data processing unit 552 is used to calculate the amplitude and direction of the grayscale gradient of each pixel in the denoised skeleton CT image, and perform a non-maximum suppression operation on the grayscale gradient amplitude;

The pixel point classification unit 553 is used to divide the pixel points of the skeletal CT image into strong edge points, weak edge points and non-edge points according to the gray gradient amplitude after the non-maximum suppression operation;

The edge point extraction unit 554 is used to obtain the weak edge point connected with the strong edge point from the weak edge point, and set the acquired weak edge point and the strong edge point connected with the strong edge point as the edge point of the bone CT image ;as well as

The edge point connection unit 555 is configured to perform a connection operation on the edge points to obtain the edge of the bone CT image.

In this embodiment of the present invention, each unit of the device for extracting bone edges from a bone CT image can be implemented by corresponding hardware or software units. Each unit can be an independent software or hardware unit, or can be integrated into a software and hardware unit. Here It is not intended to limit the present invention.

Embodiment 4:

FIG. 6 shows the structure of the medical device provided by the fourth embodiment of the present invention. For the convenience of description, only the part related to the embodiment of the present invention is shown.

The medical device 6 of the embodiment of the present invention includes a processor 60 , a memory 61 , and a computer program 62 stored in the memory 61 and executable on the processor 60 . When the processor 60 executes the computer program 62, it implements the steps in the above embodiments of the bone edge extraction method for each bone CT image, for example, steps S101 to S104 shown in FIG. 1 . Alternatively, when the processor 60 executes the computer program 62, the functions of the units in the above-mentioned apparatus embodiments are realized, for example, the functions of the units 41 to 44 shown in FIG. 4 and the functions of the units 51 to 55 shown in FIG. 5 .

In this embodiment of the present invention, when the processor 60 executes the computer program 62 to implement the steps in the above-mentioned embodiments of the method for extracting bone edges from each bone CT image, it receives a request for bone edge extraction from a bone CT image input by a user, and the bone edge extraction The request includes the corresponding bone CT image. According to the bone edge extraction request, use the preset data fitting function to fit the grayscale data of the bone CT image to obtain the data fitting curve of the bone CT image. Data fitting curve, calculate the average gray value of the bone edge area in the bone CT image, set the high threshold parameter and low threshold parameter of the preset bone edge detection algorithm according to the calculated gray average value, and use the bone edge detection algorithm The edge extraction is performed on the bone CT image, thereby improving the accuracy and speed of bone edge extraction in the bone CT image, thereby improving the efficiency of bone edge extraction.

For specific steps implemented by the processor 60 in the medical device 6 when executing the computer program 62, reference may be made to the description of the method in the first embodiment, and details are not repeated here.

Embodiment 5:

In an embodiment of the present invention, a computer-readable storage medium is provided, and the computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the above-mentioned bone edge extraction method for each bone CT image is implemented in the embodiments steps, for example, steps S101 to S104 shown in FIG. 1 . Or, when the computer program is executed by the processor, the functions of the units in the above apparatus embodiments, for example, the functions of units 41 to 44 shown in FIG. 4 and the functions of units 51 to 55 shown in FIG. 5 are realized.

In this embodiment of the present invention, a bone edge extraction request from a bone CT image input by a user is received, and the bone edge extraction request includes a corresponding bone CT image. According to the bone edge extraction request, a preset data fitting function is used to fit the bone CT image. According to the data fitting curve of the bone CT image, the average gray value of the bone edge area in the bone CT image is calculated, and according to the calculated gray average value, the The preset high threshold parameters and low threshold parameters of the bone edge detection algorithm are set, and the bone edge detection algorithm is used to extract the edge of the bone CT image, thereby improving the accuracy and speed of bone edge extraction from the bone CT image, thereby improving the bone edge. Efficiency of edge extraction.

When the computer program is executed by the processor, the method for extracting the bone edge of the bone CT image can further refer to the description of the steps in the foregoing method embodiments, which will not be repeated here.

The computer-readable storage medium of the embodiments of the present invention may include any entity or device capable of carrying computer program codes, recording medium, for example, memory such as ROM/RAM, magnetic disk, optical disk, flash memory, and the like.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (10)

1. a bone edge extraction method of bone CT image, is characterized in that, described method comprises the steps: receiving a skeleton CT image skeleton edge extraction request input by a user, where the skeleton edge extraction request includes a corresponding skeleton CT image; According to the bone edge extraction request, use a preset data fitting function to fit the grayscale data of the bone CT image to obtain a data fitting curve of the bone CT image; According to the data fitting curve of the bone CT image, the average gray value of the bone edge region in the bone CT image is calculated; wherein, after the data fitting curve of the bone CT image is obtained by fitting, first The size of the gray value, the curve segment corresponding to the bone edge region is obtained from the data fitting curve of the bone CT image, and then according to the obtained data corresponding to the curve segment, the bone edge region in the bone CT image is calculated. grayscale mean; According to the gray mean value obtained by the calculation, the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm are set, and the bone edge detection algorithm is used to perform edge extraction on the bone CT image. 2. The method of claim 1, wherein the step of using a preset data fitting function to fit the grayscale data of the bone CT image comprises: Fit a function with mixed data

