Cardiovascular and cerebrovascular disease risk early warning method and system based on multi-mode ultrasonic features

Abstract

The invention discloses a cardiovascular and cerebrovascular disease risk early warning method and system based on multi-modal ultrasonic features. The method comprises the following steps: acquiring multi-modal ultrasonic feature data of a target object; based on a data analysis model, analyzing according to the ultrasonic characteristic data to obtain cardiovascular risk parameters; determining a risk level of the target object according to a preset multi-level risk threshold rule and the cardiovascular risk parameters; and determining alarm equipment and alarm data corresponding to the target object according to the risk level and a preset corresponding relationship between the risk level and an alarm strategy. Therefore, accurate cardiovascular risk grading early warning based on multi-mode ultrasound can be realized, the timeliness and pertinence of early intervention of cardiovascular diseases are improved, and the risk of cardiovascular events caused by risk misjudgment or alarm delay is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features. Background Technology

[0002] With the rapid growth in demand for early screening of cardiovascular diseases, medical institutions are increasingly emphasizing accurate early warning of cardiovascular risks. Current technologies typically collect single-modality ultrasound data, calculate cardiovascular risk parameters using fixed thresholds or simple models, and send warning information based on a unified alarm mechanism to support disease intervention. However, existing solutions lack comprehensive analysis of multi-modality ultrasound characteristic data, accurate calculation of risk parameters, and multi-level threshold matching and dynamic allocation of alarm devices for risk levels. This makes it difficult to achieve timely, tiered warnings. Commonly used, crude or delayed alarm strategies cannot adapt to individual differences and risk severity, resulting in insufficient targeted early intervention. This can easily lead to cardiovascular events due to misjudgment of risk or delayed alarms, limiting the preventative effect and clinical application value of ultrasound screening. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features, which can realize accurate cardiovascular risk classification and early warning based on multimodal ultrasound, improve the timeliness and pertinence of early intervention of cardiovascular diseases, and reduce the risk of cardiovascular events caused by risk misjudgment or alarm delay.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features, the method comprising: Acquire ultrasonic feature data of multiple modalities of the target object; Based on the data analysis model, cardiovascular risk parameters are obtained by analyzing the ultrasound feature data. The risk level of the target object is determined based on the preset multi-level risk threshold rules and the cardiovascular risk parameters. Based on the risk level and the correspondence between the preset risk level and alarm strategy, the alarm device and alarm data corresponding to the target object are determined.

[0005] As an optional implementation, in the first aspect of the invention, the ultrasound feature data is morphological feature data, dynamic feature data, or physiological feature data.

[0006] As an optional implementation, in the first aspect of the invention, the morphological feature data includes at least one of carotid intima-media thickness, vascular plaque area, and vessel dilatation; and / or, the physiological features include at least one of blood pressure variability and heart rate variability.

[0007] As an optional implementation, in a first aspect of the invention, the kinetic characteristic data includes at least one of peak systolic velocity, end-diastolic velocity, drag index, and pulsatility index.

[0008] As an optional implementation, in the first aspect of the invention, the step of analyzing the ultrasound feature data based on a data analysis model to obtain cardiovascular risk parameters includes: Data fusion is performed on all the ultrasound feature data to obtain multimodal fused data; The user parameters of the target object and the multimodal fusion data are input into the trained risk prediction model to obtain the output cardiovascular risk parameters; the risk prediction model is trained using a training dataset that includes multiple training cardiovascular ultrasound data and corresponding cardiovascular risk labels.

[0009] As an optional implementation, in the first aspect of the present invention, the risk prediction model includes a cross-prediction model and a time-series prediction model; the cross-prediction model is used to extract cross-correlation features from the ultrasound feature data of any two modalities and output a predicted first risk probability based on the extracted cross-correlation features; the time-series prediction model analyzes and predicts time-series data formed by the multimodal fusion data at multiple time points to output a predicted symptom time and a second risk probability; the cardiovascular risk parameters are specifically calculated through the following steps: The cross-risk parameter is obtained by weighted summation of all the first risk probabilities; wherein the calculation weight corresponding to each first risk probability is proportional to the historical association frequency corresponding to the two ultrasound feature data; the historical association frequency is the proportion of records in the historical diagnostic records that are diagnosed with cardiovascular disease and have both ultrasound feature data abnormalities out of the total number of records. Calculate the time difference between the predicted symptom time and the current time, and calculate the time weight that is proportional to the time difference; The time-series risk parameter is obtained by multiplying the time weight and the second risk probability. The cardiovascular risk parameter is obtained by multiplying the cross-risk parameter and the time-series risk parameter.

