CN116230201A - Method and device for processing multisource physiological data - Google Patents
Method and device for processing multisource physiological data Download PDFInfo
- Publication number
- CN116230201A CN116230201A CN202211584820.0A CN202211584820A CN116230201A CN 116230201 A CN116230201 A CN 116230201A CN 202211584820 A CN202211584820 A CN 202211584820A CN 116230201 A CN116230201 A CN 116230201A
- Authority
- CN
- China
- Prior art keywords
- physiological data
- data
- physiological
- screening
- medical
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y20/00—Information sensed or collected by the things
- G16Y20/40—Information sensed or collected by the things relating to personal data, e.g. biometric data, records or preferences
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y40/00—IoT characterised by the purpose of the information processing
- G16Y40/20—Analytics; Diagnosis
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Primary Health Care (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to a method and a device for processing multisource physiological data. The invention can comprehensively analyze and process physiological data with different quality, frequency and other aspects from various monitoring equipment sources, and provides simple, clear and real information for assisting medical staff in decision diagnosis.
Description
Technical Field
The invention belongs to the field of physiological data processing, and particularly relates to a method and a device for processing multi-source physiological data.
Background
With the development of related scientific technologies, such as computer software and hardware technology, signal and image processing technology, display and recording technology, the technical ability of acquiring new monitored signals and parameters from various instruments is continuously improved. However, with the increasing number of monitoring devices for perioperative, ICU, intensive and emergency monitoring, the variety of physiological data acquired for patients is increasingly complex, thus causing the following problems:
first, medical staff typically observe mechanical ventilation waveforms and data of a patient through a ventilator, and observe monitoring data of electrocardio, blood pressure and the like of the patient through other devices (such as a monitor), so that a plurality of instruments are usually needed to display detected physiological parameters, and the scattered display of the different instruments clearly increases the observation burden for the medical staff.
Secondly, the same physiological data can be used for obtaining values with different precision and frequency by using different sensors, measuring methods and algorithms on a plurality of devices. For example, for Heart Rate (HR), there may be a source of SpO at the same time 2 The value of (2), the value derived from ECG, or the value derived from Pluse, it is difficult for medical staff to distinguish the value and availability of data only at the display end.
Thirdly, the monitored signal is often interfered by electromagnetic noise of the monitored environment, poor contact of the probe, movement of the patient and other factors, so that an analysis result is incorrect. Moreover, because different modes (different sensors, different algorithms and the like) are used for acquiring different monitored signals, the sensitivity degrees of different acquisition modes to various interferences are different, and the error degrees of various obtained physiological data are different. Therefore, medical staff is expected to be able to recognize physiological data with lower errors and higher reliability.
Fourth, the vast amount of physiological data streams recorded by various monitoring devices and the data mixed therein with large errors, which causes that comprehensive analysis and screening of the data streams takes a lot of effort and time, and it is difficult for medical staff to effectively or quickly utilize the information.
At present, although documents mention classifying physiological data to reduce the burden of medical staff in processing the data, these methods often do not involve quality analysis and evaluation of physiological data from multiple sources, resulting in physiological data of different quality, such as data with larger errors, and even erroneous data being collected and classified, which cannot effectively improve the efficiency of the medical staff in using the data, and may even lead to erroneous judgment.
Therefore, there is a need to improve the above-mentioned drawbacks in the prior art, so that when the physiological data from various monitoring devices, such as source, quality, frequency, etc. are faced, the obtained various vital sign signals can be comprehensively analyzed, so as to provide simpler, clear and real data information, truly exert the functions of multi-data source and multi-parameter monitoring, and improve the processing and analysis quality of the physiological data.
Disclosure of Invention
In order to improve the above problems, the present invention provides a method for processing multi-source physiological data, wherein the processing method comprises the following steps:
s1: acquiring physiological data of multiple sources;
s2: screening the multi-source physiological data obtained in the step S1 according to physiological data mapping rules;
s3: expanding the physiological data screened in the step S2;
wherein the physiological data mapping rule may be a physiological data processing rule formulated based on screening conditions including, but not limited to, one, two, three or more selected from the group consisting of medical scenario, source of physiological data, and/or quality of physiological data.
