Intelligent robot system of intelligent health care center

Abstract

The invention provides an intelligent robot system of an intelligent health care center, and the system comprises an intelligent rehabilitation robot module which assists a user in rehabilitation training through a rehabilitation assistance robot and a generated personalized postoperative rehabilitation scheme; the intelligent recuperating robot module is used for carrying out recuperating interaction with the user through a recuperating robot and the generated personalized recuperating scheme; the intelligent accompanying robot module is used for carrying out accompanying interaction with the user through an accompanying robot and the generated personalized accompanying scheme; the intelligent health management module is used for generating a personalized health management scheme of the user and supervising and evaluating the personalized health management scheme executed by the user through a health management supervision terminal; the 5G remote module is used for receiving health care management data transmitted by the target terminal through 5G in real time; processing the health care management data through the local intelligent terminal to generate a health care data processing result; and a health care data processing result is transmitted to the target terminal in real time through 5G, so that remote health care management is realized.

Description

Technical Field

[0001] This application pertains to the field of health and wellness, and particularly relates to an intelligent robot system for a smart health and wellness center. Background Technology

[0002] As the population ages, the demand for health and wellness services among the elderly is gradually increasing. On the one hand, most health and wellness institutions have low levels of intelligent technology in their facilities, making it difficult to provide high-quality services. On the other hand, with the development of technology, various health and wellness service institutions also need to develop towards intelligence and specialization to meet the health and wellness needs of their clients.

[0003] Currently, my country's supply of smart elderly care services is insufficient, the proportion is low, and the quality is not high. Elderly people, due to limited physical conditions, often have poor self-care abilities and require immediate attention when health problems arise. Therefore, designing an intelligent robot system for smart elderly care centers has become an urgent problem to be solved in this field. Summary of the Invention

[0004] This application provides an intelligent robot system for a smart health and wellness center, which can meet users' personalized needs for rehabilitation, recuperation, and companionship, and can also realize remote health and wellness, providing users with a comprehensive and intelligent health and wellness experience.

[0005] In a first aspect, embodiments of this application provide an intelligent robot system for a smart elderly care center, comprising: The intelligent rehabilitation robot module is used to generate personalized postoperative rehabilitation plans for users, and to assist users in rehabilitation training through rehabilitation assistive robots and personalized postoperative rehabilitation plans; as well as to conduct rehabilitation monitoring and assessment during the user's rehabilitation training process. The intelligent rehabilitation robot module is used to generate personalized rehabilitation plans for users, and to interact with users through the rehabilitation robot and personalized rehabilitation plans; as well as to monitor and evaluate users during the rehabilitation process; The intelligent companion robot module is used to generate personalized companion plans for users, and to interact with users through companion robots and personalized companion plans; as well as to monitor and evaluate users during the companionship process; The intelligent health management module is used to generate personalized health management plans for users and to supervise and evaluate the implementation of these plans by users through a health management monitoring terminal. The 5G remote module is used to receive health and wellness management data transmitted in real time by the target terminal via 5G; process the health and wellness management data through a local smart terminal to generate health and wellness data processing results; and transmit the health and wellness data processing results to the target terminal in real time via 5G to instruct the target terminal to use the health and wellness data processing results to interact with the user, thereby realizing remote health and wellness management; wherein, the health and wellness management data includes the user's basic physiological data and / or basic psychological data collected in real time.

[0006] Optionally, the intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized rehabilitation plan generation model to obtain a personalized postoperative rehabilitation plan associated with the user, output by the personalized rehabilitation plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized rehabilitation plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a first feature fusion module, and multiple first output modules; the multiple first output modules include a rehabilitation training plan output module, a nutrition plan output module, and a mental health plan output module.

[0007] Optionally, the intelligent rehabilitation robot module is specifically used for: Control the rehabilitation assistive robot to assist the user in rehabilitation training according to the rehabilitation training plan in the personalized postoperative rehabilitation program; and, Control the mental health interactive terminal and assist the user in mental health recovery according to the mental health plan in the personalized postoperative rehabilitation plan.

[0008] Optionally, the intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a second feature fusion module, and multiple second output modules; each second output module is used to output the personalized parameters corresponding to the user when using each intelligent care robot.

[0009] Optionally, the intelligent rehabilitation robot module is specifically used for: Control each intelligent therapy robot to interact with the user according to the user's personalized parameters; Real-time acquisition of physiological data collected during user interaction during the therapy process; Based on the physiological data, a user's recuperation and monitoring results are generated.

[0010] Optionally, the intelligent companion robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a third feature fusion module, and multiple third output modules; the multiple third output modules include a voice interaction output module, an entertainment interaction plan output module, and a nursing plan output module.

[0011] Optionally, the intelligent companion robot module is specifically used for: Control the intelligent voice interaction terminal to conduct voice interaction with the user according to the voice interaction scheme in the personalized care plan; and Control the intelligent entertainment interaction terminal to engage in entertainment interaction with the user according to the entertainment interaction plan in the personalized care plan; and... Control the intelligent nursing terminal to provide care to the user according to the nursing plan in the personalized care plan.

