Method, device, apparatus and readable medium for processing cardiac monitoring data

Abstract

Embodiments of the present disclosure provide methods, devices, equipment and readable media for processing cardiac monitoring data. The method includes: acquiring at least one set of sequences associated with the cardiac vital signs signal of the target user from the target sensor; determining whether valid vital signs data is included in at least one set of sequences based on a first machine learning model; in response to valid vital signs data being included in at least one set of sequences, determining at least one target feature associated with the valid vital signs data; and determining a judgment result for the cardiac vital signs signal of the target user by providing at least one target feature and at least one set of sequences to a second machine learning model. In this way, the difficulty of detection can be reduced, and the convenience of the detection process can be improved.

Description

Technical Field

[0001] The present disclosure relates to the fields of artificial intelligence and medical health technology, and more specifically, to a method, apparatus, device and computer-readable storage medium for processing cardiac monitoring data. Background Art

[0002] Currently, the application of artificial intelligence technology to the monitoring and processing of various physiological parameters of the human body (such as blood pressure, heart rate, pulse, etc.) has become a research hotspot. Heart failure (hereinafter referred to as heart failure) is a serious manifestation of various heart diseases. Early detection of heart failure and daily monitoring are particularly important for users at risk of heart failure. Summary of the invention

[0003] In a first aspect of the present disclosure, a method for processing cardiac monitoring data is provided. The method comprises: acquiring at least one set of sequences associated with a cardiac vital sign signal of a target user from a target sensor; determining whether valid vital sign data is included in at least one set of sequences based on a first machine learning model; in response to valid vital sign data being included in at least one set of sequences, determining at least one target feature associated with the valid vital sign data; and determining a judgment result for the cardiac vital sign signal of the target user by providing at least one target feature and at least one set of sequences to a second machine learning model.

[0004] In a second aspect of the present disclosure, a device for processing cardiac monitoring data is provided. The device includes: an acquisition module configured to acquire at least one set of sequences associated with a cardiac vital sign signal of a target user from a target sensor; a data determination module configured to determine whether at least one set of sequences includes valid vital sign data based on a first machine learning model; a feature determination module configured to determine at least one target feature associated with the valid vital sign data in response to at least one set of sequences including valid vital sign data; and a providing module configured to determine a judgment result for the cardiac vital sign signal of the target user by providing at least one target feature and at least one set of sequences to a second machine learning model.

[0005] In a third aspect of the present disclosure, an electronic device is provided. The electronic device includes at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, and when the instructions are executed by the at least one processing unit, the electronic device executes the method of the first aspect of the present disclosure.

[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein a computer program is stored on the computer-readable storage medium, and the computer program can be executed by a processor to perform the method according to the first aspect of the present disclosure.

[0007] In a fifth aspect of the present disclosure, a computer program product is provided, comprising computer executable instructions, wherein the computer executable instructions can be executed by a processor to perform the method according to the first aspect of the present disclosure.

[0008] It should be understood that the contents described in the summary of the invention are not intended to limit the key features or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In the following, in conjunction with the accompanying drawings and with reference to the following detailed description, the above and other features, advantages and aspects of various implementations of the present disclosure will become more apparent. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1A A schematic diagram showing an example environment in which embodiments of the present disclosure can be implemented;

[0011] Figure 1B A schematic diagram showing an example environment in which a terminal device detects a cardiac vital sign signal according to some embodiments of the present disclosure;

[0012] Figure 2 A schematic diagram showing the structure of a target sensor according to some embodiments of the present disclosure;

[0013] Figure 3 A schematic diagram showing at least one sequence according to some embodiments of the present disclosure;

[0014] Figure 4 A schematic diagram showing a process of heart failure risk identification according to some embodiments of the present disclosure;

[0015] Figure 5A-5C A schematic diagram showing an example interface for unstarted heart failure detection according to some embodiments of the present disclosure;

[0016] Figure 6A-6D A schematic diagram showing an example interface in a heart failure detection process according to some embodiments of the present disclosure;

[0017] Figure 7A-7C A schematic diagram showing an example interface for displaying a judgment result according to some embodiments of the present disclosure is shown;

[0018] Figure 8 A flowchart showing a process of cardiac monitoring data processing according to some embodiments of the present disclosure;

[0019] Fig. 9 A schematic structural block diagram showing an apparatus for processing cardiac monitoring data according to certain embodiments of the present disclosure; and

[0020] Fig.10 A block diagram of an electronic device is shown in which one or more embodiments of the present disclosure may be implemented. DETAILED DESCRIPTION

[0021] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0022] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below.

[0023] It should be noted that the acquisition, storage and application of user personal information involved in the technical solution of the present disclosure are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0024] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0025] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information, so that the user can independently choose whether to provide personal information to software or hardware such as electronic devices, applications, servers or storage media that execute operations of the technical solution of the present disclosure based on the prompt message.

