A remote diagnosis and treatment system and diagnosis and treatment method based on the Internet of Things

Abstract

The present invention relates to the technical field of Internet of Things, and specifically to a remote diagnosis and treatment system and a diagnosis and treatment method based on the Internet of Things. The remote diagnosis and treatment system includes: a matching module, a diagnosis module, a distribution module and an effect monitoring module. The matching module is used to identify the type of disease and condition of the patient, and then match the corresponding doctor. The diagnosis module is used to output an independent diagnosis and treatment plan according to the type of disease and the level of condition, and compare it with the doctor's diagnosis and treatment plan uploaded by the doctor to confirm the diagnosis and treatment plan. The distribution module is used to generate an order according to the diagnosis and treatment plan and complete the order delivery. The effect monitoring module is used to determine whether the patient's diagnosis and treatment is effective, and when the patient achieves the expected diagnosis and treatment effect, the diagnosis and treatment data is recorded and uploaded. The present invention evaluates the patient's condition and the doctor's medical skills respectively, and then matches the patient with a suitable doctor, which not only improves the diagnosis and treatment success rate and the diagnosis and treatment efficiency, but also avoids the waste of doctor resources, and has high timeliness, accuracy, robustness and safety.

Description

Technical Field

[0001] The present invention relates to the technical field of remote diagnosis and treatment, and in particular to a remote diagnosis and treatment system based on the Internet of Things and a remote diagnosis and treatment method based on the Internet of Things. Background Art

[0002] In today's digital and intelligent era, the medical field is undergoing unprecedented changes. The emergence of remote diagnosis and treatment systems based on the Internet of Things has brought new opportunities for the accessibility and quality improvement of medical services, and the development of remote diagnosis and treatment has far-reaching significance. The remote diagnosis and treatment system also provides rich data and cases for medical research and education. Doctors can analyze a large amount of remote diagnosis and treatment data, summarize the laws of diseases and treatment effects, and promote the progress of medicine. At the same time, medical students can also broaden their horizons and accumulate clinical experience by participating in the process of remote diagnosis and treatment.

[0003] As an innovative medical device, the TCM Four Diagnostic Instrument combines modern technology with traditional TCM diagnostic methods. It collects information such as the patient's pulse, tongue, and complexion, and uses advanced sensors and data analysis technology to provide objective and accurate data support for TCM diagnosis. The four diagnostic methods emphasized by TCM, namely, observation, auscultation, inquiry, and palpation, have been digitized and remotely operated with the assistance of the TCM Four Diagnostic Instrument. Doctors can use these detailed data to make accurate diagnoses and lay the foundation for developing personalized treatment plans.

[0004] However, in the prior art, there is still a lack of remote TCM diagnosis and treatment systems based on the four diagnostic instruments of TCM, and the diagnosis and treatment of TCM remains in offline diagnosis and treatment. At the same time, the development of remote diagnosis and treatment systems also faces some challenges. For example, the lack of supervision of doctors and patients can easily lead to medical accidents, making it difficult to hold them accountable, and the risks of remote diagnosis and treatment are difficult to control. In addition, in existing remote diagnosis and treatment, patients and doctors are randomly matched. Even if the corresponding doctors can be found based on the basic department scope, it is difficult to match the doctor's level with the patient's condition, and there is often a waste of doctor resources or difficulty in making a correct diagnosis. The accuracy and efficiency of remote diagnosis are difficult to meet expectations. In addition, there is a lack of continuous monitoring of patients after remote diagnosis, and the safety of patients is difficult to guarantee when there are different physical conditions or misjudgments. Summary of the invention

[0005] Based on this, it is necessary to propose a remote diagnosis and treatment system and method based on the Internet of Things to address the problems of low diagnosis and treatment efficiency, low doctor-patient matching, and low diagnostic accuracy in existing remote diagnosis and treatment.

[0006] The present invention is implemented through the following technical scheme: A remote diagnosis and treatment system based on the Internet of Things includes: an information collection module, a matching module, a diagnosis module, a distribution module and an effect monitoring module.

