Household intelligent medical interaction method and system and storage medium

Abstract

The invention discloses a household intelligent medical interaction method and system and a storage medium. Comprising a data acquisition interface module used for connecting one or more health detection devices and acquiring one or more health detection data; the semantic mapping module is used for judging whether the health detection data is abnormal or not, and if the health detection data is abnormal, analyzing an abnormal reason based on the personalized life data of the user to obtain a semantic mapping result; the natural language generation module is used for converting the semantic mapping result into a spoken health interpretation text by adopting a preset extensible template library; and the dialogue interaction module is used for identifying a user voice instruction, identifying a user emotion tag and an identity tag based on the user voice instruction, determining a personalized output mode based on the emotion tag and the identity tag, and returning a health explanation text by adopting the personalized output mode. According to the method, the measured value is directly converted into oral health interpretation, personalized output can be carried out according to the emotion and identity of the user, and the user experience is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent medical treatment, in particular to a family intelligent medical treatment interaction method and system and a storage medium. BACKGROUND

[0002] With the increasing attention to health, family health detection devices such as sphygmomanometers and blood glucose meters have gradually become popular. These devices can provide users with real-time health data conveniently and quickly. However, the current health detection devices only directly display detection data, and ordinary users lack medical knowledge and are difficult to understand the clinical significance of the data. The user experience is low. And cross-device data is difficult to manage and record uniformly, which seriously affects the efficiency and quality of family health management. SUMMARY

[0003] The embodiments of the present application provide a family intelligent medical treatment interaction method and system and a storage medium to at least solve the technical problem that home medical detection devices are difficult to interact efficiently with users in the related art.

[0004] According to an aspect of an embodiment of the present application, a family intelligent medical treatment interaction system is provided, comprising:

[0005] A data acquisition interface module is configured to connect one or more health detection devices and acquire one or more health detection data.

[0006] A semantic mapping module is configured to determine whether the health detection data is abnormal, and if the health detection data is abnormal, analyze the abnormal reason based on user personalized life data to obtain a semantic mapping result.

[0007] A natural language generation module is configured to convert the semantic mapping result into a colloquial health interpretation text by using a preset extensible template library.

[0008] A dialogue interaction module is configured to recognize a user voice instruction, recognize a user emotion label and an identity label based on the user voice instruction, determine a personalized output mode based on the emotion label and the identity label, and return the health interpretation text by using the personalized output mode.

[0009] In an embodiment, the system further comprises a data processing module, which comprises:

[0010] A protocol adaptive parsing unit is configured to identify protocols used by different devices, load a corresponding regular template according to the identified device protocol, parse corresponding health detection data by using the regular template, and map the parsed result to a unified medical data standard format.

[0011] A time sequence alignment unit is configured to perform time calibration and synchronization on the parsed data.

[0012] In an embodiment, the semantic mapping module comprises:

[0013] an anomaly judgment unit configured to obtain a preset normal value interval, a high value interval and a low value interval, match the health detection data with the normal value interval, the high value interval and the low value interval respectively, and obtain a judgment result;

[0014] a personalized correction unit configured to obtain the health detection data of a historical period of a user, calculate statistical data of the health detection data, and correct the judgment result based on the statistical data;

[0015] a scenario semantic unit configured to obtain user personalized life data, and analyze an abnormal reason based on the user personalized life data.

[0016] In an embodiment, obtaining the user personalized life data comprises:

[0017] obtaining a geographic location of the user, and collecting weather data corresponding to the geographic location;

[0018] obtaining a detection time, a diet log and an activity record of the user;

[0019] determining the user personalized life data based on the detection time, the diet log, the activity record and the weather data.

[0020] In an embodiment, the dialogue interaction module comprises:

[0021] a voice recognition unit configured to recognize a user voice instruction;

[0022] an identity recognition unit configured to extract user voiceprint information based on the voice instruction, and determine a user identity tag based on the voiceprint information and / or image information;

[0023] an emotion recognition unit configured to determine a user emotion tag based on the user voice instruction, a facial image and detection data;

[0024] an output unit configured to determine a personalized output mode based on the emotion tag and the identity tag, and return the health interpretation text in the personalized output mode.