[0026] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information is sent to the user in a manner such as a pop-up window, in which the prompt information can be presented in text form. In addition, the pop-up window can also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0027] As used herein, the term "model" can learn the association between the corresponding input and output from the training data, so that after the training is completed, the corresponding output can be generated for a given input. The generation of the model can be based on machine learning technology. Deep learning is a machine learning algorithm that processes inputs and provides corresponding outputs by using multi-layer processing units. A neural network model is an example of a model based on deep learning. In this article, "model" may also be referred to as "machine learning model", "learning model", "machine learning network" or "learning network", and these terms are used interchangeably in this article.

[0028] A "neural network" is a machine learning network based on deep learning. A neural network is capable of processing inputs and providing corresponding outputs, and typically includes an input layer and an output layer and one or more hidden layers between the input layer and the output layer. Neural networks used in deep learning applications typically include many hidden layers, thereby increasing the depth of the network. The layers of a neural network are connected in sequence so that the output of the previous layer is provided as input to the next layer, where the input layer receives the input of the neural network and the output of the output layer serves as the final output of the neural network. Each layer of a neural network includes one or more nodes (also called processing nodes or neurons), each of which processes input from the previous layer.

[0029] Generally, machine learning can be roughly divided into three stages, namely the training stage, the testing stage, and the application stage (also called the inference stage). In the training stage, a given model can be trained using a large amount of training data, and the parameter values ​​are continuously updated iteratively until the model can obtain consistent inferences that meet the expected goals from the training data. Through training, the model can be considered to be able to learn the association between input and output (also called input-to-output mapping) from the training data. The parameter values ​​of the trained model are determined. In the testing stage, the test input is applied to the trained model to test whether the model can provide the correct output, thereby determining the performance of the model. The testing stage can sometimes be integrated into the training stage. In the application or inference stage, the trained model can be used to process the actual model input based on the parameter values ​​obtained through training to determine the corresponding model output.

[0030] Heart failure is a serious manifestation or late stage of various heart diseases. Currently, there are about 330 million people at risk of heart failure, of which 13.7 million are confirmed to have heart failure. In addition, relevant industry data show that if heart failure is diagnosed, the risk of death within 5 years is as high as 51.8%. At the same time, the shortage of cardiologists also poses a challenge to the examination of heart failure. However, the number of cardiologists is obviously insufficient, with an average of only 4.8 cardiologists per 100,000 people, which is a serious test for medical resources. If there is a simplified home detection plan for heart failure, it can alleviate the pressure on cardiology medical resources to a certain extent, and early detection and daily monitoring of heart failure are also important for users at risk of heart failure.

[0031] Traditionally, heart failure disease detection solutions are generally conducted through medical examinations in hospitals. Therefore, the diagnosis of heart failure relies on a combination of blood tests, cardiac ultrasound and other examinations and the doctor's professional knowledge, and cannot be detected at any time. In addition, this detection solution not only relies on professionals and equipment, but also fails to detect potential heart failure risks in a timely manner. However, for heart disease, in many cases, the body has no obvious symptoms before the disease occurs. This does not mean that the heart is not at risk of heart failure. Therefore, many times, users may not feel any discomfort, so they do not detect it in time, resulting in the failure to detect the risk of heart failure in time.

[0032] In view of this, an embodiment of the present disclosure provides a solution for processing cardiac monitoring data. The solution includes: obtaining at least one set of sequences associated with the cardiac vital sign signal of the target user from the target sensor; determining whether valid vital sign data is included in at least one set of sequences based on a first machine learning model; if valid vital sign data is included in at least one set of sequences, determining at least one target feature associated with the valid vital sign data; and then, determining the judgment result for the cardiac vital sign signal of the target user by providing at least one target feature and at least one set of sequences to a second machine learning model. In this way, the difficulty of detection can be reduced, and the convenience of the detection process can be improved.

[0033] Figure 1A A schematic diagram of an example environment 100A in which embodiments of the present disclosure can be implemented is shown. In the example environment 100A, a target sensor 120 is installed in a terminal device 110. A user 140 can interact with the target sensor 120 via the terminal device 110 and / or an attached device of the terminal device 110.

[0034] In some embodiments, the target sensor 120 may be used to acquire at least one set of sequences associated with the cardiac vital signs of the target user. Figure 1AIn the environment 100A, if the target sensor 120 is in an active state, the terminal device 110 may present a page 150 for interacting with the user 140. The page 150 may include various types of pages, such as an information interaction page, a detection page, a detection result presentation page, and the like.

[0035] In some embodiments, the terminal device 110 communicates with the server 130 to provide services to the target sensor 120. The terminal device 110 can be any type of mobile terminal, fixed terminal or portable terminal, including mobile phones, wearable devices, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication systems (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio broadcast receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. In some embodiments, the terminal device 110 can also support any type of interface for the user (such as "wearable" circuits, etc.). The server 130 can be various types of computing systems / servers that can provide computing capabilities, including but not limited to mainframes, edge computing nodes, computing devices in cloud environments, and the like.