[0007] The information collection module includes a collection unit and an authentication unit. The collection unit is used to collect patient information and doctor information. The authentication unit is used to confirm the reliability of doctor information and patient information.

[0008] The matching module includes a condition evaluation unit, a doctor evaluation unit and a matching unit. The condition evaluation unit is used to determine the type of disease and the level of the disease according to the collected patient information. The doctor evaluation unit is used to determine the doctor's diagnosis and treatment scope and ability score according to the collected doctor information, and update it according to the doctor's diagnosis and treatment data. The matching unit is used to recommend the corresponding doctor according to the patient's disease type and condition level, and select the doctor with the highest matching degree to perform remote diagnosis and treatment on the patient according to the doctor's reply or work schedule.

[0009] The diagnosis module is used to output an independent diagnosis and treatment plan based on the type of disease and the level of the condition, and compare it with the doctor's diagnosis and treatment plan uploaded by the doctor. If the plans are consistent, the doctor's diagnosis and treatment plan is output; otherwise, a consultation is initiated and the consultation diagnosis and treatment plan is output.

[0010] The distribution module is used to generate orders based on the actual treatment plan output, and deliver the corresponding medicines and medical equipment to the patient's reserved delivery address after the patient confirms the order and pays.

[0011] The effect monitoring module is used to determine whether the patient's diagnosis and treatment is effective based on the health data fed back by the patient, and to record and upload the diagnosis and treatment data when the patient achieves the expected diagnosis and treatment effect.

[0012] The above-mentioned remote diagnosis and treatment system not only realizes remote diagnosis and treatment between doctors and patients, and solves the high cost and low efficiency problems in offline diagnosis and treatment, but also effectively prevents doctors from misjudging and improves the accuracy of remote diagnosis based on the types of diseases and levels of illness of patients independently identified by the system. In addition, by evaluating the medical skills of doctors, doctors can match suitable patients in the medical field they are good at, which can not only improve the success rate and efficiency of diagnosis and treatment, but also avoid the waste of doctor resources, and has high timeliness, accuracy and robustness. By monitoring the health status of patients during treatment, it is possible to timely judge whether the diagnosis and treatment of patients is effective, avoid or reduce the deterioration of patients' conditions due to misdiagnosis, and improve the safety of remote diagnosis and treatment.

[0013] Furthermore, the collection unit includes a patient collection terminal, a doctor collection terminal, and a big data collection terminal. The patient collection terminal is used to collect the patient's health data. The doctor collection terminal is used to collect the doctor's medical data. The big data collection terminal is used to collect the diagnosis and treatment big data based on doctors and patients.

[0014] Furthermore, health data includes real-time health data and historical health data. Real-time health data is acquired and uploaded by the four diagnostic instruments. Historical health data is screened by the diagnosis and treatment big data or uploaded by the patient and then uploaded after confirmation by the authentication unit. Medical data includes basic information of doctors and historical diagnosis and treatment data.

[0015] Furthermore, the judgment process of the condition evaluation unit is as follows: Construct a condition evaluation model based on a convolutional neural network. Use the patient health data in the diagnosis and treatment big data as input, and the disease type and condition level as output to train the condition evaluation model, and retain the model parameters that meet the test accuracy. Input the health data into the condition evaluation model to obtain the corresponding disease type and condition level.

[0016] Furthermore, the evaluation process of the doctor evaluation unit is as follows: Construct a medical skill scoring model based on machine learning. Use the medical skill data in the diagnosis and treatment big data as input and the doctor's score as output to train the medical skill scoring model and retain the model parameters that meet the test accuracy. Input the medical skill data into the medical skill scoring model to obtain the corresponding score. According to the disease types and corresponding scores selected by the doctor, the disease types with scores higher than the preset minimum score are retained and displayed in the diagnosis and treatment system. Set the experience weight and update the doctor's score according to the diagnosis and treatment data completed by the doctor in the diagnosis and treatment system.