[0010] As an optional implementation, in the first aspect of the present invention, determining the alarm device and alarm data corresponding to the target object based on the correspondence between the risk level and the preset risk level and alarm strategy includes: For each candidate alarm device, calculate the matching degree between the candidate alarm device and the risk level; The candidate alarm device with the highest matching degree is determined as the alarm device corresponding to the target object; the alarm device is the target object's terminal device, family member's terminal device, or emergency center terminal device; An alarm data containing the cardiovascular risk parameters is generated, and the alarm data is sent to the alarm device to trigger an alarm.

[0011] As an optional implementation, in the first aspect of the invention, calculating the matching degree between the candidate alarm device and the risk level includes: Obtain the average historical risk level of each record in the historical alarm records of the candidate alarm device; Calculate the logarithmic value corresponding to the difference between the risk level and the average value of the risk level to obtain the matching degree between the candidate alarm device and the risk level.

[0012] A second aspect of this invention discloses a cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features, the system comprising: The acquisition module is used to acquire ultrasonic feature data of multiple modalities of the target object; The analysis module is used to analyze the ultrasound feature data based on the data analysis model to obtain cardiovascular risk parameters; The determination module is used to determine the risk level of the target object based on preset multi-level risk threshold rules and the cardiovascular risk parameters; The alarm module is used to determine the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk level and alarm strategy.

[0013] As an optional implementation, in a second aspect of the invention, the ultrasound feature data is morphological feature data, dynamic feature data, or physiological feature data.

[0014] As an optional implementation, in a second aspect of the invention, the morphological feature data includes at least one of carotid intima-media thickness, vascular plaque area, and vessel dilatation; and / or, the physiological features include at least one of blood pressure variability and heart rate variability.

[0015] As an optional implementation, in a second aspect of the invention, the kinetic characteristic data includes at least one of peak systolic velocity, end-diastolic velocity, drag index, and pulsatility index.

[0016] As an optional implementation, in a second aspect of the invention, the specific method by which the analysis module analyzes the ultrasound feature data based on a data analysis model to obtain cardiovascular risk parameters includes: Data fusion is performed on all the ultrasound feature data to obtain multimodal fused data; The user parameters of the target object and the multimodal fusion data are input into the trained risk prediction model to obtain the output cardiovascular risk parameters; the risk prediction model is trained using a training dataset that includes multiple training cardiovascular ultrasound data and corresponding cardiovascular risk labels.

[0017] As an optional implementation, in a second aspect of the invention, the risk prediction model includes a cross-prediction model and a time-series prediction model; the cross-prediction model is used to extract cross-correlation features from the ultrasound feature data of any two modalities and output a predicted first risk probability based on the extracted cross-correlation features; the time-series prediction model analyzes and predicts time-series data formed by the multimodal fusion data at multiple time points to output a predicted symptom time and a second risk probability; the cardiovascular risk parameters are specifically calculated through the following steps: The cross-risk parameter is obtained by weighted summation of all the first risk probabilities; wherein the calculation weight corresponding to each first risk probability is proportional to the historical association frequency corresponding to the two ultrasound feature data; the historical association frequency is the proportion of records in the historical diagnostic records that are diagnosed with cardiovascular disease and have both ultrasound feature data abnormalities out of the total number of records. Calculate the time difference between the predicted symptom time and the current time, and calculate the time weight that is proportional to the time difference; The time-series risk parameter is obtained by multiplying the time weight and the second risk probability. The cardiovascular risk parameter is obtained by multiplying the cross-risk parameter and the time-series risk parameter.