[0036] Figure 1B A schematic diagram of an example environment 100B for detecting a cardiac vital sign signal by a terminal device in some embodiments of the present disclosure is shown. Figure 1A Example environment 100B is described with reference to environment 100A.

[0037] In some embodiments, the terminal device 110 is configured with a target sensor 120. In some examples, the terminal device 110 may indicate a mobile phone. Figure 1B As shown, the user 140 can lie flat on the plane 160 according to the example posture in the figure, and place the terminal device 110 at a predetermined position (e.g., a predetermined area 170) of the user 140. In some examples, the plane 160 can indicate a plane parallel to the horizontal ground. In some examples, the predetermined area 170 can indicate the chest position of the user 140 (where the sternum is located). Then, after confirming that the terminal device 110 is placed in the predetermined area 170, it starts to detect at least one set of sequences associated with the cardiac vital signs signal of the user 140.

[0038] It should be understood that the structure and function of the various elements in the environments 100A and 100B are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure.

[0039] Some example embodiments of the present disclosure will continue to be described with reference to the accompanying drawings. It should be understood that the pages shown in the accompanying drawings are merely examples, and various page designs may actually exist. The various graphic elements in the page may have different arrangements and different visual representations, one or more of which may be omitted or replaced, and one or more other elements may also exist. The embodiments of the present disclosure are not limited in this respect. In addition, hereinafter, the example embodiments will be described primarily with respect to the terminal device 110. It should be understood that the actions described with respect to the terminal device 110 may be performed by the terminal device 110 in collaboration with its service end (e.g., server 130).

[0040] Figure 2 FIG. 2 is a schematic diagram of a target sensor structure 200 according to some embodiments of the present disclosure. The structure 200 may be implemented at the terminal device 110. For ease of discussion, reference will be made to FIG. Figure 1A Environment 100A and Figure 1B The structure 200 is described with reference to the environment 100B.

[0041] During the heartbeat, the heart contracts and relaxes alternately, which may be accompanied by twisting and rotation. These movements can be transmitted to the skin surface through the chest, so that corresponding signals can be detected by the target sensor 120. When the heart pumping function of a user at risk of heart failure is reduced, the impact of the heart on the chest wall will also be weakened.

[0042] like Figure 2 As shown in FIG. 1 , the target sensor 120 may include an accelerometer for measuring linear acceleration and a gyroscope for measuring angular velocity to detect multi-directional linear and rotational motion. Specifically, the data collected by the target sensor 120 may include X and Y corresponding to the accelerometers. 1 , Y 1 , Z 1 Direction and gyro X 2 , Y 2 , Z 2 In some examples, the target sensor 120 may be configured in the terminal device 110 .

[0043] Specifically, the program for collecting cardiac vibration data can be pre-configured in the target sensor 120 or the terminal device 110 (including the target sensor 120). In some embodiments, if it is determined that the target sensor 120 can be placed in an effective position for collecting cardiac vital signs signals, the terminal device 110 can obtain at least one set of sequences within a predetermined time interval. Then, the terminal device 110 determines the judgment result for the cardiac vital signs signal of the target user based on at least one set of sequences. In some embodiments, if it is determined that the target sensor 120 is not placed in a valid position, the terminal device 110 provides a prompt to adjust the position of the target sensor 120. In some embodiments, if the time for which the cardiac vital signs signal is not collected within the predetermined time interval exceeds a threshold duration, the terminal device 110 provides a prompt for re-detection. In some examples, the threshold duration may indicate 60s.

[0044] In some embodiments, the cardiac sign signal is a heart rhythm detection signal of the target user. In some examples, the heart rhythm detection signal may indicate a subtle vibration signal transmitted from the heart to the chest wall. In some examples, the effective position may indicate the chest position of the target user (where the sternum is located). In some examples, the target sensor 120 may indicate a micro-electromechanical sensor, such as a micro-electromechanical system (MEMS).

[0045] Figure 3 A schematic diagram of at least one sequence according to some embodiments of the present disclosure is shown. For ease of discussion, reference will be made to Figure 1A Environment 100A and Figure 1B The schematic diagram is described using environment 100B.

[0046] like Figure 3 As shown, at least one sequence collected by the target sensor 120 can be respectively obtained from the X 2 , Y 2 , Z 2 Orientation and accelerometer X 1 , Y 1 , Z 1 direction and record the detected chest wall vibration.

[0047] In some examples, the terminal device 110 can detect users with a risk of heart failure, and provide the corresponding six sets of sample sequences collected to the server to determine the characteristics of the cardiac vital signs signal with a risk of heart failure, so as to establish a standard feature for identifying heart failure. Then, the standard can be used to determine whether the user has a risk of heart failure.