[0012] Optionally, the intelligent health management module is specifically used for: The user's health management data is input into a pre-trained personalized health management plan generation model to obtain a personalized health management plan output by the personalized health management plan generation model; wherein, the health management data includes structured physical examination data; The personalized health management solution generation model includes an encoder, a personalized condition injection module, and a decoder. The encoder captures the correlation between health management data through a multi-head self-attention mechanism. The personalized condition injection module generates condition vectors based on the health management data and uses a gating mechanism to fuse the condition vectors with the encoder output to generate personalized weights. The decoder ensures that the output personalized health management solution is related to the input data through an encoder-decoder attention mechanism and ensures the coherence of the output personalized health management solution through a self-attention mechanism.

[0013] Optionally, the 5G remote module is specifically used for: Based on the local user data corresponding to the target terminal and the health and wellness management data transmitted in real time by the target terminal, generate the health and wellness data processing results corresponding to the user. Local user data includes user attribute information and medical records, including historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data.

[0014] Optionally, the 5G remote module is specifically used for: If the user's corresponding health and wellness data processing result is suspected to be abnormal, the health and wellness data processing result and data supplementation instructions are transmitted to the target terminal in real time via 5G, so as to instruct the target terminal to use the health and wellness data processing result to interact with the user and obtain supplementary monitoring data.

[0015] Optionally, the 5G remote module is specifically used for: User attribute information, local user data, health management data, and supplementary monitoring data are input into a pre-trained intelligent pre-diagnosis model to obtain the pre-diagnosis results output by the intelligent pre-diagnosis model. The intelligent pre-diagnosis model includes an attribute feature extraction module, a historical feature extraction module, a real-time feature extraction module, a fourth feature fusion module, and multiple fourth output modules. Each of the fourth output modules is used to output the probability information of the user currently having each preset category of disease.

[0016] Secondly, embodiments of this application provide a control method for an intelligent robot in a smart health and wellness center, the control method being used to implement the functions of the intelligent robot system in the smart health and wellness center as described in any embodiment of the first aspect.

[0017] Thirdly, embodiments of this application provide an electronic device, which includes: a processor and a memory storing computer program instructions; The processor executes the computer program instructions to implement a control method for the intelligent robot in the smart health and wellness center.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement a control method for an intelligent robot in a smart elderly care center.

[0019] The intelligent robot system, control method, equipment, and computer-readable storage medium of the smart health and wellness center in this application embodiment provide an intelligent robot system for the smart health and wellness center that includes intelligent rehabilitation function, intelligent convalescence function, intelligent companionship function, and 5G remote function, which can meet the personalized health and wellness needs of users and improve the operational efficiency of the health and wellness center. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the architecture of an intelligent robot system for a smart elderly care center provided in one embodiment of this application; Figure 2 This is an overall architecture diagram, including details of each module, of the intelligent robot system used in a smart elderly care center provided in one embodiment of this application; Figure 3 This is an example of the interaction process of the 5G remote module used in the intelligent robot system of the smart health care center provided in one embodiment of this application; Figure 4 This application provides a schematic diagram illustrating the process of intelligent rehabilitation in an intelligent robot system of a smart elderly care center according to an embodiment of the present application. Figure 5 This application provides a schematic diagram illustrating the process of intelligent care in a smart health and wellness center using an intelligent robot system according to an embodiment of the present application. Figure 6 This application provides a schematic diagram illustrating the process of intelligent companionship in an intelligent robot system of a smart elderly care center according to an embodiment of the present application. Figure 7 This application shows a schematic diagram illustrating the process of 5G remote interaction in the intelligent robot system of a smart elderly care center according to an embodiment of the present application; Figure 8 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0022] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0023] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0024] To address the problems of existing technologies, this application provides an intelligent robot system for a smart elderly care center. The intelligent robot system for a smart elderly care center provided in this application will be described below. Figure 1 This is a schematic diagram of the architecture of an intelligent robot system for a smart elderly care center according to an embodiment of this application. The intelligent robot system for the smart elderly care center includes an intelligent rehabilitation robot module, an intelligent convalescent robot module, an intelligent companion robot module, an intelligent health management module, and a 5G remote module; wherein: The intelligent rehabilitation robot module is used to generate personalized postoperative rehabilitation plans for users, and to assist users in rehabilitation training through rehabilitation assistive robots and personalized postoperative rehabilitation plans; as well as to conduct rehabilitation monitoring and assessment during the user's rehabilitation training process. The intelligent rehabilitation robot module is used to generate personalized rehabilitation plans for users, and to interact with users through the rehabilitation robot and personalized rehabilitation plans; as well as to monitor and evaluate users during the rehabilitation process; The intelligent companion robot module is used to generate personalized companion plans for users, and to interact with users through companion robots and personalized companion plans; as well as to monitor and evaluate users during the companionship process; The intelligent health management module is used to generate personalized health management plans for users and to supervise and evaluate the implementation of these plans by users through a health management monitoring terminal. The 5G remote module is used to receive health and wellness management data transmitted in real time by the target terminal via 5G; process the health and wellness management data through a local smart terminal to generate health and wellness data processing results; and transmit the health and wellness data processing results to the target terminal in real time via 5G to instruct the target terminal to use the health and wellness data processing results to interact with the user, thereby realizing remote health and wellness management; wherein, the health and wellness management data includes the user's basic physiological data and / or basic psychological data collected in real time.