[0018] As an optional implementation, in a second aspect of the invention, the alarm module determines the specific method by which it determines the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk level and alarm strategy, including: For each candidate alarm device, calculate the matching degree between the candidate alarm device and the risk level; The candidate alarm device with the highest matching degree is determined as the alarm device corresponding to the target object; the alarm device is the target object's terminal device, family member's terminal device, or emergency center terminal device; An alarm data containing the cardiovascular risk parameters is generated, and the alarm data is sent to the alarm device to trigger an alarm.

[0019] As an optional implementation, in a second aspect of the invention, the alarm module calculates the matching degree between the candidate alarm device and the risk level in the following specific ways: Obtain the average historical risk level of each record in the historical alarm records of the candidate alarm device; Calculate the logarithmic value corresponding to the difference between the risk level and the average value of the risk level to obtain the matching degree between the candidate alarm device and the risk level.

[0020] A third aspect of this invention discloses another cardiovascular and cerebrovascular disease risk warning system based on multimodal ultrasound features, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention acquires multiple modal ultrasound feature data of the target object and calculates cardiovascular risk parameters based on a data analysis model. It determines the risk level according to multi-level risk threshold rules and matches the corresponding alarm device to send alarm data. This enables accurate cardiovascular risk classification and early warning based on multimodal ultrasound, improves the timeliness and pertinence of early intervention for cardiovascular diseases, and reduces the risk of cardiovascular events caused by risk misjudgment or alarm delay. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another cardiovascular and cerebrovascular disease risk warning system based on multimodal ultrasound features disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for early warning of cardiovascular and cerebrovascular diseases based on multimodal ultrasound features. By acquiring multimodal ultrasound feature data of the target object and calculating cardiovascular risk parameters based on a data analysis model, the risk level is determined according to multi-level risk threshold rules, and corresponding alarm devices are matched to send alarm data. This enables accurate cardiovascular risk classification and early warning based on multimodal ultrasound, improving the timeliness and targeting of early intervention for cardiovascular diseases, and reducing the risk of cardiovascular events caused by risk misjudgment or alarm delays. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features, as disclosed in an embodiment of the present invention. Figure 1 The described method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 1 As shown, the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features may include the following operations: 101. Obtain ultrasonic feature data of multiple modalities of the target object.

[0032] Optionally, the ultrasound feature data can be morphological feature data, dynamic feature data, or physiological feature data.

[0033] Optional, morphological data may include at least one of carotid intima-media thickness, plaque area, and vessel dilatation.

[0034] Optional, physiological characteristics include at least one of blood pressure variability and heart rate variability.

[0035] Optionally, the kinetic data may include at least one of peak systolic velocity, end-diastolic velocity, drag index, and pulsatility index.

[0036] 102. Based on the data analysis model, analyze the ultrasound feature data to obtain cardiovascular risk parameters.

[0037] Optionally, the cardiovascular risk parameter can be a continuous risk probability of 0-1 or a comprehensive risk score; this invention does not limit this.

[0038] 103. Determine the risk level of the target subject based on the preset multi-level risk threshold rules and cardiovascular risk parameters.

[0039] Optionally, the risk level can be one of four levels: low risk, medium risk, high risk, or extremely high risk. This invention does not limit the risk level.

[0040] 104. Based on the risk level and the pre-defined correspondence between the risk level and alarm strategy, determine the alarm devices and alarm data corresponding to the target object.

[0041] Optionally, the alarm device can be a smart bracelet worn by the patient, a family member's mobile app, or a terminal of a nearby emergency center; this invention does not limit the scope of the device.

[0042] As can be seen, the above-described embodiments of the invention acquire multiple modal ultrasound feature data of the target object and calculate cardiovascular risk parameters based on a data analysis model. They then determine the risk level according to multi-level risk threshold rules and match the corresponding alarm device to send alarm data. This enables accurate cardiovascular risk classification and early warning based on multimodal ultrasound, improves the timeliness and pertinence of early intervention for cardiovascular diseases, and reduces the risk of cardiovascular events caused by risk misjudgment or alarm delay.

[0043] As an optional embodiment, the above steps, including analyzing ultrasound feature data based on a data analysis model to obtain cardiovascular risk parameters, include: All ultrasound feature data are fused to obtain multimodal fused data; The user parameters and multimodal fusion data of the target object are input into the trained risk prediction model to obtain the output cardiovascular risk parameters.