[0048] Figure 4FIG. 4 is a schematic diagram of a process 400 for identifying a risk of heart failure according to some embodiments of the present disclosure. The process 400 may be implemented at the terminal device 110. For ease of discussion, reference will be made to Figure 1A Environment 100A and Figure 1B Process 400 is described with reference to environment 100B.

[0049] like Figure 4 As shown, process 400 may include a heart rhythm detection process and a heart failure discrimination process. In the heart failure detection process, the terminal device 110 may use a heart rhythm detection model to determine whether the six groups of sequences detected by the target sensor 120 are valid vital sign data. Then, in the heart failure discrimination process, the terminal device 110 may use a heart failure discrimination model to determine whether the participant has a risk of heart failure.

[0050] In some embodiments, the terminal device 110 may obtain at least one set of sequences associated with the cardiac vital sign signal of the target user from the target sensor 120. In some examples, the at least one set of sequences may respectively indicate the X 1 , Y 1 , Z 1 Direction and gyro X 2 , Y 2 , Z 2 Six sequences of directions.

[0051] In some embodiments, the terminal device 110 can determine whether valid vital sign data is included in at least one set of sequences based on a first machine learning model. In some embodiments, the first machine learning model is a machine learning model based on a decision tree model. In some embodiments, the first machine learning model is stored in the format of Open Neural Network Exchange (ONNX). It will be understood that this is merely exemplary and not restrictive. Then, during the model reasoning process, the terminal device 110 can use the first machine learning model to determine whether valid vital sign data is included in at least one set of sequences detected by the target sensor 120.

[0052] In some embodiments, the cardiac vital sign signal is a heart rhythm detection signal of the target user. In some embodiments, the terminal device 110 can determine a set of three-dimensional data corresponding to the heart rhythm detection signal based on at least one set of sequences. In some embodiments, the dimension of a set of three-dimensional data is related to the period, amplitude and area of ​​the heart rhythm detection signal. In some examples, a basic feature extraction operation can be performed on at least one set of sequences to determine a set of three-dimensional data. Then, the terminal device 110 can input a set of three-dimensional data into a first machine learning model. If the output of the first machine learning model indicates that a set of three-dimensional data includes valid heart rhythm data, the terminal device 110 can determine that at least one set of sequences includes valid vital sign data.

[0053] In some examples, at least one set of sequences may be divided into a plurality of data segments, and the plurality of data segments may be judged whether they contain valid vital sign data. In some examples, the valid vital sign data may indicate heartbeat rhythm data. In other examples, the fact that at least one set of sequences does not include valid vital sign data may indicate that at least one set of sequences includes non-heartbeat rhythm data. It is understood that this is merely exemplary and not restrictive.

[0054] In some embodiments, if at least one set of sequences includes valid vital sign data, at least one target feature associated with the valid vital sign data is determined. In some embodiments, the terminal device 110 can determine at least one target feature of the heart rhythm detection signal by performing Fourier transform on the valid vital sign data. In some embodiments, the at least one target feature can be characterized by the real part, imaginary part and phase of the heart rhythm detection signal obtained by Fourier transform.

[0055] In some embodiments, the terminal device 110 determines the judgment result for the target user's cardiac vital sign signal by providing at least one target feature and at least one set of sequences to a second machine learning model. In some examples, the second machine learning model may be a mixed expert model (MoE). In some examples, the second machine learning model may be based on at least one of convolution (CONV), residual shrinkage block unit (RSBu), global average pooling (GAP), long short-term memory network (LSTM), and gated addition tree ensemble (GATE).

[0056] Specifically, CONV can be used to extract features from the model input through the convolution kernel operation in the convolution layer, RSBu can be used to improve model performance and training effects by introducing residual connections and contraction operations, and the global maximum pooling operation can be used to extract the maximum value of each channel of the feature map to obtain a one-dimensional vector with the same number of channels. Each element of this vector represents the maximum activation value on the corresponding channel (that is, a global feature representation of each channel). LSTM can be used to output prediction features corresponding to the model input. In addition, gated add tree ensemble is a deep learning architecture that is combined with a differentiable and nonlinear decision tree set and reweighted through self-attention to predict the desired output.

[0057] In some embodiments, the second machine learning model can be determined in the following manner. The terminal device 110 uses the target sensor 120 to determine at least one set of sample sequences associated with at least one user having a preset feature. If the at least one set of sample sequences indicates valid vital sign data, the terminal device 110 determines at least one sample target feature associated with the at least one set of sample sequences. Then, the terminal device 110 determines the second machine learning model based on the at least one set of sample sequences and the at least one sample target feature.

[0058] In some embodiments, the terminal device 110 obtains a judgment result of the cardiac vital sign signal of the target user from the output of the second machine learning model, and the judgment result indicates whether the target user is at risk of heart failure.

[0059] In some examples, the terminal device 110 may provide at least one target feature and at least one set of sequences to the second machine learning model in the form of a feature graph. Then, the features obtained from several parts of the second machine learning model are spliced. Then, the spliced ​​features are paired with a gated addition tree set to output a predicted judgment result.