According to an embodiment of the invention, the physiological data is physiological data or physiological parameters of a living human or animal body.
According to an embodiment of the invention, the physiological data may be selected from physiological parameter data related to the health status of a living human or animal bodyOne, two, three or more, such as carbon dioxide removal data, lung protective ventilation data, oxygenation data, tissue oxygen and perfusion data, hemodynamic data, blood pressure data, intracranial pressure and brain relaxant data, brain oxygen and brain perfusion data, electroencephalogram data, and the like. As examples, the physiological data may be selected from one, two, three or more of the following including but not limited to: temperature (Temp), heart Rate (HR), systolic pressure (SBP), diastolic pressure (DBP), mean pressure (MBP), central Venous Pressure (CVP), positive end-expiratory pressure (PEEP), pulse oxygen saturation (SpO) 2 ) End-tidal carbon dioxide (EtCO) 2 ) Tissue oxygen saturation (StO) 2 ) Tidal Volume (TV), respiratory Rate (RR), ventilation per Minute (MV), concentration of inhaled oxygen (FiO) 2 ) Heart volume (CO), heart rate index (CI), systemic Vascular Resistance (SVR), electrocardiogram (ECG), pulse rate (Pluse), volume per Stroke (SV), variability per Stroke (SVV), etc.
The manner in which physiological data is obtained according to embodiments of the present invention may use a manner known to those skilled in the art, such as a detection instrument or a monitoring instrument connected to physiological data by a wired or wireless (e.g., cloud) manner.
According to an embodiment of the present invention, in the physiological data mapping rule, the screening based on the medical scene is preferably screening the category of the physiological data based on the medical scene.
According to an embodiment of the invention, in the physiological data mapping rules, the screening based on medical scenarios is performed before the screening based on physiological data sources.
According to an embodiment of the invention, the physiological data mapping rules, the screening based on physiological data sources is performed before the screening based on physiological data quality.
According to an embodiment of the invention, step S2 comprises the steps of:
s2.1: based on the medical scene, the category of the physiological data is screened for the first time, and the physiological data of the required category is obtained;
s2.2: based on the source of the physiological data, performing secondary screening on the physiological data of the required category obtained in the step S2.1 to obtain the physiological data of the required source;
s2.3: and (3) performing third screening on the physiological data of the required source obtained in the step S2.2 based on the quality of the physiological data to obtain the physiological data of the required quality.
According to an embodiment of the present invention, in step S2.1, physiological data of a desired category is screened out based on medical scenarios such as clinical scenarios, manually input or according to electronic medical records, clinical experience, departments, surgical "images". As an example, in a clinical scenario of lung function monitoring, the required physiological data categories may be selected from one, two, three or more of the following: basic vital sign monitoring parameters such as Heart Rate (HR), body temperature, systolic pressure (SBP), diastolic pressure (DBP), mean pressure (MBP), pulse oxygen saturation (SpO) 2 ) End-tidal carbon dioxide (EtCO) 2 ) Etc.; respiratory monitoring parameters, such as Positive End Expiratory Pressure (PEEP), tidal Volume (TV), respiratory Rate (RR), ventilation per Minute (MV), concentration of inhaled oxygen (FiO) 2 ) Etc.
According to embodiments of the present invention, multiple sources or multiple sources refer to sources from more than two different sources.
In step S2.2, certain physiological data screened based on medical scenarios, such as clinical scenarios, may be obtained from different sources, according to embodiments of the present invention.
According to embodiments of the present invention, the different sources of physiological data may be selected from different sources of physiological data due to factors such as sensing principles, measurement methods, measurement sites or algorithms of calculation.
Among the physiological data of different origin, those known to the person skilled in the art, which are closer to the true value of the physiological state of the patient, can be determined according to factors such as their measurement principle, calculation principle, measurement site, medical knowledge, monitoring device or monitoring module, according to an embodiment of the present invention.
According to embodiments of the present invention, some physiological data may exist in a "gold standard" acquisition format. For example, a "gold standard" for the way blood pressure values are measured may be the way blood pressure is obtained through a floating catheter. For physiological parameters for which the "gold standard" acquisition method exists, when data is acquired by the "gold standard" method, the data acquired by the "gold standard" method is preferable.