[0025] In an embodiment, the system further comprises:

[0026] an active intervention module configured to analyze whether a user has a trend abnormality based on the health detection data of a preset period, and perform active early warning based on a hierarchical early warning strategy in the case that the user has the trend abnormality.

[0027] In an implementation, the active intervention module comprises:

[0028] An anomaly detection unit is configured to record health detection data of the preset time period by using a sliding window, and detect whether there is a trend anomaly of the user based on the health detection data in the sliding window.

[0029] A hierarchical early warning unit is configured to determine a warning level and a corresponding early warning strategy in the case of an anomaly, and perform active early warning based on the early warning strategy.

[0030] In an implementation, the method further comprises:

[0031] A storage module is configured to store the health detection data, and generate a trend analysis report according to different time dimensions based on the health detection data.

[0032] According to another aspect of the embodiments of the present application, a home intelligent medical interaction method is provided, which comprises:

[0033] Recognizing a user voice instruction;

[0034] Connecting one or more health detection devices to collect one or more health detection data;

[0035] Judging whether the health detection data is abnormal, and if the health detection data is abnormal, analyzing an abnormal reason based on user individualized life data to obtain a semantic mapping result;

[0036] Converting the semantic mapping result into a colloquial health explanation text by using a preset extensible template library;

[0037] Identifying a user emotion label and an identity label based on the user voice instruction, determining an individualized output mode based on the emotion label and the identity label, and returning the health explanation text by using the individualized output mode.

[0038] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided, which stores a computer program, wherein the computer program is set to execute the home intelligent medical interaction method when running.

[0039] The technical scheme provided by the embodiments of the present application can have the following beneficial effects:

[0040] The family intelligent medical interaction system of the present application realizes the intelligent management of the whole process from data collection to personalized health interpretation by integrating multiple modules. The data collection interface module can connect multiple health detection devices and collect various health detection data, solving the problem of difficult unified management of cross-device data. This enables users to centrally manage data from different devices, making it easy to conduct comprehensive analysis. The natural language generation module converts complex health data and analysis results into colloquial health interpretation text, enabling ordinary users to easily understand the clinical significance of the data and enhancing their acceptance and understanding of health data. Further, the system determines the personalized output mode by identifying the emotional label and identity label of the user, and can provide more thoughtful and appropriate responses according to the emotional state and identity background of different users. This significantly improves the efficiency and quality of family health management, providing a comprehensive, convenient and personalized health management solution for users. BRIEF DESCRIPTION OF DRAWINGS

[0041] The drawings described herein are intended to provide further understanding of the present application, form a part of the present application, and the illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0042] Figure 1 is a schematic diagram of an optional family intelligent medical interaction system according to an embodiment of the present application;

[0043] Figure 2 is a schematic diagram of a family intelligent medical interaction system according to an embodiment of the present application;

[0044] Figure 3 is a schematic diagram of a family intelligent medical interaction method according to an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to better enable those skilled in the art to understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] This application addresses the problems existing in the prior art by proposing a home-based intelligent medical interaction system. This system is primarily deployed in the home environment and connects to various health monitoring devices such as blood pressure monitors and blood glucose meters via a data acquisition interface module to collect health data in real time. The system utilizes a semantic mapping module to analyze the data and identify anomalies, combining this with the user's personalized lifestyle data to provide the reasons for the anomalies. A natural language generation module converts the analysis results into spoken text, and a dialogue interaction module recognizes the user's voice commands and emotional identity tags, outputting personalized health explanations to help users better understand and manage their own health. This system is suitable for long-term home health management scenarios.

[0048] The following is a detailed description of the home intelligent medical interaction system according to an embodiment of this application, with reference to the accompanying drawings. Figure 1 As shown, the system mainly includes the following modules: data acquisition interface module, semantic mapping module, natural language generation module, and dialogue interaction module.

[0049] The data acquisition interface module is used to connect to one or more health monitoring devices and collect one or more health monitoring data.