[0060] In some examples, the accuracy, recall (sensitivity), and precision (specificity) of the first machine learning model and the second machine learning model can be determined by the following formulas, respectively:

[0061] Accuracy = (K 1 +K 2 ) / K×100% (1)

[0062] Among them, K 1 represents the number of users with risk of heart failure who are judged to have risk of heart failure, K 2 The judgment result of users who are not at risk of heart failure is the number of people who are not at risk of heart failure, and K represents the total number of samples.

[0063] Recall = K 1 / (K 1 +K 3 ) × 100% (2)

[0064] Among them, K 1 represents the number of users with risk of heart failure who are judged to have risk of heart failure, K 3 Indicates the number of users with heart failure risk who are judged as not having heart failure risk.

[0065] Precision = K 2 / (K 2 +K 4 ) × 100% (3)

[0066] Among them, K 2The number of users who are not at risk of heart failure is determined to be not at risk of heart failure, K 4 Indicates the number of users who are not at risk of heart failure but are judged to be at risk of heart failure.

[0067] By using the collected data of the user group without heart failure risk and the user group with heart failure to determine the judgment result of the heart vital sign signal for each user using this solution, the accuracy rate is about 94.58%, the recall rate is about 92.16%, and the precision rate is about 95.39%.

[0068] The data of accuracy, recall and precision in some embodiments of the present disclosure are as follows: Table 1-Table 3:

[0069] Table 1 Model performance data table

[0070]

[0071]

[0072] Table 2 Data table of judgment results of different types of users

[0073]

[0074] Table 3 Model performance data

[0075]

[0076] Figure 5A-5C Schematic diagrams of example interfaces 500A-500C of unstarted heart failure detection according to some embodiments of the present disclosure are shown. Figure 5A-5C Example interfaces 500A-500C are described. For ease of discussion, reference will be made to Figure 1A Environment 100A and Figure 1B The interfaces 500A-500C are described with reference to the environment 100B.

[0077] like Figure 5A The example interface 500A shown may include a detection display area 510. In some examples, the terminal device 110 may present an image 512 for displaying a preset detection gesture in the detection display area 510. It is understood that this is merely exemplary and not limiting.

[0078] The terminal device 110 may also present prompt information for detection on the interface 500A, such as "Place the target sensor flat on the sternum", "Keep the body still and breathe steadily", "Please wear a thin piece of clothing", etc. It is understood that this is merely exemplary and not restrictive, and may actually be any reasonable prompt information.

[0079] The terminal device 110 may also present precautions for testing on the interface 500A, such as "The test results of this product are for reference only. They cannot be used for medical diagnosis and treatment purposes. If you feel unwell, please use professional equipment for testing or seek medical attention immediately."

[0080] The terminal device 110 may also present a "start detection" control 513 on the interface 500A. In some examples, the user may predetermine a trigger operation to trigger the control 513 to start detection. In some examples, the predetermined trigger operation may indicate one of clicking, double-clicking, or hovering. It is understood that this is merely exemplary and not restrictive.

[0081] After detecting that the control 513 is triggered, the terminal device 110 may present the interface 500B.

[0082] like Figure 5B As shown, the terminal device 110 may present a prompt 521 for displaying the detection status on the interface 500B, such as “Detecting the placement of the target sensor, please wait.” It is understood that this is merely exemplary and not restrictive.

[0083] The terminal device 110 may also present a "end detection" control 522 in the interface 500B. In some examples, the user may predetermine a trigger operation to trigger the control 522 to end the detection. In some examples, the predetermined trigger operation may indicate one of clicking, double-clicking, or hovering. It is understood that this is merely exemplary and not restrictive.

[0084] After the terminal device 110 presents the interface 500B, if the terminal device 110 detects that the target sensor is not placed in a valid position for collecting cardiac vital sign signals, the interface 500C may be presented.

[0085] like Figure 5C The example interface 500C shown may present a prompt 531 for showing that the position of the target sensor is abnormal, such as “No heart rhythm detected, please try again!” It can be understood that this is merely exemplary and not limiting.

[0086] The terminal device 110 may present a "start new detection" control 532 in the interface 500C. In some examples, the user may predetermine a trigger operation to trigger the control 532 to start a new detection. In some examples, the predetermined trigger operation may indicate one of clicking, double-clicking, or hovering. It is understood that this is merely exemplary and not restrictive.

[0087] Figure 6A-6D Schematic diagrams showing example interfaces 600A-600D in a heart failure detection process according to some embodiments of the present disclosure. Figure 6A-6DExample interfaces 600A-600D are described. For ease of discussion, reference will be made to Figure 1A Environment 100A and Figure 1B The interfaces 600A-600D are described with reference to the environment 100B.

[0088] like Fig. 6A In the example interface 600A shown, the terminal device 110 can display the detection time (e.g., 120s) in area 610, and can also present a detection status prompt 611 (e.g., "Data collection, please remain still") and a collected cardiac vital sign signal image 612 on the interface 600A. It is understood that this is merely exemplary and not restrictive.