[0017] Furthermore, the diagnosis module includes an intelligent diagnosis unit, an inquiry unit, a comprehensive diagnosis unit and a consultation unit. The intelligent diagnosis unit is used to output the corresponding autonomous diagnosis and treatment plan according to the type of disease and the level of condition output by the condition evaluation module. The inquiry unit is used to output the doctor's diagnosis results and the corresponding doctor's diagnosis and treatment plan. The comprehensive diagnosis unit is used to determine whether the doctor's diagnosis results and the doctor's diagnosis and treatment plan are consistent with the autonomous diagnosis and treatment plan. If so, the actual diagnosis and treatment plan is confirmed, otherwise the consultation unit is activated. The consultation unit is used to match at least two doctors with the diagnosed doctors to form a consultation group according to the type of disease and condition, jointly diagnose the patient remotely or offline, and record the diagnosis and treatment data after the diagnosis.

[0018] Furthermore, the diagnostic process of the intelligent diagnosis unit is as follows: Construct an intelligent diagnosis model based on a convolutional neural network. Take the basic information and health data of patients in the diagnosis and treatment big data as input, and the diagnosis and treatment plan as output, train the intelligent diagnosis model, and retain the model parameters that meet the test accuracy. Input the disease type, disease level, and basic information of the patient output by the disease evaluation unit into the intelligent diagnosis model to obtain the corresponding diagnosis and treatment plan.

[0019] Furthermore, the consultation process of the consultation unit is as follows: the matching unit outputs at least two doctors with the highest matching degree, and forms a consultation group with the doctor who has made the diagnosis. The patient and the consultation group jointly confirm the online consultation time. The doctors in the consultation group conduct online consultations with the patients respectively. If the consultation results are consistent, a common diagnosis and treatment plan is output. Otherwise, an offline consultation is conducted. The patient and the consultation group confirm the offline consultation time. If the offline consultation confirms the common diagnosis and treatment plan, the consultation data is recorded and uploaded, and jointly certified by all doctors in the consultation group. Otherwise, the patient is regarded as a patient with a difficult and complicated disease, and a superior consultation group is formed for consultation until the diagnosis and treatment plan is confirmed.

[0020] Furthermore, the monitoring process of the effect monitoring module is as follows: Construct a diagnosis and treatment effect prediction model based on deep learning. Take the patient health data, diagnosis and treatment plan and diagnosis and treatment progress in the diagnosis and treatment big data as input, and the diagnosis and treatment effect and diagnosis and treatment efficiency as output, train the diagnosis and treatment effect prediction model, and retain the model parameters that meet the test accuracy. Input the real-time monitored diagnosis and treatment progress, patient health data and corresponding diagnosis and treatment plan into the diagnosis and treatment effect prediction model to obtain the diagnosis and treatment effect and diagnosis and treatment efficiency, and judge whether the diagnosis and treatment effect is in line with the diagnosis and treatment plan. If it is in line with the diagnosis and treatment plan, the diagnosis and treatment plan is valid. If it is not in line with the diagnosis and treatment plan, the diagnosis and treatment plan is invalid, and the consultation unit is started.

[0021] The present invention also provides a remote diagnosis and treatment method based on the Internet of Things, comprising the following steps:

[0022] Patients actively upload real-time health data and request diagnosis and treatment.

[0023] The constructed disease assessment model is used to identify the type of disease and the level of disease of the patient. The corresponding doctor is matched according to the type of disease and the level of disease of the patient.

[0024] The corresponding autonomous diagnosis and treatment plan is output according to the disease type and disease level output by the disease evaluation model. The doctor conducts remote diagnosis and treatment and uploads the doctor's diagnosis and treatment plan.

[0025] Determine whether the self-diagnosis and treatment plan is consistent with the doctor's diagnosis and treatment plan. If yes, confirm the diagnosis and treatment plan. Otherwise, consult with the patient. After the consultation, directly upload the diagnosis and treatment plan.