[0025] It should be noted that the intelligent robot (terminal) used in the intelligent robot system disclosed in this application can be an embodied intelligent robot (terminal), and the model used can be updated and iterated based on real-time feedback and collected data to optimize the user experience.

[0026] Figure 2 This is an overall architecture diagram, including details of each module, of the intelligent robot system used in a smart elderly care center according to one embodiment of this application. Figure 2 The smart health and wellness center includes five major functional modules: rehabilitation, convalescence, companionship, 5G remote care, and health management.

[0027] In some embodiments, the intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized rehabilitation plan generation model to obtain a personalized postoperative rehabilitation plan associated with the user, output by the personalized rehabilitation plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized rehabilitation plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a first feature fusion module, and multiple first output modules; the multiple first output modules include a rehabilitation training plan output module, a nutrition plan output module, and a mental health plan output module.

[0028] The user attribute information may include age, gender, weight, height, BMI, occupation, history of underlying diseases, allergies, etc.; the historical diagnostic records may include diagnostic description information, such as "left knee arthritis"; the medical imaging data may include knee X-rays, MRI, CT scans, etc.; the historical medical test data may include complete blood count data, blood biochemistry data, urine test results, etc.; the historical surgical records may include surgical type, surgical time, postoperative rehabilitation records, postoperative complication records, etc.; the historical medication records may include the name, dosage, frequency, and time of medication taken, etc.

[0029] In some embodiments, the attribute feature extraction module included in the personalized rehabilitation plan generation model can be composed of two MLP layers, with the hidden layer sizes set to 64 and 32 respectively, the L2 weight decay of the output stage being 0.001 to improve generalization, the activation function being set to the ReLU function, the training batch being set to 32 to balance memory and efficiency, and the output being a 32-dimensional feature vector.

[0030] In some embodiments, the personalized rehabilitation plan generation model includes a patient visit feature extraction module, which may consist of a text encoder, an image encoder, a laboratory data MLP, a medication record MLP, and a fusion submodule. The input to the text encoder can be a text-based diagnostic / surgical record, and the text encoder can be the encoder in a pre-trained medical BERT model, with an output dimension of 768. The input to the image encoder can be a medical image, and the image encoder type can be a pre-trained ResNet network with fully connected layers removed (e.g., ResNet-50, ResNet-101, etc.). The encoder's output dimension is set to 2048; the test data MLP can consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the medication record MLP can also consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the fusion submodule is used to concatenate the outputs of the text encoder, image encoder, test data MLP, and medication record MLP, and perform dimensionality reduction through the MLP fusion layer, outputting a 128-dimensional feature vector.

[0031] In some embodiments, the feature fusion module included in the personalized rehabilitation plan generation model takes as input a 32-dimensional feature vector output by the attribute feature extraction module and a 128-dimensional feature vector output by the medical visit feature extraction module, and then passes them through a concatenation process, a fully connected layer, and a Swish activation layer to obtain the output 128-dimensional fused feature vector.

[0032] In some embodiments, the rehabilitation training plan output module included in the personalized rehabilitation plan generation model takes the fused feature vector as input and outputs a personalized rehabilitation plan containing exercise type, intensity, and frequency (e.g., "3 sets of knee flexion and extension × 10 times / day, moderate intensity"). The model architecture of the rehabilitation training plan output module includes an LSTM sequence generator, and the decoder therein can be set to a 64-unit LSTM.

[0033] In some embodiments, the personalized rehabilitation plan generation model includes a nutrition plan output module, which takes a fused feature vector as input and outputs a nutrition plan containing dietary recommendations. The model architecture of the nutrition plan output module includes an MLP classifier, and the number of output categories can be set to 20, and the number of hidden layers can be set to 64.

[0034] In some embodiments, the personalized rehabilitation plan generation model includes a mental health plan output module, which takes a fused feature vector as input and outputs a mental health plan containing psychological intervention suggestions. The model architecture of the mental health plan output module includes a Transformer decoder, and the number of layers of the Transformer decoder can be set to 2.

[0035] In some embodiments, when training the personalized rehabilitation plan generation model, the loss functions corresponding to the three output modules can be weighted and calculated to obtain the weighted total loss, and training can be performed based on the weighted total loss. According to clinical importance, the weights corresponding to the rehabilitation training plan output module, nutrition plan output module, and mental health plan output module can be set to 0.4, 0.3, and 0.3, respectively.

[0036] In some embodiments, the intelligent rehabilitation robot module is specifically used for: Control the rehabilitation assistive robot to assist the user in rehabilitation training according to the rehabilitation training plan in the personalized postoperative rehabilitation program; and, Control the mental health interactive terminal and assist the user in mental health recovery according to the mental health plan in the personalized postoperative rehabilitation plan.

[0037] In some embodiments, the intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a second feature fusion module, and multiple second output modules; each second output module is used to output the personalized parameters corresponding to the user when using each intelligent care robot.