[0050] In one implementation, the data acquisition interface module is used to connect to one or more health monitoring devices such as a blood pressure monitor, blood glucose meter, thermometer, pulse oximeter, smart bracelet, and fetal heart monitor to collect one or more health monitoring data.

[0051] Specifically, the data acquisition interface module integrates multiple communication protocols such as Bluetooth Low Energy (BLE), Wi-Fi, and USB to connect with different types of health monitoring devices. During device connection, the module first uses a device fingerprint recognition algorithm to quickly and accurately match and determine the specific model and brand of the connected device. This provides precise device information support for subsequent data acquisition and processing, ensuring the system can adapt and optimize for different devices.

[0052] In an embodiment, the device fingerprint identification collects input data such as BLE broadcast packet fields, USB VID / PID, Wi-Fi MAC address prefix, extracts the original feature vector, and matches the local drive table. After calculating the device fingerprint hash, the input is input into a lightweight classifier to predict the device category and confidence. If the confidence exceeds the preset threshold, the online analysis strategy is triggered or the user is requested to confirm, while maintaining an online-updatable device drive table to achieve accurate device identification and management.

[0053] As shown in Figure 2 The system of the present application also includes a data processing module, which includes:

[0054] A protocol adaptive parsing unit is configured to identify the protocols used by different devices, load the corresponding regular templates according to the identified device protocols, parse the corresponding health detection data using the regular templates, and map the parsing results to a unified medical data standard format.

[0055] Specifically, after identifying the connected device information, the system will load the regular expression template corresponding to the protocol from the protocol template library. The protocol template library is a collection of multiple protocol parsing rules, and each protocol has its own regular expression template, field mapping table, and byte order and unit information. The regular expression template is used to accurately match and extract the key fields in the data packet.

[0056] Further, the received data is compared with the regular expression templates in the protocol template library one by one to find the matching template. The key fields such as measurement values and timestamps in the data are extracted using the matched regular expression template. According to the field mapping table and the byte order / unit information, the extracted field values are converted to a unified unit and format. The parsed and converted data is mapped to a unified medical data standard format, such as the FHIR (Fast Healthcare Interoperability Resources) Observation resource. FHIR is an internationally recognized medical information exchange standard that ensures interoperability of data between different systems.

[0057] For data formats that do not match existing templates, the system enables a semi-supervised learning mechanism. By analyzing the structure and content of the unmatched data, a new regular expression template is automatically generated and stored in the protocol template library.

[0058] It also includes a time alignment unit for time calibration and synchronization of the parsed data. The system performs time calibration and synchronization storage of medical data from different devices based on event timestamps and network time protocol, ensuring the time consistency and accuracy of the data.

[0059] The system of the present application further comprises a semantic mapping module for determining whether the health detection data is abnormal, and if the health detection data is abnormal, analyzing the abnormal reason based on the user's personalized life data to obtain a semantic mapping result.

[0060] In an embodiment, the semantic mapping module comprises:

[0061] An abnormality determination unit is configured to obtain a preset normal value interval, a high value interval and a low value interval, match the health detection data with the normal value interval, the high value interval and the low value interval respectively, and obtain a determination result.

[0062] Specifically, the system obtains the normal, high and low value intervals of the health indicators from authoritative medical guidelines, matches the real-time collected health detection data with these intervals, determines the normal, high or low state of the values according to the matching results, and outputs the results. For example, the medical guidelines stipulate that the normal blood pressure interval is 90-139 mmHg for systolic pressure and 60-89 mmHg for diastolic pressure, the high interval is 140-159 mmHg for systolic pressure and 90-99 mmHg for diastolic pressure, and the low interval is less than 90 mmHg for systolic pressure and less than 60 mmHg for diastolic pressure. If the systolic pressure is detected to be 145 mmHg and the diastolic pressure is 92 mmHg, the system determines that the blood pressure data is in a high state after matching.

[0063] A personalized correction unit is configured to obtain the health detection data of the user in a historical period, calculate the statistical data of the health detection data, and correct the determination result based on the statistical data.