[0026] A delivery order is generated based on the diagnosis and treatment plan, and the diagnosis and treatment plan and required medicines and medical devices are sent to the delivery address provided by the patient.

[0027] Monitor the patient's treatment process, determine whether the treatment plan is effective, and record and upload treatment data when the patient achieves the expected treatment effect.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present invention not only realizes remote diagnosis and treatment between doctors and patients, and solves the high cost and low efficiency problems in offline diagnosis and treatment, but also effectively prevents doctors from misjudging and improves the accuracy of remote diagnosis based on the types of diseases and levels of illness of patients independently identified by the system. In addition, by evaluating the medical skills of doctors, doctors can match suitable patients in the medical field they are good at, which can not only improve the success rate and efficiency of diagnosis and treatment, but also avoid the waste of doctor resources, and has high timeliness, accuracy and robustness. By monitoring the health status of patients during treatment, it is possible to timely judge whether the diagnosis and treatment of patients is effective, avoid or reduce the deterioration of the patient's condition due to misdiagnosis, and improve the safety of remote diagnosis and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a schematic diagram of the structure of a remote diagnosis and treatment system based on the Internet of Things according to Embodiment 1 of the present invention;

[0031] Figure 2 This is a step diagram of the remote diagnosis and treatment method based on the Internet of Things according to Example 2 of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] It should be noted that when a component is referred to as being "mounted on" another component, it may be directly on the other component or there may be a central component. When a component is considered to be "set on" another component, it may be directly set on the other component or there may be a central component at the same time. When a component is considered to be "fixed to" another component, it may be directly fixed on the other component or there may be a central component at the same time.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "or / and" used herein includes any and all combinations of one or more of the related listed items.

[0035] Example 1: Please refer to Figure 1 , this embodiment provides a remote diagnosis and treatment system based on the Internet of Things, including: an information collection module, a matching module, a diagnosis module, a distribution module and an effect monitoring module.

[0036] The information collection module includes a collection unit and an authentication unit. The collection unit is used to collect patient information and doctor information. Specifically, the collection unit includes a patient collection terminal, a doctor collection terminal, and a big data collection terminal.

[0037] The patient collection terminal is used to collect the patient's health data. Health data includes real-time health data and historical health data. In this embodiment, real-time health data is acquired and uploaded by the four diagnostic instruments. The four diagnostic instruments are Chinese medicine auxiliary diagnosis and treatment instruments that can collect the patient's pulse data, tongue data, facial data and inquiry data. During the remote diagnosis and treatment process, the patient can choose to have a video call with the doctor for diagnosis and treatment, or choose to have the doctor view the real-time health data for direct diagnosis. Historical health data is screened by the diagnosis and treatment big data or uploaded by the patient and then uploaded after confirmation by the authentication unit.

[0038] The doctor collection terminal is used to collect the doctor's medical data. Medical data includes basic information of doctors and historical diagnosis and treatment data. Among them, medical data can be obtained through big data screening, or it can be uploaded by the doctor on his own, and then the doctor registration is completed after certification by the certification unit, and the corresponding basic information is retained as medical data. In addition, the remote diagnosis and treatment cases completed by registered doctors are also saved as historical diagnosis and treatment data.

[0039] The big data collection terminal is used to collect the diagnosis and treatment big data based on doctors and patients. The diagnosis and treatment big data can be obtained through the network, and after confirmation by the authentication unit, it is pre-processed and obtained by deleting duplicate data, missing data, and erroneous data. Among them, the diagnosis and treatment big data includes multiple independent historical diagnosis and treatment data, and each historical diagnosis and treatment data includes the corresponding patient information, doctor information and diagnosis and treatment process information.

[0040] The authentication unit is used to confirm the reliability of doctor information and patient information. The authentication unit can be manually confirmed or authenticated through intelligent recognition. For example, the authentication unit can be jointly managed by multiple managers. During authentication, the doctor information and patient information are confirmed by at least two managers. Alternatively, the doctor information and patient information are identified by intelligent recognition software. For example, in the submitted doctor information and patient information, the official seal confirmed by the relevant organization or department can be identified. For example, after the doctor uploads the physician qualification certificate, the intelligent recognition software identifies the validity of the qualification certificate and retains the corresponding doctor information after confirming its validity.