[0089] After the detection starts on the presentation interface 600A, if an abnormality is found, the presentation interface 600B is presented on the terminal device 110 .

[0090] like Figure 6B As shown, the terminal device 110 can present the unused remaining detection time (e.g., 110s) in area 620 of the interface 600B. Furthermore, the terminal device 110 can present a prompt 621 for displaying abnormal conditions (e.g., data collection is abnormal, please reposition the target sensor) and a collected cardiac vital sign signal image 622 on the interface 600B.

[0091] like Figure 6C As shown, the terminal device 110 can present the remaining detection time (e.g., 60s) in area 630 of interface 600B. Furthermore, the terminal device 110 can present a prompt 631 (e.g., data collection in progress, please remain still) for displaying the collection status and a collected cardiac vital sign signal image 632 on interface 600C.

[0092] After the cardiac vital sign signal collection is completed, the example interface 600D may be presented on the terminal device 110 .

[0093] like Fig.6D As shown, the terminal device can present the remaining detection time (eg, 0s) in area 640 of interface 600D. Furthermore, the terminal device 110 can present a prompt 641 (eg, detection completed, chart being generated) for showing that the collection is completed on interface 600D.

[0094] Figure 7A-7C Schematic diagrams of example interfaces 700A-700C for displaying judgment results according to some embodiments of the present disclosure are shown. Figure 7A-7C Example interfaces 700A-700C are described. For ease of discussion, reference will be made to Figure 1A Environment 100A and Figure 1B The interfaces 700A-700C are described with reference to the environment 100B.

[0095] If it is determined that the target user has no obvious abnormality, that is, the target user has no risk of heart failure, the terminal device 110 may Fig. 7A The area 710 in the example interface 700A shown presents the judgment result. Specifically, the area 710 may also include judgment result prompt information 711 (for example, your result has no obvious abnormality) and a cardiac vital sign signal image 712.

[0096] If the target user is determined to be suspected of having a risk of heart failure, the terminal device 110 may Figure 7B The area 720 in the example interface 700B shown presents the judgment result. Specifically, the area 720 may also include judgment result prompt information 721 (for example, your result has no obvious abnormality) and a cardiac vital sign signal image 722.

[0097] If insufficient data is collected, the terminal device 110 may Figure 7C Area 730 in the example interface 700C shown presents a prompt 731 asking the user to retest (eg, insufficient data has been collected, please retest).

[0098] In summary, the embodiments of the present disclosure can use the target sensor to detect the heart vital signs of the target user to determine whether the user has a risk of heart failure. In this way, the difficulty of detection can be reduced and the convenience of the detection process can be improved.

[0099] Figure 8 FIG. 8 is a flowchart showing a process 800 of processing cardiac monitoring data according to some embodiments of the present disclosure. The process 800 may be implemented at the terminal device 110. For ease of discussion, reference will be made to FIG. Figure 1A Environment 100A and Figure 1B Process 800 is described with reference to environment 100B.

[0100] In block 810 , the terminal device 110 acquires at least one set of sequences associated with a cardiac vital sign signal of a target user from a target sensor.

[0101] In box 820, the terminal device 110 determines whether at least one group of sequences includes valid vital sign data based on the first machine learning model.

[0102] In block 830 , in response to valid vital sign data being included in at least one set of sequences, the terminal device 110 determines at least one target feature associated with the valid vital sign data.

[0103] In box 840, the terminal device 110 determines a judgment result of the cardiac vital sign signal for the target user by providing at least one target feature and at least one set of sequences to a second machine learning model.

[0104] In some embodiments, acquiring at least one set of sequences includes: determining whether the target sensor is placed at a valid position for acquiring cardiac vital sign signals; and if it is determined that the target sensor is placed at a valid position, acquiring at least one set of sequences within a predetermined time interval.

[0105] In some embodiments, process 800 further includes providing a prompt to adjust the position of the target sensor if it is determined that the target sensor is not positioned in a valid position.

[0106] In some embodiments, process 800 further includes: providing a prompt for retesting if the time for which the cardiac vital sign signal is not collected within the predetermined time interval exceeds a threshold time period.

[0107] In some embodiments, the cardiac vital sign signal is a cardiac rhythm detection signal of the target user. In other embodiments, determining whether at least one set of sequences includes valid vital sign data includes: determining a set of three-dimensional data corresponding to the cardiac rhythm detection signal based on at least one set of sequences, wherein the dimension of the set of three-dimensional data is related to the period, amplitude and area of ​​the cardiac rhythm detection signal; inputting the set of three-dimensional data into a first machine learning model, wherein the first machine learning model is a machine learning model based on a decision tree model; and in response to an output of the first machine learning model indicating that a set of three-dimensional data includes valid cardiac rhythm data, determining that at least one set of sequences includes valid vital sign data.