[0064] Specifically, the health detection data of the user in a historical period is obtained, for example, the health detection data of the user in the last 30 days or the last 100 times is obtained. The statistical quantities such as mean and standard deviation of these data are calculated, and then the preliminary determination result is corrected based on these statistical quantities. If the current detection value exceeds the historical mean plus 1.5 times the standard deviation, the detection result is marked as "abnormal relative to personal history", so as to realize the health state determination more in line with the individual characteristics of the user.

[0065] A scenario semantic unit is configured to obtain the user's personalized life data, and analyze the abnormal reason based on the user's personalized life data.

[0066] Specifically, the geographic location of the user is obtained, and the weather data corresponding to the geographic location is collected, including temperature, humidity, air quality, etc. The detection time, diet log and activity record of the user are obtained. The user's personalized life data is determined based on the detection time, diet log, activity record and weather data.

[0067] Further, the user's personalized life data is analyzed to find the abnormal reasons. By comprehensively analyzing the measurement time (such as before meal, after meal, after exercise), recent behavior log (including diet, medication, exercise record), and environmental factors such as weather and geography, the health detection data is analyzed to find the abnormal reasons. For example, if the user measures the blood glucose within 30 minutes after meal and the value increases, the system will combine the measurement time and diet log to output the explanation "may be related to the meal just eaten", thereby providing the user with more practical health suggestions and avoiding misunderstanding caused by single data interpretation.

[0068] In summary, the semantic mapping module can analyze whether the data is abnormal based on the detection data, and correct the results and analyze the reasons. More accurate measurement results and reasons are obtained.

[0069] The natural language generation module is configured to convert the semantic mapping results into colloquial health interpretation texts by using a preset extensible template library.

[0070] Specifically, the natural language generation module converts the semantic mapping results into colloquial health interpretation texts by using a preset extensible template library. The module selects the corresponding template and fills in the specific data according to different semantic mapping results to generate easy-to-understand health interpretations. For example, if it is detected that the user's heart rate is 85 beats per minute, which is higher than the average of 75 beats per minute last week, and shows an upward trend, the module will generate the following text: "Today's heart rate is 85 beats per minute, which is higher than the average of 75 beats per minute last week. In the last 7 days, the heart rate shows an upward trend. Suggestions: pay attention to rest and avoid strenuous exercise. If the heart rate continues to increase, please seek medical attention immediately." The reason analysis can also be output if the semantic mapping result analyzes that the reason is "may be related to the strenuous exercise just completed". The colloquial explanation helps the user better understand the health data and take appropriate measures.

[0071] The dialog interaction module is also included, which is configured to recognize user voice instructions, identify user emotion labels and identity labels based on the user voice instructions, determine a personalized output mode based on the emotion labels and identity labels, and return health interpretation texts by using the personalized output mode.

[0072] In one embodiment, the dialog interaction module includes a voice recognition unit configured to recognize user voice instructions. In one example scenario, the user can actively initiate a voice query, for example, the user asks "how is today's blood pressure and sleep condition".

[0073] The identity recognition unit is configured to extract user voiceprint information based on the voice instructions, and determine the user identity label based on the voiceprint information and / or image information.

[0074] In an embodiment, the system is equipped with a camera to capture the user's face image. The user's identity can be recognized based on the captured face image.

[0075] The user's voiceprint information can also be extracted based on the user's voice instruction. The system receives the user's voice instruction and pre-processes it, including noise reduction, voice activity detection, etc., to ensure the quality of the voice signal. Then, voiceprint features such as Mel Frequency Cepstral Coefficients (MFCC) are extracted from the processed voice signal, which can reflect the unique voice characteristics of the speaker. Then, the extracted voiceprint features are compared with the pre-stored user voiceprint template, usually using a similarity measurement method to calculate the similarity between the features. If the similarity exceeds the set threshold, it is considered that the voice instruction comes from a registered user, and its identity label is determined. If it is below the threshold, it is determined as an unknown user. In addition, the system can also combine image information (such as face recognition) to further enhance the accuracy and security of identity recognition, achieving multi-modal identity authentication.

[0076] An emotion recognition unit is used to determine the user's emotion label based on the user's voice instruction, face image, and detection data.