[0041] The matching module includes a condition evaluation unit, a doctor evaluation unit and a matching unit. The condition evaluation unit is used to determine the patient's disease and condition level based on the collected patient information. In the diagnosis and treatment big data, each disease and the corresponding condition are first classified by manual classification. Among them, the disease can also be classified so that each doctor can match the diagnosis and treatment field in which he is good at. In this embodiment, the disease is first classified according to the degree of harm to the human body, and the disease is divided into general diseases, low-risk diseases, and high-risk diseases. Then, for higher-level diseases, such as low-risk diseases and high-risk diseases, the disease level is further divided, such as low-risk diseases are divided into low-level conditions and high-level conditions, and high-risk diseases are divided into low-level conditions, intermediate conditions, high-level conditions and special conditions.

[0042] In this embodiment, the judgment process of the condition evaluation unit is as follows:

[0043] Construct a disease evaluation model based on convolutional neural networks. Convolutional neural networks extract the features of patients' health data and input these features into the evaluation model to obtain the corresponding disease types and disease levels. Convolutional neural networks can automatically learn the potential features of data without manual feature selection.

[0044] The patient health data in the diagnosis and treatment big data is used as input, and the disease type and condition level are used as output to train the condition evaluation model, and the model parameters that meet the test accuracy are retained. In this embodiment, 2000 historical diagnosis and treatment data are randomly selected from the big data, and 70% of them are used as the training set, and the remaining 30% are used as the test set to train the condition evaluation model.

[0045] Input the health data into the disease evaluation model to obtain the corresponding disease type and disease level.

[0046] The disease assessment model based on convolutional neural networks has strong automatic feature extraction capabilities, does not require manual feature selection, and can process complex medical data such as imaging and physiological indicators, and can achieve rapid and standardized disease assessment.

[0047] The doctor evaluation unit is used to determine the doctor's diagnosis and treatment scope and ability score based on the collected doctor information, and update it according to the doctor's diagnosis and treatment data.

[0048] The evaluation process of the doctor evaluation unit is as follows:

[0049] Construct a medical skill scoring model based on machine learning. In this embodiment, a medical skill scoring model based on random forest is used, which includes multiple decision trees, each of which outputs a scoring result. The final medical skill score is obtained by weighting the outputs of multiple decision trees.

[0050] The medical skills data in the diagnosis and treatment big data is used as input, and the doctor's score is used as output to train the medical skills scoring model, and the model parameters that meet the test accuracy are retained. In this embodiment, the main criteria for doctor scoring include physician grade and diagnosis and treatment experience. Among them, the physician grade is obtained by an industry-recognized grade assessment, such as the doctor's professional grade assessment, the physician's certification by relevant organizations or departments, etc. Diagnosis and treatment experience includes recent (such as within two years) general diagnosis and treatment data and long-term (since the start of practice) major diagnosis and treatment data. For example, the doctor's cure data for a certain difficult and complicated disease or a major medical accident, etc., all need to be recorded for a long time as an important basis for the doctor's medical skills score.

[0051] Input the medical skill data into the medical skill scoring model to obtain the corresponding score.

[0052] The medical skill scoring model based on machine learning is more objective, fair and accurate, avoiding large deviations in the evaluation of doctors' medical skills due to subjective impressions, and can reasonably match patients with different levels of conditions to avoid doctors being unable to solve patients' conditions, resulting in low diagnosis and treatment efficiency, or doctors with higher medical skills diagnosing patients with lower levels, resulting in a waste of doctor resources.