As an example, for physiological parameters RR of interest for a lung function monitoring clinical scenario, there may be three different categories in the data mapping rule selected from: RR (CO) 2 ) RR (total), RR (spont). RR (total) can be used as the preferred physiological data by those skilled in the art based on relevant medical knowledge.
According to an embodiment of the present invention, in step S2.2, after the second screening of the desired class of physiological data obtained in step S2.1, if the desired physiological data is a plurality of physiological data of different sources, these physiological data of different sources may be ranked.
According to embodiments of the present invention, physiological data may be ranked according to their importance or reliability.
Preferably, the ranking may be performed using a ranking model, resulting in an order of importance or reliability of the physiological parameters in the corresponding medical scenario. For example, the ranking model may be a ranking model based on importance or reliability.
According to embodiments of the invention, the ranking model may be modeled based on a learning ranking machine learning model such as PointWise, listWise, or based on a conventional model such as LR, xgBoost, and design feature engineering. For example, the ranking model may be built based on an initial training dataset built from existing medical knowledge.
According to an embodiment of the present invention, in step 2.3, the quality of the physiological data obtained in step S2.2 may be analyzed according to the signal quality of the physiological data.
According to an embodiment of the invention, in step 2.3, the signal quality of the physiological data represents the signal quality of the acquired physiological data.
It will be appreciated by those skilled in the art that the signal quality of the physiological data may be affected by factors such as the physical measurement method, calculation method, measurement environment, etc. of the physiological data. Preferably, the signals may differ in signal quality due to signal noise. For example, the signal quality may be analyzed by "quality index" or "noise index" to select a signal with less signal noise and better quality.
According to the embodiment of the invention, in step 2.3, the quality of the physiological data obtained in step 2.2 can be analyzed according to the signal noise of the physiological data and in combination with the data frequency or other factors, and the physiological data with the required quality can be obtained by screening according to the quality of the physiological data.
According to the embodiment of the invention, when the conversion amount is required to be observed and monitored, physiological data can be screened according to the conversion frequency in the step 2.3.
As an example, for physiological parameters RR of interest in a clinical scenario of lung function monitoring, among RR (total) data of different sources, an RR (total) derived from a ventilator is preferentially selected after physiological parameter mass analysis.
According to an embodiment of the invention, the quality index of the physiological data, preferably the signal quality index of the physiological data, may correspond one-to-one to physiological parameters available to the monitoring device as a source of said physiological data.
According to the embodiment of the invention, the signal quality indexes of the related physiological data acquired by the accessible monitoring device and the signal quality indexes of the physiological parameters acquired by other acquisition paths are uniformly measured according to the measurement or calculation principle of the physiological data acquired by the accessible monitoring device. For example, SQI, sqi_bis, sqi_scvo2, sqi_icg may be obtained by a michaeli T series monitor, sqi_bis may be obtained by a lidaco hemodynamic machine, eeg _bis_group_ SQI _val may be obtained by a GES5 monitor, etc., and the signal quality index of these physiological data may be uniformly measured with the signal quality index of the corresponding physiological data obtained by other acquisition means.
According to the embodiment of the invention, according to the measurement principle of the physiological data acquired by the accessible monitoring device, the correlation coefficient between the recorded event and the physiological data error can be calculated, and the correlation coefficient is applied to the analysis or calculation process of the physiological data signal quality index.
According to embodiments of the present invention, acquired physiological data may be grouped according to the channel of measurement or calculation based on the measurement or calculation principle of the physiological data acquired by an accessible monitoring device. Preferably, the signal quality indices of physiological data within the same group may be considered to have a correlation. For example, in the data on the hydrodynamic properties, the signal quality index of the physiological parameter data such as arterial pressure waveform and HR, SBP, DBP, MAP is considered to have correlation based on the physiological parameter such as HR, SBP, DBP, MAP calculated from arterial pressure waveform.
According to embodiments of the present invention, physiological parameter data quality may be ranked in combination with a calculated physiological data signal quality index according to a frequency of physiological parameter data required for a medical scenario. Preferably, higher ranked physiological data is preferentially output.