[0044] Optionally, the risk prediction model is trained using a training dataset that includes multiple training cardiovascular ultrasound data and corresponding cardiovascular risk labels.

[0045] Optionally, the training dataset is derived from 50,000 clinical ultrasound examination reports and follow-up diagnostic results; this invention does not impose any limitations on this.

[0046] Optionally, the data fusion can be achieved through feature splicing, attention mechanism fusion, or a combination of early fusion and late fusion; this invention does not impose any limitations on this method.

[0047] Optionally, the user parameters may include age, gender, body mass index, blood pressure, smoking history or family medical history, which are not limited in this invention.

[0048] Optionally, the risk prediction model can employ deep neural networks (such as CNN-LSTM) or ensemble learning algorithms (such as XGBoost / RandomForest). Its input layer receives multidimensional feature vectors and basic patient information, the hidden layer performs nonlinear feature cross and temporal correlation analysis, and the output layer outputs a probability score of stroke or myocardial infarction occurring in the future. Specifically, the model supports online updates and automatically fine-tunes individual baseline parameters as user data accumulates, thereby improving the accuracy of personalized predictions.

[0049] As can be seen, through the above optional embodiments, multimodal fusion data is obtained by fusing all ultrasound feature data, and cardiovascular risk parameters are output by combining the user parameter input with the trained risk prediction model. This enables accurate risk quantification assessment based on fused data and personalized parameters, improves the accuracy and individual adaptability of risk parameter calculation, and reduces the risk of risk assessment bias caused by modal isolation or neglect of user parameters.

[0050] As an optional embodiment, the risk prediction model in the above steps includes a cross-prediction model and a time-series prediction model; the cross-prediction model is used to extract cross-correlation features from ultrasound feature data of any two modalities and output a predicted first risk probability based on the extracted cross-correlation features; the time-series prediction model analyzes and predicts time-series data formed by multimodal fusion data at multiple time points to output a predicted symptom time and a second risk probability; the cardiovascular risk parameters are specifically calculated through the following steps: The cross-risk parameter is obtained by weighted summation of all first risk probabilities; optionally, the calculation weight corresponding to each first risk probability is proportional to the historical association frequency corresponding to the two ultrasound feature data; the historical association frequency is the proportion of records in the historical diagnostic records that are diagnosed with cardiovascular disease and have both ultrasound feature data abnormalities out of the total number of records. Calculate the time difference between the predicted symptom time and the current time, and calculate the time weight that is proportional to the time difference; The time-series risk parameter is obtained by multiplying the time weight and the second risk probability. The cardiovascular risk parameters are obtained by multiplying the cross-risk parameter and the time-series risk parameter.

[0051] Optionally, the cross-prediction model can be a multi-head self-attention structure to capture the correlation features between different modalities, and the time-series prediction model can be a long short-term memory network to analyze dynamic changes within the cardiac cycle. This invention does not impose any limitations.

[0052] Optionally, the historical association frequency can be derived from 100,000 historical cases based on desensitization and statistical algorithms; this invention does not limit this.

[0053] As can be seen, through the above optional embodiments, the cross-prediction model extracts cross-correlation features between modalities to output the first risk probability, the time-series prediction model analyzes multi-time-point fusion data to output symptom time and the second risk probability, and the cardiovascular risk parameters are obtained based on the product of historical correlation frequency-weighted cross-risk and time-weighted time-series risk. This achieves accurate parameter calculation of cross-time-series dual risk fusion, improves the comprehensiveness and foresight of risk assessment, and reduces the risk of risk parameter distortion caused by a single prediction path.

[0054] As an optional embodiment, the step above, determining the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk level and alarm strategy, includes: For each candidate alarm device, calculate the matching degree between the candidate alarm device and the risk level; The candidate alarm device with the highest matching degree is determined as the alarm device corresponding to the target object; optionally, the alarm device is the target object's terminal device, family member terminal device, or emergency center terminal device; It generates alarm data including cardiovascular risk parameters and sends the alarm data to the alarm device to trigger an alarm.

[0055] Optionally, the alarm data may include risk level, key abnormal indicators, recommended measures, or location information; this invention does not impose any limitations.