[0077] Specifically, the present application proposes an emotion recognition method based on multi-modal data. The user's voice signal, face video image, and physiological parameters are collected. Voiceprint features are extracted from the voice signal, expression feature vectors are extracted from the face video image, and physiological parameters are analyzed. The emotion dynamic adaptation engine uses cross-modal Transformer or Bayesian fusion method to fuse these multi-modal feature vectors, calculates the user's emotion label and confidence, and outputs the user's emotion distribution and confidence. For example, the system analyzes and obtains the user's emotion distribution as follows: anxiety: 0.8, calm: 0.2, and gives the confidence. It indicates that the user is likely to be in an anxious state.

[0078] For example, when the user's voice tone is raised, the facial expression is tense, and the heart rate is accelerated, the system integrates these features and determines that the user may be in an anxious state through the model, and outputs the corresponding emotion label and confidence.

[0079] An output unit is used to determine a personalized output method based on the emotion label and the identity label, and returns a health interpretation text using the personalized output method.

[0080] Specifically, the user's emotion label and identity label are obtained. Different output methods are set according to the identity of different users (such as the elderly, young people, doctors, patients, etc.), including volume setting, speech rate setting, etc.

[0081] According to the user's emotional state (such as anxiety, calm, anger, etc.), adjust the speech speed, tone, word choice, and interaction strategy to better calm or guide the user. In combination with the above strategies, obtain health interpretation text from the natural language generation module and return it to the user through speech synthesis or text display.

[0082] For example, for an elderly person: use simple and easy-to-understand language, slower speech speed, higher volume, repeat key information to ensure understanding. If the user is a male elderly person, the system will say: "Grandpa, your blood pressure is a bit high, today's measurement is 145, which is higher than usual. Have you been feeling unwell recently? If so, it's best to go to the hospital."

[0083] For example, for a user who is anxious: use soothing language, slower speech speed, gentle tone, and provide specific suggestions. If the user is anxious, the system will say: "I understand that you may be a bit worried right now, your blood sugar is a bit high, but don't worry, it may be related to your recent diet. It is recommended that you eat less sweets and exercise more, and if you are still not at ease, you can consult a doctor.

[0084] In one embodiment, it further comprises:

[0085] The active intervention module is configured to analyze whether the user has a trend abnormality based on the health detection data of the preset period, and to actively warn based on a hierarchical warning strategy in the case of a trend abnormality.

[0086] In one embodiment, the active intervention module comprises:

[0087] The anomaly detection unit is configured to record the health detection data of the preset period using a sliding window, and to detect whether the user has a trend abnormality based on the health detection data within the sliding window.

[0088] Specifically, a sliding window is maintained to record the user's daily average health detection data for the past 14 days or the last 100 measurement data. First, the Z-score is calculated based on the data within the sliding window, and if the Z-value is greater than 3, the data is determined to be abnormal. The exponential weighted moving average (EWMA) method can also be used to detect the trend of the data, and if the EWMA value exceeds the set threshold for K consecutive times, a trend abnormality alarm is triggered, thus discovering and warning the user's trend health abnormality in a timely manner.

[0089] Optionally, personalized threshold learning can also be performed. Dynamic threshold adjustment is achieved through an online learning mechanism. Specifically, every time 30 new sample data are added, an online learning process is triggered to extract features such as measurement values, historical mean values, and standard deviations from the 30 new sample data.

[0090] The system updates the threshold offset using a linear regression or a lightweight neural network model. Linear regression or a lightweight neural network is selected as the learning model. The model is trained using the extracted features and corresponding labels (doctor's annotations) to update the model parameters. The model outputs the threshold offset for the current user. The threshold for anomaly detection is adjusted based on the output of the model.

[0091] During the model training process, the user's historical health data and the doctor's professional annotations are combined as supervised information, and a loss function containing false positive and false negative cost weighting is used to optimize the model parameters. This learning method can dynamically adjust the threshold for anomaly detection according to the individual characteristics and actual health status of the user, thereby improving the accuracy and adaptability of anomaly detection, reducing false positives and false negatives, and providing more accurate health monitoring services for users.