[0038] In some embodiments, the attribute feature extraction module included in the personalized care plan generation model can be composed of two MLP layers, with the hidden layer sizes set to 64 and 32 respectively, the L2 weight decay of the output stage being 0.001 to improve generalization, the activation function being set to the ReLU function, the training batch being set to 32 to balance memory and efficiency, and the output being a 32-dimensional feature vector.

[0039] In some embodiments, the personalized healthcare plan generation model includes a patient visit feature extraction module, which may consist of a text encoder, an image encoder, a laboratory data MLP, a medication record MLP, and a fusion submodule. The input to the text encoder can be a text-based diagnostic / surgical record, and the text encoder can be the encoder in a pre-trained medical BERT model, with an output dimension of 768. The input to the image encoder can be a medical image, and the image encoder type can be a pre-trained ResNet network with fully connected layers removed (e.g., ResNet-50, ResNet-101, etc.). The encoder's output dimension is set to 2048; the test data MLP can consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the medication record MLP can also consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the fusion submodule is used to concatenate the outputs of the text encoder, image encoder, test data MLP, and medication record MLP, and perform dimensionality reduction through the MLP fusion layer, outputting a 128-dimensional feature vector.

[0040] In some embodiments, the feature fusion module included in the personalized care plan generation model takes as input a 32-dimensional feature vector output by the attribute feature extraction module and a 128-dimensional feature vector output by the medical visit feature extraction module, and then passes them sequentially through a concatenation process, a fully connected layer, and a Swish activation layer to obtain the output 128-dimensional fused feature vector.

[0041] In some embodiments, the intelligent rehabilitation robot module is specifically used for: Control each intelligent therapy robot to interact with the user according to the user's personalized parameters; Real-time acquisition of physiological data collected during user interaction during the therapy process; Based on the physiological data, a user's recuperation and monitoring results are generated.

[0042] In some embodiments, the intelligent rehabilitation robot module is specifically used for: Based on the user's TCM four diagnostic methods data, a pre-constructed TCM knowledge graph, and a pre-trained TCM diagnostic and treatment model, a personalized TCM treatment plan associated with the user is generated; wherein, the TCM four diagnostic methods data includes tongue appearance data, pulse appearance data, consultation data, and facial color data; The construction of the TCM knowledge graph includes the following steps: Traditional Chinese medicine (TCM) knowledge is extracted from TCM literature, modern medical records, and the experience of TCM experts, and classified according to symptoms, syndromes, prescriptions, and physiotherapy methods to construct a TCM text dataset containing different categories. The TCM text dataset is cleaned of abnormal data and missing data is filled in. Then, text segmentation, stop word removal and entity recognition are performed in sequence to obtain an entity set containing symptoms, syndromes, prescriptions and physiotherapy methods. Based on the entities contained in the entity set, construct a knowledge structure including disease-syndrome-prescription and disease-syndrome-physiotherapy method, and construct a knowledge graph based on the knowledge structure.

[0043] In some embodiments, the TCM diagnosis and treatment model may include a multimodal data processing module, a knowledge graph module, and an intelligent diagnosis and decision-making module. The multimodal data processing module is used to extract and fuse features from the four diagnostic methods (inspection, auscultation and olfaction, palpation, and olfaction) data input to the model for subsequent constitution identification and other processing. The knowledge graph module is used to augment the input data to enrich the semantic information in the subsequent intelligent diagnosis and decision-making process, and to assist in the model's inference and output stages, such as verifying the model's initial output results based on the content of the knowledge graph to improve the accuracy of the output results. The intelligent diagnosis and decision-making module is used to perform intelligent TCM diagnosis based on the knowledge graph's inference engine and multimodal data, and further match suitable treatment robot usage schemes based on the diagnostic results.

[0044] In some embodiments, the TCM diagnosis and treatment model can be obtained by sequentially training, supervising fine-tuning, applying a reward model, and performing reinforcement learning on the basis of a generative model; wherein... In the pre-training stage, open-source models such as Llama-4 can be used as the base model, and TCM knowledge graphs can be used for pre-training to enable the pre-trained model to understand TCM terminology and related concepts.

[0045] During the supervised fine-tuning phase, structured training data based on medical case data can be used to fine-tune the pre-trained model obtained after pre-training. The training data can be, for example, the input: "Patient has stomach pain, dry mouth and red tongue, thready and rapid pulse, and slightly yellow complexion", and the output: "Diagnosis: Spleen and stomach yin deficiency type; Treatment: Nourishing yin and harmonizing the stomach; Prescription: Modified Sha Shen Mai Dong Tang".

[0046] In the reward model phase, a reward model based on TCM expert standards can be constructed to evaluate the accuracy and rationality of the diagnostic and treatment / physiotherapy recommendations output by the model.

[0047] During the reinforcement learning phase, the model can be optimized through reinforcement learning to provide diagnostic and treatment suggestions that are more in line with traditional Chinese medicine theory. On the other hand, the accuracy of diagnosis can be continuously optimized by combining feedback from smart health and wellness centers on actual user experiences.