According to the embodiment of the invention, in the step S3, the expansion of the physiological data means that the physiological data screened in the step S2 is further processed by a further algorithm and/or a visualization means, so as to provide further information required by a medical scene or a medical staff.
According to embodiments of the present invention, the medical scenario or the further information required by the healthcare worker includes, but is not limited to, one, two, three or more of the following:
(1) Calculating and obtaining numerical data from waveform data through a correlation algorithm; CO, CI, HR, SBP, DBP and/or MAP data may be obtained, for example, by arterial pressure waveform calculations;
(2) Basic statistical analysis data of physiological data; for example, median, quartile, mean, variance, etc.;
(3) Accumulation effects of physiological data over time; for example, AUC value calculations, etc.;
(4) Correlation between physiological data;
(5) A procedure or result in a medical scenario.
The invention also provides a display method of the multi-source physiological data, wherein the display method comprises the processing method of the multi-source physiological data and the following steps:
s4: and (3) displaying the physiological data obtained in the step S3 and/or further information required by medical scenes or medical staff.
According to an embodiment of the present invention, in step S4, the physiological data and/or the medical scene obtained in step S3 or further information required by the medical staff is presented to the user via a display device.
The invention also provides a processing device of the physiological data, which can be used for the processing method of the physiological data or the display method of the physiological data.
According to an embodiment of the invention, the processing device of physiological data comprises the following modules:
a physiological data transmission module for acquiring physiological data of multiple sources, such as step S1 described above;
a physiological data screening module, configured to screen the multi-source physiological data acquired by the physiological data transmission module according to the physiological data mapping rule, for example, step S2 described above;
a physiological data expansion module for expanding the physiological data screened by the physiological data screening module, for example, step S3 described above; the method comprises the steps of,
an optional physiological data display module for displaying the physiological data obtained by the physiological data screening module or the physiological data expansion module and/or further information required by the medical scene or the medical staff, for example, step S4 described above.
According to an embodiment of the present invention, the physiological data transmission module may include an element connected to a detecting instrument or a monitoring instrument that provides physiological data, so as to transmit data collected or stored by the detecting instrument or the monitoring instrument to a processing device of the physiological data to acquire the physiological data. For example, the physiological data transmission module may be a wired or wireless (e.g., cloud) transmission module as known in the art.
According to an embodiment of the invention, the processing means of physiological data may further comprise an interaction module to allow a user to control the operation of at least one other module, such as the opening, closing, pausing or continuing of the operation of the other module; alternatively, the interaction module may allow the user to process physiological data, such as: the physiological data is manually entered, deleted, modified, classified or marked. The interaction module can input control instructions through a touch screen or mechanical keys.
It will be appreciated by those skilled in the art that the processing means of physiological data may include additional transmission elements to transmit the data or information generated in the modules described above to other devices, as desired.
The above modules in the physiological data processing apparatus according to the embodiment of the present invention may further include one or more data storage elements to store data or information according to the needs of each module.
The present invention also provides a method for preventing and/or treating a disease, comprising preventing and/or treating a disease using the physiological data processing method, the physiological data display method, or the physiological data processing apparatus.
The invention also provides a physiological data processing method, a physiological data display method or a physiological data processing device.
The invention also provides the use of the physiological data processing method, the physiological data display method or the physiological data processing device in preventing and/or treating diseases.
The invention also provides the use of the physiological data processing method, the physiological data display method or the physiological data processing device in patient management.
The invention also provides the use of a processing device of the above physiological data in the manufacture of a system for the prevention and/or treatment of a disease, or in the manufacture of a system for perioperative management of a patient.