[0108] In some embodiments, the second machine learning model is determined in the following manner: using a target sensor to determine at least one group of sample sequences associated with at least one user having preset characteristics; in response to at least one group of sample sequences indicating valid vital sign data, determining at least one sample target feature associated with at least one group of sample sequences; and determining the second machine learning model based on at least one group of sample sequences and at least one sample target feature.

[0109] In some embodiments, the cardiac vital sign signal is a heart rhythm detection signal of the target user. In other embodiments, determining at least one target feature includes: determining at least one target feature of the heart rhythm detection signal by performing Fourier transform on the effective vital sign data, wherein the at least one target feature is characterized by the real part, imaginary part and phase of the heart rhythm detection signal obtained by Fourier transform.

[0110] In some embodiments, determining the judgment result includes: obtaining a judgment result of a cardiac vital sign signal for the target user from an output of a second machine learning model, the judgment result indicating whether the target user is at risk of heart failure.

[0111] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes.

[0112] Fig. 9A schematic structural block diagram of an apparatus 900 for processing cardiac monitoring data according to some embodiments of the present disclosure is shown. The apparatus 900 may be implemented as or included in a terminal device 110. Each module / component in the apparatus 900 may be implemented by hardware, software, firmware, or any combination thereof.

[0113] like Fig. 9 As shown, the device 900 includes an acquisition module 910, which is configured to acquire at least one group of sequences associated with the cardiac vital sign signal of the target user from the target sensor; the device 900 also includes a data determination module 920, which is configured to determine whether the at least one group of sequences includes valid vital sign data based on a first machine learning model; the device 900 also includes a feature determination module 930, which is configured to determine at least one target feature associated with the valid vital sign data in response to the at least one group of sequences including valid vital sign data; the device 900 also includes a providing module 940, which is configured to determine a judgment result for the cardiac vital sign signal of the target user by providing the at least one target feature and the at least one group of sequences to a second machine learning model.

[0114] In some embodiments, the acquisition module 910 is further configured to determine whether the target sensor is placed at a valid position for collecting cardiac vital sign signals; and if it is determined that the target sensor is placed at a valid position, acquire at least one set of sequences within a predetermined time interval.

[0115] In some embodiments, the apparatus 900 further includes a first prompt module configured to provide a prompt for adjusting the position of the target sensor if it is determined that the target sensor is not placed in a valid position.

[0116] In some embodiments, the device 900 further includes a second prompt module configured to provide a prompt for retesting if the time for which the cardiac vital sign signal is not collected within a predetermined time interval exceeds a threshold duration.

[0117] In some embodiments, the cardiac vital sign signal is a cardiac rhythm detection signal of the target user. In other embodiments, the data determination module 920 is further configured to determine a set of three-dimensional data corresponding to the cardiac rhythm detection signal based on at least one set of sequences, wherein the dimension of the set of three-dimensional data is related to the period, amplitude and area of ​​the cardiac rhythm detection signal; input the set of three-dimensional data into a first machine learning model, wherein the first machine learning model is a machine learning model based on a decision tree model; and in response to the output of the first machine learning model indicating that a set of three-dimensional data includes valid cardiac pulsation data, determine that at least one set of sequences includes valid vital sign data.

[0118] In some embodiments, the second machine learning model is determined in the following manner: using a target sensor to determine at least one group of sample sequences associated with at least one user having preset characteristics; in response to at least one group of sample sequences indicating valid vital sign data, determining at least one sample target feature associated with at least one group of sample sequences; and determining the second machine learning model based on at least one group of sample sequences and at least one sample target feature.

[0119] In some embodiments, the cardiac vital sign signal is a cardiac rhythm detection signal of the target user. In other embodiments, the feature determination module 930 is further configured to determine at least one target feature of the cardiac rhythm detection signal by performing Fourier transform on the effective vital sign data, wherein the at least one target feature is characterized by the real part, imaginary part and phase of the cardiac rhythm detection signal obtained by Fourier transform.

[0120] In some embodiments, the providing module 940 is also configured to obtain a judgment result of the cardiac vital sign signal of the target user from the output of the second machine learning model, and the judgment result indicates whether the target user is at risk of heart failure.

[0121] The units and / or modules included in the device 900 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units and / or modules can be implemented using software and / or firmware, such as machine executable instructions stored on a storage medium. In addition to or as an alternative to machine executable instructions, some or all of the units and / or modules in the device 900 can be implemented at least in part by one or more hardware logic components. As an example and not limitation, exemplary types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0122] It should be understood that one or more steps in the above method can be performed by a suitable electronic device or a combination of electronic devices. Such an electronic device or a combination of electronic devices may include, for example, Figure 1A The terminal device 110 in.

[0123] Fig.10 1 shows a block diagram of an electronic device 1000 in which one or more embodiments of the present disclosure may be implemented. It should be understood that Fig.10 The electronic device 1000 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein. Fig.10 The electronic device 1000 shown can be used to implement Figure 1A terminal device 110.