[0048] For example, the personalized TCM health care plan generated by the TCM diagnosis and treatment model may include the following diagnostic results: "Syndrome type: Spleen and stomach yin deficiency type; Constitution: Yin deficiency constitution (red tongue with little coating, thready and rapid pulse); Health risks: prone to problems such as dry mouth, insomnia, and indigestion; Etiological analysis: long-term irregular diet, depletion of stomach yin," and physiotherapy suggestions: "Physiotherapy equipment: intelligent moxibustion robot; Usage plan: Location: Zhongwan acupoint, Zusanli acupoint; Frequency: once a day, 20 minutes each time; Duration: it is recommended to use continuously for 15 days; Precautions: avoid use on an empty stomach; pay attention to temperature during use to avoid burns; those with sensitive skin should reduce the use time; Auxiliary suggestions: Diet: eat more foods that nourish yin and moisten dryness, such as pears, white fungus, and lilies; Massage: massage Zusanli acupoint for 10 minutes every day; Sleep: maintain a regular schedule and avoid staying up late."

[0049] Thus, the trained TCM diagnostic and therapeutic model has the following advantages: 1. Multimodal data fusion: Combining multi-dimensional data such as tongue appearance, facial appearance, pulse appearance, and medical history to improve diagnostic accuracy.

[0050] 2. Knowledge Graph Driven: Achieve precise association between symptoms, syndrome types, and prescriptions through traditional Chinese medicine knowledge graphs.

[0051] 3. Intelligent matching of rehabilitation robots: Intelligently matching TCM diagnosis results with rehabilitation robot usage plans.

[0052] 4. Continuous learning mechanism: Based on user feedback and expert evaluation, continuously optimize diagnosis and suggestions.

[0053] 5. Adapted to health and wellness scenarios: Designed specifically for smart health and wellness centers, it boasts a high degree of intelligence and meets users' personalized needs.

[0054] In some embodiments, the intelligent companion robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a third feature fusion module, and multiple third output modules; the multiple third output modules include a voice interaction output module, an entertainment interaction plan output module, and a nursing plan output module.

[0055] In some embodiments, the attribute feature extraction module included in the personalized care solution generation model can be composed of two MLP layers, with the hidden layer sizes set to 64 and 32 respectively, the L2 weight decay of the output stage being 0.001 to improve generalization, the activation function being set to the ReLU function, the training batch being set to 32 to balance memory and efficiency, and the output being a 32-dimensional feature vector.

[0056] In some embodiments, the personalized care companion generation model includes a medical visit feature extraction module, which may consist of a text encoder, an image encoder, a laboratory data MLP, a medication record MLP, and a fusion submodule. The input to the text encoder can be a text-based diagnostic / surgical record, and the text encoder can be the encoder in a pre-trained medical BERT model, with an output dimension of 768. The input to the image encoder can be a medical image, and the image encoder type can be a pre-trained ResNet network with fully connected layers removed (e.g., ResNet-50, ResNet-101, etc.). The encoder's output dimension is set to 2048; the test data MLP can consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the medication record MLP can also consist of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the fusion submodule is used to concatenate the outputs of the text encoder, image encoder, test data MLP, and medication record MLP, and perform dimensionality reduction through the MLP fusion layer, outputting a 128-dimensional feature vector.

[0057] In some embodiments, the feature fusion module included in the personalized care plan generation model takes as input a 32-dimensional feature vector output by the attribute feature extraction module and a 128-dimensional feature vector output by the medical visit feature extraction module, and then passes them through a concatenation process, a fully connected layer, and a Swish activation layer to obtain the output 128-dimensional fused feature vector.

[0058] In some embodiments, the personalized care solution generation model includes a voice interaction output module, which takes a fused feature vector as input and outputs a chat care solution that includes text topic classification. The model architecture of the voice interaction output module includes an MLP classifier, and the output dimension of the voice interaction output module can be set to 20.

[0059] In some embodiments, the personalized companionship solution generation model includes an entertainment interaction solution output module, which takes a fused feature vector as input and outputs an entertainment interaction solution that includes activity recommendation classification. The model architecture of the entertainment interaction solution output module includes an MLP classifier, and the output dimension of the entertainment interaction solution output module can be set to 20.

[0060] In some embodiments, the personalized care plan generation model includes a care plan output module, which takes a fused feature vector as input and outputs a care plan that includes a care suggestion classification. The model architecture of the care plan output module includes an MLP classifier, and the output dimension of the care plan output module can be set to 20.

[0061] In some embodiments, the intelligent companion robot module is specifically used for: Control the intelligent voice interaction terminal to conduct voice interaction with the user according to the voice interaction scheme in the personalized care plan; and Control the intelligent entertainment interaction terminal to engage in entertainment interaction with the user according to the entertainment interaction plan in the personalized care plan; and... Control the intelligent nursing terminal to provide care to the user according to the nursing plan in the personalized care plan.