Advantageous effects
The physiological data processing method, the physiological data display method or the physiological data processing device can comprehensively analyze and process physiological data with different quality, frequency and other aspects from various monitoring equipment sources, and provide simple, clear and real information for assisting medical staff in making decision diagnosis. The physiological data processing method, the physiological data display method or the physiological data processing device can automatically classify and comb the mass physiological parameter data of various monitoring devices into limited categories according to medical scenes, source classification and data quality filtration, so that the difficulty that medical staff need to pay attention to the mass physiological parameter data with different qualities in the prior art is greatly reduced, the burden of the medical staff is obviously reduced, the working efficiency of the medical staff is improved, and the management of the monitoring data, and the more effective and more reliable acquisition of the data and information are facilitated. Moreover, by adopting the physiological data processing method, the physiological data display method or the physiological data processing device, the physiological parameter source selection can be dynamically and automatically updated according to actual conditions (such as the change of medical scenes, the condition of real-time connection equipment, the condition of equipment sensors, the condition of medical instruments in use and the like), a physiological parameter list is maintained and displayed, and medical staff is helped to quickly acquire reliable data and information so as to make professional judgment as soon as possible.
Drawings
FIG. 1 is a schematic diagram of a method for processing multisource physiological data in an embodiment of the present invention.
Fig. 2 is a schematic diagram of a data mapping rule in an embodiment of the present invention.
FIG. 3 is a schematic diagram of a physiological data quality analysis of data mapping rules in an embodiment of the present invention.
Fig. 4 is a schematic diagram of the expanded acquisition CO, CI, HR, SBP, DBP, MAP according to the waveform calculation of the waveform arterial pressure in the embodiment of the present invention.
FIG. 5 is a schematic diagram of the extended acquisition of statistical analysis data based on physiological data calculations in an embodiment of the present invention.
FIG. 6 is a schematic diagram showing the effect of expanding the acquisition of physiological parameter accumulation over time according to the calculation of physiological data in the embodiment of the present invention.
FIG. 7 is a schematic illustration of the expansion of correlation between acquired physiological parameter data based on physiological data calculations in an embodiment of the present invention, wherein FIG. 7A is a physiological two-dimensional scatter plot and FIG. 7B is a physiological timing diagram.
Fig. 8 is a schematic diagram showing the process and results of the expanded acquisition of the vascular clamp experimental scenario according to the physiological data calculation in the embodiment of the present invention.
FIG. 9 is a schematic diagram of a physiological parameter reliability ordering model.
Fig. 10 is a schematic diagram of a physiological parameter data quality analysis.
Fig. 11 is a schematic diagram of physiological data processing in the vascular clamp experimental scenario in example 3.
Fig. 12 is a graph showing the results of expansion and display of physiological data in the vascular clamp experimental scenario in example 3.
Fig. 13 is a schematic diagram of physiological data processing in the liquid impact experimental scenario in example 4.
Fig. 14 is a schematic diagram showing the results of expansion and display of physiological data in the liquid impact experimental scenario in example 4.
Detailed Description
The technical scheme of the invention will be further described in detail below with reference to specific embodiments. It is to be understood that the following examples are illustrative only and are not to be construed as limiting the scope of the invention. All techniques implemented based on the above description of the invention are intended to be included within the scope of the invention.
Example 1
The embodiment provides a processing device of physiological data, which includes the following modules: a physiological data transmission module for acquiring physiological data of multiple sources, such as step S1 in example 2 below; a physiological data screening module for screening the multi-source physiological data acquired by the physiological data transmission module according to the physiological data mapping rule, for example, step S2 in embodiment 2 below; a physiological data expansion module for expanding the physiological data screened by the physiological data screening module, for example, step S3 in example 2 below; and a physiological data display module for displaying the physiological data obtained by the physiological data screening module or the physiological data expansion module and/or further information required by the medical scene or the medical staff, for example, step S4 in embodiment 2 below.
The physiological data transmission module comprises an element connected with a detecting instrument or a monitoring instrument for providing physiological data, so that the data collected or stored by the detecting instrument or the monitoring instrument is transmitted to a processing device of the physiological data to acquire the physiological data. The physiological data transmission module may be a wired or wireless (e.g., cloud) transmission module as known in the art. The physiological data processing device further comprises an interaction module to allow a user to control the operation of at least one other module, such as the opening, closing, pausing or continuing operation of the other module; alternatively, the interaction module may allow the user to process physiological data, such as: the physiological data is manually entered, deleted, modified, classified or marked. The interaction module can input control instructions through a touch screen or mechanical keys. The physiological data processing device also includes one or more data storage elements to store data or information as needed by each module. The physiological data transmission module comprises elements connected with a detection instrument and a monitoring instrument for providing physiological data, so that the data collected or stored by the detection instrument or the monitoring instrument are transmitted to a processing device of the physiological data to acquire the physiological data. The physiological data transmission module is a wired or wireless transmission module known in the art, for example, wireless transmission through the cloud.