[0124] like Fig.10 As shown, the electronic device 1000 is in the form of a general electronic device. The components of the electronic device 1000 may include, but are not limited to, one or more processors or processing units 1010, a memory 1020, a storage device 1030, one or more communication units 1040, one or more input devices 1050, and one or more output devices 1060. The processing unit 1010 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 1020. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capability of the electronic device 1000.

[0125] The electronic device 1000 typically includes a plurality of computer storage media. Such media may be any available media accessible to the electronic device 1000, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 1020 may be a volatile memory (e.g., a register, a cache, a random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 1030 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data and may be accessed within the electronic device 1000.

[0126] The electronic device 1000 may further include additional removable / non-removable, volatile / non-volatile storage media. Fig.10 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. Memory 1020 may include a computer program product 1025 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0127] The communication unit 1040 implements communication with other electronic devices through a communication medium. Additionally, the functions of the components of the electronic device 1000 can be implemented in a single computing cluster or multiple computing machines that can communicate through a communication connection. Therefore, the electronic device 1000 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0128] The input device 1050 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 1060 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 1000 may also communicate with one or more external devices (not shown) through the communication unit 1040 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device 1000, or communicate with any device that allows the electronic device 1000 to communicate with one or more other electronic devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).

[0129] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above.

[0130] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0131] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0132] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0133] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0134] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.

Claims

1. A method for processing heart monitoring data, comprising: Acquire at least one set of sequences associated with cardiac vital sign signals of a target user from a target sensor; Based on the first machine learning model, determining whether the at least one set of sequences includes valid vital sign data; In response to valid vital sign data being included in the at least one set of sequences, determining at least one target feature associated with the valid vital sign data; as well as By providing the at least one target feature and the at least one group of sequences to a second machine learning model, a judgment result of the cardiac vital sign signal of the target user is determined.

2. The method according to claim 1, wherein obtaining the at least one set of sequences comprises: determining whether the target sensor is placed in a valid position for collecting the cardiac vital sign signal; as well as If it is determined that the target sensor is placed at the effective position, the at least one set of sequences within a predetermined time interval is acquired.

3. The method according to claim 2, further comprising: If it is determined that the target sensor is not placed in the valid position, a prompt is provided to adjust the position of the target sensor.

4. The method according to claim 2, further comprising: If the time for which the cardiac vital sign signal is not collected within the predetermined time interval exceeds a threshold time period, a prompt for re-detection is provided.

5. The method according to claim 1, wherein the cardiac vital sign signal is a heart rhythm detection signal of the target user, and determining whether the at least one set of sequences includes the valid vital sign data comprises: Determine a set of three-dimensional data corresponding to the heart rhythm detection signal based on the at least one set of sequences, wherein the dimension of the set of three-dimensional data is related to the period, amplitude and area of ​​the heart rhythm detection signal; Inputting the set of three-dimensional data into the first machine learning model, wherein the first machine learning model is a machine learning model based on a decision tree model; as well as In response to an output of the first machine learning model indicating that the set of three-dimensional data includes valid cardiac pulsation data, it is determined that the at least one set of sequences includes the valid vital sign data.

6. The method according to claim 1, wherein the second machine learning model is determined in the following manner: Determining, using a target sensor, at least one set of sample sequences associated with at least one user having a preset characteristic; In response to the at least one set of sample sequences indicating valid vital sign data, determining at least one sample target feature associated with the at least one set of sample sequences; and The second machine learning model is determined based on the at least one group of sample sequences and the at least one sample target feature.

7. The method according to claim 1, wherein the cardiac sign signal is a heart rhythm detection signal of the target user, and wherein determining the at least one target feature comprises: The at least one target feature of the heart rhythm detection signal is determined by performing Fourier transform on the effective vital sign data, and the at least one target feature is characterized by the real part, imaginary part and phase of the heart rhythm detection signal obtained by the Fourier transform.

8. The method according to claim 7, wherein determining the judgment result comprises: The judgment result of the cardiac vital sign signal for the target user is obtained from the output of the second machine learning model, and the judgment result indicates whether the target user has a risk of heart failure.

9. A device for processing heart monitoring data, comprising: An acquisition module, configured to acquire at least one set of sequences associated with a cardiac vital sign signal of a target user from a target sensor; A data determination module is configured to determine whether the at least one set of sequences includes valid vital sign data based on a first machine learning model; a feature determination module, configured to determine at least one target feature associated with the valid vital sign data in response to the at least one set of sequences including the valid vital sign data; as well as A module is provided, configured to determine a judgment result for the cardiac vital sign signal of the target user by providing the at least one target feature and the at least one group of sequences to a second machine learning model.

10. An electronic device comprising: at least one processing unit; as well as At least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, cause the electronic device to perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the method according to any one of claims 1 to 8.

12. A computer program product comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 8.

Images