[0056] As can be seen, through the above optional embodiments, by calculating the matching degree between candidate alarm devices and the current risk, the device with the highest matching degree is selected as the alarm device and alarm data containing risk parameters is sent, thereby realizing accurate alarm device selection and information push based on historical matching, improving the pertinence and response efficiency of alarms, and reducing the risk of early warning failure due to incompatible alarm devices.

[0057] As an optional embodiment, the step of calculating the matching degree between the candidate alarm device and the risk level in the above steps includes: Obtain the average historical risk level of each record in the historical alarm records of the candidate alarm device; Calculate the logarithmic value corresponding to the difference between the risk level and the average level to obtain the matching degree between the candidate alarm device and the risk level.

[0058] Optionally, the logarithmic value can be log(1 + |diff|), where diff is the difference between the risk level and the average level. This calculation operation can ensure that the greater the difference, the lower the matching degree. This invention does not limit this.

[0059] As can be seen, through the above optional embodiments, the logarithm of the difference between the average risk level of the historical alarm records of candidate alarm devices and the current risk level is calculated as the matching degree, so as to accurately select suitable alarm devices in the future, realize accurate alarm device selection and information push based on historical matching, and improve the targeting and response efficiency of alarms.

[0060] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features disclosed in an embodiment of the present invention. Figure 2The described multimodal ultrasound-based cardiovascular and cerebrovascular disease risk early warning system can be applied to data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). Figure 2 As shown, the cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features may include: The acquisition module 201 is used to acquire ultrasonic feature data of multiple modalities of the target object.

[0061] Analysis module 202 is used to analyze ultrasound feature data based on a data analysis model to obtain cardiovascular risk parameters.

[0062] The determination module 203 is used to determine the risk level of the target object based on preset multi-level risk threshold rules and cardiovascular risk parameters.

[0063] The alarm module 204 is used to determine the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between the risk level and the alarm strategy.

[0064] As can be seen, the above-described embodiments of the invention acquire multiple modal ultrasound feature data of the target object and calculate cardiovascular risk parameters based on a data analysis model. They then determine the risk level according to multi-level risk threshold rules and match the corresponding alarm device to send alarm data. This enables accurate cardiovascular risk classification and early warning based on multimodal ultrasound, improves the timeliness and pertinence of early intervention for cardiovascular diseases, and reduces the risk of cardiovascular events caused by risk misjudgment or alarm delay.

[0065] As an optional embodiment, the ultrasound feature data may be morphological feature data, dynamic feature data, or physiological feature data.

[0066] As can be seen, the above optional embodiments limit the types of ultrasound feature data to accurately characterize the cardiovascular condition of the subject, assist in achieving accurate cardiovascular risk classification and early warning based on multimodal ultrasound, and improve the timeliness and pertinence of early intervention for cardiovascular diseases.

[0067] As an optional embodiment, the morphological features include at least one of carotid intima-media thickness, plaque area, and vessel dilatation; and / or, the physiological features include at least one of blood pressure variability and heart rate variability.

[0068] As can be seen, the above optional embodiments define the content of morphological or physiological features to accurately characterize the cardiovascular morphology of the subject, assisting in the realization of accurate cardiovascular risk grading and early warning based on multimodal ultrasound, and improving the timeliness and pertinence of early intervention for cardiovascular diseases.

[0069] As an optional embodiment, the kinetic characteristic data includes at least one of peak systolic velocity, end-diastolic velocity, drag index, and pulsatility index.

[0070] As can be seen, the content of the dynamic characteristic data is defined through the above optional embodiments to accurately characterize the cardiovascular dynamics of the object, assist in realizing accurate cardiovascular risk classification and early warning based on multimodal ultrasound, and improve the timeliness and pertinence of early intervention for cardiovascular diseases.

[0071] As an optional embodiment, the analysis module, based on a data analysis model, analyzes ultrasound feature data to obtain cardiovascular risk parameters in the following specific ways: All ultrasound feature data are fused to obtain multimodal fused data; User parameters and multimodal fusion data of the target object are input into the trained risk prediction model to obtain the output cardiovascular risk parameters; the risk prediction model is trained using a training dataset that includes multiple training cardiovascular ultrasound data and corresponding cardiovascular risk labels.