[0062] In some embodiments, the intelligent health management module is specifically used for: The user's health management data is input into a pre-trained personalized health management plan generation model to obtain a personalized health management plan output by the personalized health management plan generation model; wherein, the health management data includes structured physical examination data; The personalized health management solution generation model includes an encoder, a personalized condition injection module, and a decoder. The encoder captures the correlation between health management data through a multi-head self-attention mechanism. The personalized condition injection module generates condition vectors based on the health management data and uses a gating mechanism to fuse the condition vectors with the encoder output to generate personalized weights. The decoder ensures that the output personalized health management solution is related to the input data through an encoder-decoder attention mechanism and ensures the coherence of the output personalized health management solution through a self-attention mechanism.

[0063] In some embodiments, health management data may include user attribute information, structured physical examination data, past medical history, family medical history, lifestyle data, psychological assessment data, etc.

[0064] In some embodiments, the encoder included in the personalized health management plan generation model can be composed of a 12-layer Transformer Encoder structure, with the number of self-attention heads set to 12, the hidden layer size set to 768, and the feedforward neural network size set to 3072. It can capture the correlation between health management data through a multi-head self-attention mechanism, stabilize the training process through layer normalization, and alleviate the gradient vanishing problem through a residual connection structure, thereby improving the accuracy and robustness of the personalized health management plan generation model.

[0065] In some embodiments, the personalized condition injection module included in the personalized health management solution generation model can generate a condition vector based on health management data and use a gating mechanism to fuse the condition vector with the encoder output; the condition vector can be multiplied with the encoder output through a linear layer to generate personalized weights.

[0066] In some embodiments, the decoder included in the personalized health management plan generation model can be composed of a 12-layer Transformer Decoder structure, with the number of self-attention heads set to 12 and the hidden layer size set to 768. It can ensure that the output personalized health management plan is related to the input data through the encoder-decoder attention mechanism, ensure the coherence of the output personalized health management plan through the self-attention mechanism, and retain the order information of each suggestion in the personalized health management plan through positional encoding.

[0067] In some embodiments, during the training of the personalized health management plan generation model, a corpus can be built based on large-scale texts in the health field such as medical literature, health guidelines, and health blogs, and the model building efficiency can be improved by using a pre-trained model and fine-tuning.

[0068] Figure 3 This is an example of the interaction process of the 5G remote module used in the intelligent robot system of the smart health care center provided in one embodiment of this application.

[0069] In some embodiments, the 5G remote module is specifically used for: Based on the local user data corresponding to the target terminal and the health and wellness management data transmitted in real time by the target terminal, generate the health and wellness data processing results corresponding to the user. Local user data includes user attribute information and medical records, including historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data.

[0070] In some embodiments, the 5G remote module is specifically used for: If the user's corresponding health and wellness data processing result is suspected to be abnormal, the health and wellness data processing result and data supplementation instructions are transmitted to the target terminal in real time via 5G, so as to instruct the target terminal to use the health and wellness data processing result to interact with the user and obtain supplementary monitoring data.

[0071] In some embodiments, the 5G remote module is specifically used for: User attribute information, local user data, health management data, and supplementary monitoring data are input into a pre-trained intelligent pre-diagnosis model to obtain the pre-diagnosis results output by the intelligent pre-diagnosis model. The intelligent pre-diagnosis model includes an attribute feature extraction module, a historical feature extraction module, a real-time feature extraction module, a fourth feature fusion module, and multiple fourth output modules. Each of the fourth output modules is used to output the probability information of the user currently having each preset category of disease.

[0072] The user attribute information may include age, gender, weight, height, BMI, occupation, history of underlying diseases, and allergy history. The local user data may include historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data. The historical diagnostic records may include diagnostic descriptions, such as "left knee arthritis." The medical imaging data may include, for example, X-rays, MRIs, and CT scans of the knee joint. The historical medical test data may include complete blood count data, blood biochemistry data, and urinalysis results. The historical surgical records may include surgical type, surgical time, postoperative rehabilitation records, and postoperative complication records. The historical medication records may include the name, dosage, frequency, and time of medication taken. The psychological data in the health management data may include anxiety scores, stress index, and mood fluctuation index. The physiological data may include heart rate, blood oxygen, blood pressure, respiratory rate, and blood glucose.

[0073] In some embodiments, the attribute feature extraction module included in the intelligent pre-diagnosis model can be composed of two MLP layers, with the hidden layer sizes set to 64 and 32 respectively, the L2 weight decay of the output stage being 0.001 to improve generalization, the activation function being set to the ReLU function, the training batch being set to 32 to balance memory and efficiency, and the output being a 32-dimensional feature vector.

[0074] In some embodiments, the historical feature extraction module included in the intelligent pre-diagnosis model may consist of a text encoder, an image encoder, a medical examination data MLP, a medication record MLP, and a fusion submodule. The input to the text encoder can be a text-based diagnostic / surgical record, and the text encoder can be the encoder in a pre-trained medical BERT model, with an output dimension of 768. The input to the image encoder can be a medical image, and the image encoder type can be a pre-trained ResNet network with fully connected layers removed (e.g., ResNet-50, ResNet-101, etc.). The output dimension of the encoder is set to 2048; the test data MLP can be composed of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the medication record MLP can be composed of two MLP layers with hidden layer sizes of 64 and 32 respectively, and the activation function is set to LeakyReLU, outputting a 32-dimensional feature vector; the fusion submodule is used to concatenate the outputs of the text encoder, image encoder, test data MLP, and medication record MLP, and perform dimensionality reduction through the MLP fusion layer, outputting a 128-dimensional feature vector.