Example 2
The present embodiment provides a method of physiological data processing using the processing apparatus of physiological data of embodiment 1, including the steps of:
(1) Acquisition of physiological data from multiple sources
The electronic medical system or a user manually inputs a medical scene, and according to the input medical scene and the supported physiological parameter types, the physiological parameter screening model is used for screening out the physiological parameters of interest in the medical scene;
and screening physiological parameter models according to the medical scenes, inputting the physiological parameter models into the medical scenes or medical scene keywords, and outputting the physiological parameter models into the physiological parameter sets of interest in the medical scenes. The model may use a ranking model to derive a ranking of the importance of each physiological parameter in the scene, followed by taking one or more physiological parameters of the ranking front. The sorting model can be modeled based on a learning sorting machine learning model such as PointWise, listWise or a traditional model such as LR, xgBoost and the like, and feature engineering is designed. The model is built based on an initial training dataset built from existing medical knowledge.
(2) Screening the multi-source physiological data obtained in the step (1) according to physiological data mapping rules
As illustrated in fig. 9, physiological parameters are screened according to source classification based on a physiological parameter reliability ranking model. The reliability ranking model is manually built according to the known medical common knowledge. The above-described ranking model structure is built based on basic medical common sense. For example, there may be a "gold standard" acquisition modality for some physiological parameters (e.g., the modality of acquiring a blood pressure waveform by a floating catheter may be referred to as a "gold standard" for blood pressure value measurement), for which the data acquired by the "gold standard" modality is considered the most reliable. For another example, for respiratory parameters, the ventilator-derived parameter data is considered more reliable than the monitor-derived parameter data.
As illustrated in fig. 10, a physiological parameter data quality analysis was performed. Wherein the time series abnormal data (abnormal) analysis model functions to analyze and identify non-physiological abnormal data in the time series data and to assign the non-physiological abnormal data to the lowest signal quality index. The non-physiological abnormal data refers to abnormal data caused by electromagnetic interference, interference data caused by physical movement of the device, noise data caused by skeletal muscle tremor, electrical interference and electrode movement occurring in ECG, and the like, for example. The time sequence analysis model is constructed based on a physiological parameter data and time sequence abnormal data (abnormal) analysis model with outlier data (outlier) removed, and aims to obtain the signal quality index of the physiological parameter data through analysis of a historical value and a real-time value of the physiological parameter data, and parameters output by the model are the physiological parameter signal quality index. The timing analysis model may be an autoregressive AR (p) model, a moving average MA (q) model, or the like. And integrating the physiological parameter data signal quality and the physiological parameter data frequency, scoring the physiological parameter data, and taking the data with the highest score.
Expanding and displaying the physiological data screened in the step S2: and (3) further processing the physiological data screened in the step (S2) through a further algorithm and/or a visual means, so as to provide further information required by medical scenes or medical staff, and presenting the physiological data and/or the information through a display module. Further information required by the medical scenario or healthcare worker includes: (1) Calculating and obtaining numerical data from waveform data through a correlation algorithm; CO, CI, HR, SBP, DBP and/or MAP data may be obtained, for example, by arterial pressure waveform calculations; (2) basic statistical analysis data of physiological data; for example, median, quartile, mean, variance, etc.; (3) the cumulative effect of physiological data over time; for example, AUC value calculations, etc.; (4) correlation between physiological data; (5) procedure or outcome in a medical scenario.
Example 3
The embodiment provides a physiological data processing method in a vascular clamp experimental scene. As shown in fig. 11, in a blood vesselIn the clamping experiment, physiological data StO is screened according to clinical scenes 2 And screening physiological data according to the source, carrying out quality analysis on the screened physiological data, and expanding and displaying the screened physiological data to obtain a process and a result display in the scene.