[0092] The hierarchical early warning unit is used to determine the early warning level and the corresponding early warning strategy in the presence of abnormal conditions, and actively warn based on the early warning strategy.

[0093] If there is a trend anomaly, the corresponding early warning level is determined based on the relationship between the abnormal value and the preset early warning level. The early warning level includes emergency, high risk, and attention.

[0094] If it is an emergency state, the instantaneous value exceeds the critical threshold, the system immediately prompts the user to seek emergency medical treatment through voice, and at the same time sends an alarm information to the family members or emergency telephone, ensuring that the user can quickly obtain professional medical assistance.

[0095] If it is a high-risk state, when there are K consecutive obvious anomalies or trend warnings, the system suggests the user to contact the doctor and sends a continuous reminder to prompt the user to seek medical treatment or consult professional advice in time.

[0096] If it is an attention state, for single or slight abnormal conditions, the system suggests the user to observe temporarily and retest according to the plan to further confirm the health status, avoiding excessive worry or delay.

[0097] It also supports multi-channel notification, which can be local voice reminder, mobile phone push or sending family group message. This hierarchical early warning mechanism can provide targeted early warning and suggestions according to the severity of the anomaly, ensuring that users can get timely and appropriate health intervention in different situations.

[0098] In one embodiment, it further includes a storage module for storing health detection data, and generating trend analysis reports based on the health detection data in different time dimensions.

[0099] In one embodiment, the system further includes a storage module for storing health detection data. The storage module can classify and organize the data according to different time dimensions, such as day, week, month, and year. Based on these stored data, the system can generate trend analysis reports to help users and doctors understand the trend of health indicators over time. For example, by analyzing the long-term changes in blood pressure, blood sugar, and other data, potential health problems can be identified and prevention recommendations can be provided, thereby achieving more comprehensive and in-depth health management.

[0100] The interactive system of the present application first converts complex professional medical data into user-friendly colloquial explanations, enabling ordinary users to clearly understand their own health status and enhancing their participation and understanding of health management. Second, the system can automatically identify and integrate health device data from different brands and protocols, solving the problem of unified management of cross-device data and providing a comprehensive and convenient health management platform for users. In addition, combined with the user's historical health data, the system can provide personalized health recommendations to help users better manage their health. Finally, by detecting the trend of health data, the system can actively remind users or family members to timely identify potential health problems and achieve early intervention, improving the efficiency and quality of health management.

[0101] According to another aspect of the embodiments of the present application, a home intelligent medical interactive method is also provided. As shown in the method includes the following steps: Figure 3

[0102] S101 identifies the user's voice instruction.

[0103] S102 connects one or more health detection devices to collect one or more health detection data.

[0104] S103 determines whether the health detection data is abnormal, and if the health detection data is abnormal, analyzes the abnormal reason based on the user's personalized life data to obtain a semantic mapping result.

[0105] S104 uses a preset extensible template library to convert the semantic mapping result into a colloquial health explanation text.

[0106] S105 identifies the user's emotional label and identity label based on the user's voice instruction, determines the personalized output method based on the emotional label and identity label, and returns the health explanation text using the personalized output method.

[0107] It also includes active intervention, based on the health detection data of a preset period, analyzes whether the user has a trend abnormality, and in the case of a trend abnormality, actively warns based on a graded warning strategy.

[0108] ​It should be noted that the family smart medical interaction system provided in the above embodiment is only used as an example to illustrate the division of the above functional modules when the family smart medical interaction method is executed, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the family smart medical interaction system and the family smart medical interaction method provided in the above embodiment belong to the same concept, and the implementation process is embodied in the system embodiment, which will not be described here.

[0109] According to another aspect of the embodiments of the present application, a computer readable storage medium corresponding to the family smart medical interaction method provided in the above embodiment is also provided, and the computer readable storage medium stores a computer program (i.e., a program product). When the computer program is run by a processor, the family smart medical interaction method provided in any of the above embodiments is executed.

[0110] It should be noted that examples of the computer readable storage medium can also include, but are not limited to, a phase change memory (PRAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), other types of random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, or other optical, magnetic storage medium, which will not be described one by one here.