[0075] In some embodiments, the real-time feature extraction module included in the intelligent pre-diagnosis model can be composed of a physiological branch and a psychological branch; wherein, the model architecture of the physiological branch can be a CNN model capable of processing sequential data, which can be composed of sequentially concatenated convolutional layers, pooling layers, convolutional layers, pooling layers, fully connected layers, and activation layers, and the output dimension can be set to 64; the model architecture of the psychological branch can be an MLP, and the output dimension can be set to 64; the outputs of the physiological branch and the psychological branch can be compressed after concatenation to obtain the 64-dimensional features output by the real-time feature extraction module.

[0076] In some embodiments, the output of the intelligent prediagnosis model may include the probability of developing cardiovascular and cerebrovascular diseases, which is used to characterize the probability that the user is currently in the onset period of cardiovascular and cerebrovascular diseases.

[0077] In some embodiments, the output of the intelligent prediagnosis model may include the probability of developing respiratory diseases, which is used to characterize the probability that a user is currently in the acute phase of a respiratory disease.

[0078] In some embodiments, the output of the intelligent prediagnosis model may include the probability of having a metabolic disease, which is used to characterize the probability that the user is currently in the onset period of a metabolic disease.

[0079] In some embodiments, the model architecture of the output module of the intelligent pre-diagnosis model can be an MLP, which includes a fully connected layer and the probability range of the output value is [0,1].

[0080] Figure 4 This illustration shows a flowchart of intelligent rehabilitation performed in an intelligent robot system of a smart elderly care center according to an embodiment of this application. Figure 4 As shown, it includes: S401. Generate a personalized postoperative rehabilitation plan for the user.

[0081] S402, assists users in rehabilitation training through rehabilitation assistive robots and personalized postoperative rehabilitation plans.

[0082] S403. Conduct rehabilitation monitoring and assessment during the user's rehabilitation training process.

[0083] Figure 5 This illustration shows a flowchart of a smart healthcare center's intelligent robot system performing intelligent care according to an embodiment of this application. Figure 5 As shown, it includes: S501, Generate a personalized treatment plan for the user.

[0084] S502, interacts with users through therapeutic robots and personalized therapeutic plans.

[0085] S503. Monitor and assess users during their recuperation process.

[0086] Figure 6 This illustration shows a flowchart of the intelligent robot system providing intelligent companionship in a smart elderly care center according to an embodiment of this application. Figure 6 As shown, it includes: S601, Generate a personalized care plan for the user.

[0087] S602, interacts with users through companion robots and personalized companion solutions.

[0088] S603. Monitor and assess during the user's care process.

[0089] Figure 7 This illustration shows a flowchart of a 5G remote interaction process within an intelligent robot system of a smart elderly care center, according to an embodiment of this application. Figure 7 As shown, it includes: S701. Receive health and wellness management data transmitted in real time by the target terminal via 5G; wherein, the health and wellness management data includes the user's basic physiological data and / or basic psychological data collected in real time.

[0090] S702. Process health and wellness management data through a local smart terminal to generate health and wellness data processing results.

[0091] S703. Transmit the health and wellness data processing results to the target terminal in real time via 5G, so as to instruct the target terminal to use the health and wellness data processing results to interact with the user and realize remote health and wellness management.

[0092] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown.

[0093] Electronic devices may include a processor 801 and a memory 802 storing computer program instructions.

[0094] Specifically, the processor 801 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0095] Memory 802 may include mass storage for data or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 802 may include removable or non-removable (or fixed) media. Where appropriate, memory 802 may be internal or external to an electronic device. In a particular embodiment, memory 802 may be a non-volatile solid-state memory.

[0096] In one embodiment, memory 802 may be read-only memory (ROM). In one embodiment, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0097] The processor 801 reads and executes computer program instructions stored in the memory 802 to implement the functions of the intelligent robot system of the smart health care center described in any of the above embodiments.

[0098] In one example, the electronic device may also include a communication interface 803 and a bus 810. For example, Figure 8 As shown, the processor 801, memory 802, and communication interface 803 are connected through bus 810 and complete communication with each other.

[0099] The communication interface 803 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0100] Bus 810 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 810 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0101] Alternatively, embodiments of this application may be implemented using a computer-readable storage medium. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement the functions of the intelligent robot system in the smart elderly care center described in any of the above embodiments.

[0102] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0103] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0104] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0105] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0106] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. An intelligent robot system for a smart health and wellness center, characterized in that, include: The intelligent rehabilitation robot module is used to generate personalized postoperative rehabilitation plans for users and assist users in rehabilitation training through rehabilitation assistive robots and personalized postoperative rehabilitation plans; In addition, rehabilitation monitoring and assessment are conducted during the user's rehabilitation training process; The intelligent rehabilitation robot module is used to generate personalized rehabilitation plans for users, and to interact with users through the rehabilitation robot and personalized rehabilitation plans; as well as to monitor and evaluate users during the rehabilitation process; The intelligent companion robot module is used to generate personalized companion plans for users, and to interact with users through companion robots and personalized companion plans; as well as to monitor and evaluate users during the companionship process; The intelligent health management module is used to generate personalized health management plans for users and to supervise and evaluate the implementation of these plans by users through a health management monitoring terminal. The 5G remote module is used to receive health and wellness management data transmitted in real time by the target terminal via 5G. The data on health and wellness management is processed through local smart terminals to generate health and wellness data processing results; The health and wellness data processing results are transmitted to the target terminal in real time via 5G to instruct the target terminal to use the health and wellness data processing results to interact with the user and realize remote health and wellness management; wherein, the health and wellness management data includes the user's basic physiological data and / or basic psychological data collected in real time.