[0072] As can be seen, through the above optional embodiments, multimodal fusion data is obtained by fusing all ultrasound feature data, and cardiovascular risk parameters are output by combining the user parameter input with the trained risk prediction model. This enables accurate risk quantification assessment based on fused data and personalized parameters, improves the accuracy and individual adaptability of risk parameter calculation, and reduces the risk of risk assessment bias caused by modal isolation or neglect of user parameters.

[0073] As an optional embodiment, the risk prediction model includes a cross-prediction model and a time-series prediction model. The cross-prediction model is used to extract cross-correlation features from ultrasound feature data of any two modalities and output a predicted first risk probability based on the extracted cross-correlation features. The time-series prediction model analyzes and predicts time-series data formed by multimodal fusion data at multiple time points to output a predicted symptom time and a second risk probability. The cardiovascular risk parameters are specifically calculated through the following steps: The cross-risk parameter is obtained by weighted summation of all first risk probabilities; optionally, the calculation weight corresponding to each first risk probability is proportional to the historical association frequency corresponding to the two ultrasound feature data; the historical association frequency is the proportion of records in the historical diagnostic records that are diagnosed with cardiovascular disease and have both ultrasound feature data abnormalities out of the total number of records. Calculate the time difference between the predicted symptom time and the current time, and calculate the time weight that is proportional to the time difference; The time-series risk parameter is obtained by multiplying the time weight and the second risk probability. The cardiovascular risk parameters are obtained by multiplying the cross-risk parameter and the time-series risk parameter.

[0074] As can be seen, through the above optional embodiments, the cross-prediction model extracts cross-correlation features between modalities to output the first risk probability, the time-series prediction model analyzes multi-time-point fusion data to output symptom time and the second risk probability, and the cardiovascular risk parameters are obtained based on the product of historical correlation frequency-weighted cross-risk and time-weighted time-series risk. This achieves accurate parameter calculation of cross-time-series dual risk fusion, improves the comprehensiveness and foresight of risk assessment, and reduces the risk of risk parameter distortion caused by a single prediction path.

[0075] As an optional embodiment, the alarm module determines the specific method of the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk levels and alarm strategies, including: For each candidate alarm device, calculate the matching degree between the candidate alarm device and the risk level; The candidate alarm device with the highest matching degree is determined as the alarm device corresponding to the target object; optionally, the alarm device is the target object's terminal device, family member terminal device, or emergency center terminal device; It generates alarm data including cardiovascular risk parameters and sends the alarm data to the alarm device to trigger an alarm.

[0076] As can be seen, through the above optional embodiments, by calculating the matching degree between candidate alarm devices and the current risk, the device with the highest matching degree is selected as the alarm device and alarm data containing risk parameters is sent, thereby realizing accurate alarm device selection and information push based on historical matching, improving the pertinence and response efficiency of alarms, and reducing the risk of early warning failure due to incompatible alarm devices.

[0077] As an optional embodiment, the alarm module calculates the matching degree between the candidate alarm device and the risk level in the following specific ways: Obtain the average historical risk level of each record in the historical alarm records of the candidate alarm device; Calculate the logarithmic value corresponding to the difference between the risk level and the average level to obtain the matching degree between the candidate alarm device and the risk level.

[0078] As can be seen, through the above optional embodiments, the logarithm of the difference between the average risk level of the historical alarm records of candidate alarm devices and the current risk level is calculated as the matching degree, so as to accurately select suitable alarm devices in the future, realize accurate alarm device selection and information push based on historical matching, and improve the targeting and response efficiency of alarms.

[0079] Example 3 Please see Figure 3 , Figure 3 This is another cardiovascular and cerebrovascular disease risk warning system based on multimodal ultrasound features disclosed in the embodiments of the present invention. Figure 3 The described cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features is applied in a data processing system / data processing equipment / data processing server (wherein, the server includes a local processing server or a cloud processing server). Figure 3 As shown, the cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features described in Embodiment 1.

[0080] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features described in Embodiment 1.

[0081] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features described in Embodiment 1.