Example 4
The embodiment provides a physiological data processing method in a liquid impact experimental scene. As shown in fig. 12, in the liquid impact experiment, physiological data SVV, HR, CVP, MBP is screened according to clinical scenes, then physiological data HR (Pulse), CVP, MBP (index) are screened according to sources, and the physiological data obtained by the screening are subjected to quality analysis to obtain physiological data meeting quality requirements. And then expanding and displaying the screened physiological data to obtain the process and result display in the scene.
The embodiments of the present invention have been described above by way of example. However, the scope of protection of the present application is not limited to the above exemplary embodiments. Any modifications, equivalent substitutions, improvements, etc. made by those skilled in the art, which fall within the spirit and principles of the present invention, should be included in the scope of the present invention.
Claims (10)
1. A method of processing multisource physiological data, wherein the method of processing comprises the steps of:
s1: acquiring physiological data of multiple sources;
s2: screening the multi-source physiological data obtained in the step S1 according to physiological data mapping rules;
s3: expanding the physiological data screened in the step S2;
wherein the physiological data mapping rules may be physiological data processing rules formulated based on screening conditions including, but not limited to, one, two, three or more selected from the group consisting of medical scenarios, sources of physiological data, and/or quality of physiological data;
preferably, the physiological data may be selected from one, two, three or more of physiological parameter data related to the health status of a living human or animal body, such as one, two, three or more of carbon dioxide scavenging data, lung protective ventilation data, oxygenation data, tissue oxygen and perfusion data, hemodynamic data, blood pressure data, intracranial pressure and brain relaxation data, brain oxygen and brain perfusion data, brain electrical data, and the like.
2. The method of claim 1, wherein the physiological data is obtained by means known to those skilled in the art, such as by connecting a detection or monitoring instrument of the physiological data by wire or wireless means (e.g. cloud).
3. The processing method according to claim 1 or 2, wherein in the physiological data mapping rule, the screening based on the medical scene is preferably screening for categories of physiological data based on the medical scene;
preferably, the screening based on medical scenes is performed before the screening based on physiological data sources;
preferably, the screening based on physiological data sources is performed prior to the screening based on physiological data quality.
4. A process according to any one of claims 1 to 3, wherein step S2 comprises the steps of:
s2.1: based on the medical scene, the category of the physiological data is screened for the first time, and the physiological data of the required category is obtained;
s2.2: based on the source of the physiological data, performing secondary screening on the physiological data of the required category obtained in the step S2.1 to obtain the physiological data of the required source;
s2.3: and (3) performing third screening on the physiological data of the required source obtained in the step S2.2 based on the quality of the physiological data to obtain the physiological data of the required quality.
5. The method according to any one of claims 1-4, wherein in step S2.1, the physiological data of the desired category is selected based on medical scenes such as clinical scenes, manually input or according to electronic medical records, clinical experience, departments, surgical "images".
6. The method according to any one of claims 1-5, wherein in step S2.2, certain physiological data screened based on medical scenario, such as clinical scenario, can be obtained from different sources;
among physiological data of different sources, those known to those skilled in the art are preferred, which are closer to the true value of the physiological state of the patient, as determined by factors such as the measurement principle, calculation principle, measurement site, medical knowledge, monitoring device or monitoring module, etc.;
preferably, for physiological parameters for which a "gold standard" acquisition mode exists, in the case where data is acquired by the "gold standard" mode, data acquired by the "gold standard" mode is preferable.
7. The method of any one of claims 1 to 6, wherein in step S2.2, after the second screening of the desired class of physiological data obtained in step S2.1, if the desired physiological data is a plurality of physiological data of different sources, the physiological data of different sources may be ranked;
preferably, physiological data may be ranked according to their importance or reliability;
preferably, in step 2.3, the signal quality of the physiological data represents the signal quality of the acquired physiological data;
preferably, the quality of the physiological data obtained in step S2.2 may be analyzed according to the signal noise of the physiological data, in combination with the data frequency or other factors, and the physiological data of the required quality may be obtained by screening according to the quality of the physiological data.
8. The processing method according to any one of claims 1 to 7, wherein in step S3, the physiological data is further processed by a further algorithm and/or a visual means, so as to provide further information required by the medical scene or the medical staff.