2. The intelligent robot system for a smart elderly care center according to claim 1, characterized in that, The intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized rehabilitation plan generation model to obtain a personalized postoperative rehabilitation plan associated with the user, output by the personalized rehabilitation plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized rehabilitation plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a first feature fusion module, and multiple first output modules; the multiple first output modules include a rehabilitation training plan output module, a nutrition plan output module, and a mental health plan output module.

3. The intelligent robot system for a smart elderly care center according to claim 1 or 2, characterized in that, The intelligent rehabilitation robot module is specifically used for: Control the rehabilitation assistive robot to assist the user in rehabilitation training according to the rehabilitation training plan in the personalized postoperative rehabilitation program; and, Control the mental health interactive terminal and assist the user in mental health recovery according to the mental health plan in the personalized postoperative rehabilitation plan.

4. The intelligent robot system for a smart elderly care center according to claim 1, characterized in that, The intelligent rehabilitation robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, historical surgical data, and historical medication data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a second feature fusion module, and multiple second output modules; each second output module is used to output the personalized parameters corresponding to the user when using each intelligent care robot.

5. The intelligent robot system for a smart elderly care center according to claim 4, characterized in that, The intelligent rehabilitation robot module is specifically used for: Control each intelligent therapy robot to interact with the user according to the user's personalized parameters; Real-time acquisition of physiological data collected during user interaction during the therapy process; Based on the physiological data, a user's recuperation and monitoring results are generated.

6. The intelligent robot system for a smart elderly care center according to claim 1, characterized in that, The intelligent companion robot module is specifically used for: The user's attribute information and medical data are input into a pre-trained personalized care plan generation model to obtain a personalized care plan associated with the user, output by the personalized care plan generation model; wherein, the medical data includes historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data; The personalized care plan generation model includes an attribute feature extraction module, a medical visit feature extraction module, a third feature fusion module, and multiple third output modules; the multiple third output modules include a voice interaction output module, an entertainment interaction plan output module, and a nursing plan output module.

7. The intelligent robot system for a smart elderly care center according to claim 6, characterized in that, The intelligent companion robot module is specifically used for: Control the intelligent voice interaction terminal to conduct voice interaction with the user according to the voice interaction scheme in the personalized care plan; and Control the intelligent entertainment interaction terminal to conduct entertainment interaction with the user according to the entertainment interaction plan in the personalized care plan; as well as, Control the intelligent nursing terminal to provide care to the user according to the nursing plan in the personalized care plan.

8. The intelligent robot system for a smart elderly care center according to claim 1, characterized in that, The intelligent health management module is specifically used for: The user's health management data is input into a pre-trained personalized health management plan generation model to obtain a personalized health management plan output by the personalized health management plan generation model; wherein, the health management data includes structured physical examination data; The personalized health management solution generation model includes an encoder, a personalized condition injection module, and a decoder. The encoder captures the correlation between health management data through a multi-head self-attention mechanism. The personalized condition injection module generates condition vectors based on the health management data and uses a gating mechanism to fuse the condition vectors with the encoder output to generate personalized weights. The decoder ensures that the output personalized health management solution is related to the input data through an encoder-decoder attention mechanism and ensures the coherence of the output personalized health management solution through a self-attention mechanism.

9. The intelligent robot system for a smart elderly care center according to claim 1, characterized in that, The 5G remote module is specifically used for: Based on the local user data corresponding to the target terminal and the health and wellness management data transmitted in real time by the target terminal, generate the health and wellness data processing results corresponding to the user. Local user data includes user attribute information and medical records, including historical diagnostic records, medical imaging data, historical medical test data, and historical surgical data.

10. The intelligent robot system for a smart elderly care center according to claim 9, characterized in that, The 5G remote module is specifically used for: If the user's corresponding health and wellness data processing result is suspected to be abnormal, the health and wellness data processing result and data supplementation instructions are transmitted to the target terminal in real time via 5G, so as to instruct the target terminal to use the health and wellness data processing result to interact with the user and obtain supplementary monitoring data. User attribute information, local user data, health management data, and supplementary monitoring data are input into a pre-trained intelligent pre-diagnosis model to obtain the pre-diagnosis results output by the intelligent pre-diagnosis model. The intelligent pre-diagnosis model includes an attribute feature extraction module, a historical feature extraction module, a real-time feature extraction module, a fourth feature fusion module, and multiple fourth output modules. Each of the fourth output modules is used to output the probability information of the user currently having each preset category of disease.

Images