[0082] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0083] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0084] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0085] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0089] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0090] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0091] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0092] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0093] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0094] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0095] Finally, it should be noted that the cardiovascular and cerebrovascular disease risk warning method and system based on multimodal ultrasound features disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features, characterized in that, The method includes: Acquire ultrasonic feature data of multiple modalities of the target object; Based on the data analysis model, cardiovascular risk parameters are obtained by analyzing the ultrasound feature data. The risk level of the target object is determined based on the preset multi-level risk threshold rules and the cardiovascular risk parameters. Based on the risk level and the correspondence between the preset risk level and alarm strategy, the alarm device and alarm data corresponding to the target object are determined.

2. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 1, characterized in that, The ultrasound feature data can be morphological feature data, dynamic feature data, or physiological feature data.

3. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 2, characterized in that, The morphological features include at least one of carotid intima-media thickness, plaque area, and vessel dilatation; and / or, the physiological features include at least one of blood pressure variability and heart rate variability.

4. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 2, characterized in that, The kinetic characteristic data include at least one of peak systolic velocity, end-diastolic velocity, drag index, and pulsatility index.

5. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 1, characterized in that, The method of obtaining cardiovascular risk parameters based on the data analysis model and the ultrasound feature data includes: Data fusion is performed on all the ultrasound feature data to obtain multimodal fused data; The user parameters of the target object and the multimodal fusion data are input into the trained risk prediction model to obtain the output cardiovascular risk parameters; the risk prediction model is trained using a training dataset that includes multiple training cardiovascular ultrasound data and corresponding cardiovascular risk labels.

6. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 5, characterized in that, The risk prediction model includes a cross-prediction model and a time-series prediction model. The cross-prediction model is used to extract cross-correlation features from the ultrasound feature data of any two modalities and output a predicted first risk probability based on the extracted cross-correlation features. The time-series prediction model analyzes and predicts time-series data formed by the fusion of multimodal data at multiple time points to output a predicted symptom time and a second risk probability. The cardiovascular risk parameters are specifically calculated through the following steps: The cross-risk parameter is obtained by weighted summation of all the first risk probabilities; wherein the calculation weight corresponding to each first risk probability is proportional to the historical association frequency corresponding to the two ultrasound feature data; the historical association frequency is the proportion of records in the historical diagnostic records that are diagnosed with cardiovascular disease and have both ultrasound feature data abnormalities out of the total number of records. Calculate the time difference between the predicted symptom time and the current time, and calculate the time weight that is proportional to the time difference; The time-series risk parameter is obtained by multiplying the time weight and the second risk probability. The cardiovascular risk parameter is obtained by multiplying the cross-risk parameter and the time-series risk parameter.

7. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 1, characterized in that, The step of determining the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk level and alarm strategy includes: For each candidate alarm device, calculate the matching degree between the candidate alarm device and the risk level; The candidate alarm device with the highest matching degree is determined as the alarm device corresponding to the target object; the alarm device is the target object's terminal device, family member's terminal device, or emergency center terminal device; An alarm data containing the cardiovascular risk parameters is generated, and the alarm data is sent to the alarm device to trigger an alarm.

8. The method for early warning of cardiovascular and cerebrovascular disease risks based on multimodal ultrasound features according to claim 7, characterized in that, The calculation of the matching degree between the candidate alarm device and the risk level includes: Obtain the average historical risk level of each record in the historical alarm records of the candidate alarm device; Calculate the logarithmic value corresponding to the difference between the risk level and the average value of the risk level to obtain the matching degree between the candidate alarm device and the risk level.

9. A cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features, characterized in that, The system includes: The acquisition module is used to acquire ultrasonic feature data of multiple modalities of the target object; The analysis module is used to analyze the ultrasound feature data based on the data analysis model to obtain cardiovascular risk parameters; The determination module is used to determine the risk level of the target object based on preset multi-level risk threshold rules and the cardiovascular risk parameters; The alarm module is used to determine the alarm device and alarm data corresponding to the target object based on the risk level and the preset correspondence between risk level and alarm strategy.

10. A cardiovascular and cerebrovascular disease risk early warning system based on multimodal ultrasound features, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the cardiovascular and cerebrovascular disease risk warning method based on multimodal ultrasound features as described in any one of claims 1-8.

Images