9. A method of displaying multisource physiological data, wherein the method of displaying comprises a method of processing multisource physiological data according to any one of claims 1 to 8, and the steps of:
s4: and (3) displaying the physiological data obtained in the step S3 and/or further information required by medical scenes or medical staff.
10. A processing device of physiological data, which is usable for the processing method according to any one of claims 1 to 7 or the display method according to claim 9;
preferably, the processing device of physiological data comprises the following modules:
a physiological data transmission module for acquiring physiological data of multiple sources, such as step S1 described above;
a physiological data screening module, configured to screen the multi-source physiological data acquired by the physiological data transmission module according to the physiological data mapping rule, for example, step S2 described above;
a physiological data expansion module for expanding the physiological data screened by the physiological data screening module, for example, step S3 described above; the method comprises the steps of,
an optional physiological data display module for displaying the physiological data obtained by the physiological data screening module or the physiological data expansion module and/or further information required by the medical scene or the medical staff, for example, step S4 described above.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211584820.0A CN116230201A (en) | 2022-12-09 | 2022-12-09 | Method and device for processing multisource physiological data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211584820.0A CN116230201A (en) | 2022-12-09 | 2022-12-09 | Method and device for processing multisource physiological data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116230201A true CN116230201A (en) | 2023-06-06 |
Family
ID=86589958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211584820.0A Pending CN116230201A (en) | 2022-12-09 | 2022-12-09 | Method and device for processing multisource physiological data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116230201A (en) |
-
2022
- 2022-12-09 CN CN202211584820.0A patent/CN116230201A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8924235B2 (en) | Method and apparatus for monitoring physiological parameter variability over time for one or more organs | |
| JP5584413B2 (en) | Patient monitoring system and monitoring method | |
| CA2631870C (en) | Residual-based monitoring of human health | |
| US20070027368A1 (en) | 3D anatomical visualization of physiological signals for online monitoring | |
| JP5604451B2 (en) | System and method for automatically capturing and archiving clinically meaningful vital signs | |
| CN113974589B (en) | Multimodal behavioral paradigm evaluation optimization system and cognitive ability evaluation method | |
| US20150164428A1 (en) | Method for Multi-Scale Quality Assessment for Variability Analysis | |
| DE112011100282T5 (en) | Wellness assessment system | |
| CN107438399A (en) | Angiocarpy deteriorates early warning scoring | |
| CN106037720B (en) | A medical application system of mixed continuous information analysis technology | |
| CN114027842B (en) | Objective screening system, method and device for depression | |
| WO2014150684A1 (en) | Artifact as a feature in neuro diagnostics | |
| CN109715049A (en) | For the multi-modal physiological stimulation of traumatic brain injury and the agreement and signature of assessment | |
| Techentin et al. | 1D convolutional neural networks for estimation of compensatory reserve from blood pressure waveforms | |
| Hurtado et al. | Respiratory volume monitoring: A machine-learning approach to the non-invasive prediction of tidal volume and minute ventilation | |
| CN109065163A (en) | Tcm diagnosis service platform | |
| CN118975778A (en) | An anesthesia depth monitoring system based on EEG signals | |
| CN116936072A (en) | Physiological data classification processing device and method | |
| CN117547226A (en) | Comprehensive patient anesthesia operation pain nociception assessment method | |
| CN116230201A (en) | Method and device for processing multisource physiological data | |
| Sharma et al. | Characteristics of iot health data | |
| CN210494064U (en) | A Holter, Respiration and Exercise Monitoring Device | |
| KR20220117632A (en) | Method and system for providing remote counseling service | |
| TWI789040B (en) | Device, method, and system for collecting, managing and displaying physiological data | |
| CN114073500A (en) | Analysis method and analysis device for atrial fibrillation load |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Room 108 and 109, Building A, Lin988, Dazhong Fule Village, Huairou Town, Huairou District, Beijing, 101499 Applicant after: Beijing Lantian Tenglin Medical Equipment Co.,Ltd. Address before: Room 108 and 109, Building A, Lin988, Dazhong Fule Village, Huairou Town, Huairou District, Beijing, 101499 Applicant before: Beijing Lantian tenglin Medical Instrument Co.,Ltd. |
|
| CB02 | Change of applicant information |