[0111] The computer readable storage medium provided in the above embodiments of the present application and the family smart medical interaction method provided in the embodiments of the present application are based on the same inventive concept, and have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0112] The technical features of the above embodiments can be combined in any way. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0113] The above embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A home-based intelligent medical interactive system, characterized in that, include: The data acquisition interface module is used to connect to one or more health monitoring devices and collect one or more health monitoring data. The semantic mapping module is used to determine whether the health detection data is abnormal. If the health detection data is abnormal, the module analyzes the cause of the abnormality based on the user's personalized life data to obtain the semantic mapping result. The natural language generation module is used to convert the semantic mapping results into colloquial health explanation text using a preset extensible template library; The dialogue interaction module is used to recognize user voice commands, identify user emotion tags and identity tags based on the user voice commands, determine a personalized output method based on the emotion tags and identity tags, and return the health explanation text using the personalized output method.

2. The system according to claim 1, characterized in that, It also includes a data processing module, which includes: The protocol adaptive parsing unit is used to identify the protocols used by different devices, load the corresponding regular expression template according to the identified device protocol, parse the corresponding health detection data using the regular expression template, and map the parsing results to a unified medical data standard format. The timing alignment unit is used to perform time calibration and synchronization on the parsed data.

3. The system according to claim 1, characterized in that, The semantic mapping module includes: An anomaly detection unit is used to obtain preset normal value ranges, high value ranges, and low value ranges, and to match the health detection data with the normal value ranges, high value ranges, and low value ranges respectively to obtain a detection result. A personalized correction unit is used to acquire the user's health monitoring data for historical time periods, calculate the statistical data of the health monitoring data, and correct the judgment result based on the statistical data; The contextual semantic unit is used to acquire the user's personalized life data and analyze the causes of anomalies based on the user's personalized life data.

4. The system according to claim 3, characterized in that, Obtaining the user's personalized lifestyle data includes: Obtain the user's geographic location and collect the weather data corresponding to that location; Obtain the user's testing time, diet log, and activity records; The user's personalized lifestyle data is determined based on the detection time, diet log, activity records, and weather data.

5. The system according to claim 1, characterized in that, The dialogue interaction module includes: A voice recognition unit is used to recognize user voice commands; An identity recognition unit is used to extract user voiceprint information based on the voice command, and to determine user identity tags based on the voiceprint information and / or image information; An emotion recognition unit is used to determine the user's emotion label based on the user's voice commands, facial images, and detection data; The output unit is used to determine a personalized output method based on the emotion tag and identity tag, and return the health explanation text using the personalized output method.

6. The system according to claim 1, characterized in that, Also includes: The proactive intervention module is used to analyze whether there are any trend abnormalities in the user's health monitoring data based on a preset time period, and to issue proactive warnings based on a graded warning strategy if such trend abnormalities are found.

7. The system according to claim 6, characterized in that, The active intervention module includes: An anomaly detection unit is used to record health monitoring data for the preset time period using a sliding window, and to detect whether the user has any trend-related anomalies based on the health monitoring data within the sliding window. The graded early warning unit is used to determine the early warning level and the corresponding early warning strategy when an anomaly occurs, and to issue an active early warning based on the early warning strategy.

8. The system according to claim 1, characterized in that, Also includes: The storage module is used to store the health monitoring data and generate trend analysis reports based on the health monitoring data according to different time dimensions.

9. A home-based intelligent medical interaction method, characterized in that, include: Recognize user voice commands; Connect to one or more health monitoring devices to collect one or more health monitoring data; Determine whether the health monitoring data is abnormal. If the health monitoring data is abnormal, analyze the cause of the abnormality based on the user's personalized life data to obtain a semantic mapping result. A pre-defined, extensible template library is used to transform the semantic mapping results into colloquial health explanation text; Based on the user's voice command, the system identifies the user's emotion tag and identity tag, determines a personalized output method based on the emotion tag and identity tag, and returns the health explanation text using the personalized output method.

10. A computer-readable medium, characterized in that, It stores computer-readable instructions, which are executed by a processor to implement a home intelligent medical interaction method as described in claim 9.

Images