Fitting the grayscale data of the bone edge region, the grayscale data of the bone marrow region, the grayscale data of the muscle region and the grayscale data of the background noise region in the bone CT image; Wherein, the w _G1 is the proportion of each function in the mixed data fitting function, l=1, 2, 3, 4, and the w _G1 satisfies

The f _G1 (x) represents the respective functions in the mixed data fitting functions. 3. The method of claim 2, wherein before the step of fitting the grayscale data of the bone CT image using a preset data fitting function, the method further comprises: The parameters in the data fitting function are initially estimated, and the parameters obtained by the initial estimation are optimized, and the parameters include the parameters of each Gaussian function in the data fitting function and the parameters of each Gaussian function in the weights in the data fitting function described above. 4. The method of claim 1, wherein the step of performing edge extraction on the bone CT image using the bone edge detection algorithm, comprising: using a preset filtering algorithm to perform smoothing and denoising processing on the bone CT image; Calculate the amplitude and direction of the grayscale gradient of each pixel in the denoised bone CT image, and perform a non-maximum suppression operation on the grayscale gradient amplitude; dividing the pixel points of the skeletal CT image into strong edge points, weak edge points and non-edge points according to the grayscale gradient magnitude after the non-maximum value suppression operation; Obtain the weak edge point connected with the strong edge point from the weak edge point, and set the acquired weak edge point and the strong edge point connected with the strong edge point as the bone CT the edge points of the image; A connection operation is performed on the edge points to obtain the edge of the bone CT image. 5. A bone edge extraction device of a bone CT image, wherein the device comprises: a request receiving unit, configured to receive a skeleton CT image skeleton edge extraction request input by a user, where the skeleton edge extraction request includes a corresponding skeleton CT image; a data fitting unit, configured to use a preset data fitting function to fit the grayscale data of the bone CT image according to the bone edge extraction request to obtain a data fitting curve of the bone CT image; The mean value calculation unit is configured to calculate the average gray value of the bone edge region in the bone CT image according to the data fitting curve of the bone CT image; wherein, the data fitting of the bone CT image is obtained by fitting. After the curve is combined, first, according to the size of the gray value, the curve segment corresponding to the bone edge region is obtained from the data fitting curve of the bone CT image, and then the bone CT is calculated according to the obtained data corresponding to the curve segment. the grayscale mean of the bone edge regions in the image; and The edge acquisition unit is configured to set the high threshold parameter and the low threshold parameter of the preset bone edge detection algorithm according to the gray mean value obtained by the calculation, and use the bone edge detection algorithm to perform edge detection on the bone CT image. extract. 6. The apparatus of claim 5, wherein the data fitting unit comprises: data fitting subunit for fitting functions with mixed data

Fitting the grayscale data of the bone edge region, the grayscale data of the bone marrow region, the grayscale data of the muscle region and the grayscale data of the background noise region in the bone CT image; Wherein, the w _G1 is the proportion of each function in the mixed data fitting function, the l=1, 2, 3, 4, and the w _G1 satisfies

The f _G1 (x) represents the respective functions in the mixed data fitting functions. 7. The apparatus of claim 6, wherein the apparatus further comprises: A parameter optimization unit, configured to perform initial estimation on parameters in the data fitting function, and optimize the parameters obtained by the initial estimation, where the parameters include parameters of each Gaussian function in the data fitting function and all weight of each Gaussian function in the data fitting function. 8. The apparatus of claim 5, wherein the edge acquisition unit comprises: an image denoising unit, configured to perform smooth denoising processing on the bone CT image using a preset filtering algorithm; a data processing unit, configured to calculate the magnitude and direction of the grayscale gradient of each pixel in the denoised skeleton CT image, and perform a non-maximum suppression operation on the grayscale gradient magnitude; a pixel point classification unit, configured to divide the pixel points of the bone CT image into strong edge points, weak edge points and non-edge points according to the grayscale gradient amplitude after the non-maximum suppression operation; An edge point extraction unit, configured to obtain weak edge points connected with the strong edge points from the weak edge points, and extract the obtained weak edge points connected with the strong edge points and the strong edge point is set as the edge point of the bone CT image; and An edge point connection unit, configured to perform a connection operation on the edge points to obtain the edge of the bone CT image. 9. A medical device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the computer program as claimed in the claims Steps of any one of 1 to 4 of the method. 10. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented .

Images