[0053] According to the disease types and corresponding scores selected by the doctors, the disease types with scores higher than the preset minimum score are retained and displayed in the diagnosis and treatment system. In this embodiment, the minimum score is set to 80. For any doctor, he can independently choose the type of disease to diagnose, such as internal medicine, surgery, pediatrics, acupuncture, etc., and score each type through the medical skills scoring model, and retain the disease types with scores higher than 80 to avoid problems such as doctors having difficulty in diagnosing the corresponding diseases or low diagnostic efficiency.

[0054] Set the experience weight, and update the doctor's score based on the diagnosis and treatment data completed by the doctor in the diagnosis and treatment system. The doctor's diagnosis and treatment experience includes offline diagnosis and treatment experience and online diagnosis and treatment experience. In this embodiment, the weight of online diagnosis and treatment experience is set to 0.7, and the weight of offline diagnosis and treatment experience is set to 0.3. The doctor's online diagnosis and treatment medical skill score a and offline medical skill score b are output in the medical skill evaluation model respectively, and the comprehensive medical skill score is 0.7a+0.3b, where the online and offline diagnosis and treatment data are not less than 10.

[0055] The matching unit is used to recommend corresponding doctors according to the type of disease and the level of the patient's condition, and select the doctor with the highest matching degree to provide remote diagnosis and treatment to the patient based on the doctor's reply or work schedule. After uploading health information or making a diagnosis and treatment request, the patient also selects a recent available call time period. The matching unit matches according to the schedule set by the doctor, and selects doctors who meet the general diagnosis and treatment time and have a high matching degree. Among them, the matching degree is based on the patient's disease type and condition level output by the condition evaluation unit, and the doctor with the corresponding diagnosis and treatment scope and adapted score is queried and selected in the preset score matching table. Among them, the score matching table is used to characterize the mapping relationship between the disease type and condition level and the medical skill score. For example, low-level conditions of low-risk diseases are adapted to medical skill scores of 80 points.

[0056] The diagnosis module is used to output an independent diagnosis and treatment plan based on the type of disease and the level of the condition, and compare it with the doctor's diagnosis and treatment plan uploaded by the doctor. If the plans are consistent, the doctor's diagnosis and treatment plan is output; otherwise, a consultation is initiated and the consultation diagnosis and treatment plan is output.

[0057] The diagnosis module includes an intelligent diagnosis unit, a questioning unit, a comprehensive diagnosis unit and a consultation unit. The intelligent diagnosis unit is used to output the corresponding autonomous diagnosis and treatment plan according to the disease type and disease level output by the disease evaluation module.

[0058] The diagnostic process of the intelligent diagnostic unit is as follows:

[0059] Build an intelligent diagnosis model based on convolutional neural network.

[0060] The basic information and health data of patients in the diagnosis and treatment big data are used as input, and the diagnosis and treatment plan is used as output. The intelligent diagnosis model is trained and the model parameters that meet the test accuracy are retained.

[0061] The disease type, disease level and basic information of the patient output by the disease evaluation unit are input into the intelligent diagnosis model to obtain the corresponding diagnosis and treatment plan.

[0062] In other embodiments, it is also possible to screen and integrate big data and create a treatment plan matching table to provide corresponding treatment plans for the patient's age group, gender, disease type, and condition level. The difference is that although the treatment plan matching table has a higher matching efficiency, it is difficult to effectively combine the patient's health information. Compared with the treatment plan output by the intelligent diagnosis model, the treatment plan matching table has a lower degree of fit with the patient.

[0063] The consultation unit is used to output the doctor's diagnosis results and the corresponding doctor's treatment plan. The consultation unit is based on the doctor's independent judgment, and the doctor actively uploads the diagnosis results and the corresponding treatment plan. If the doctor cannot diagnose the result based on the patient's health data alone, he can choose to make an online call diagnosis or an offline diagnosis. If the offline diagnosis still cannot diagnose the result, the diagnosis failure is selected and other doctors of the same level are diagnosed. If two consecutive doctors of the same level fail to diagnose, the doctor at the next level will be selected for diagnosis and treatment. For example, a doctor with a medical skill score of 80-85 is regarded as a first-level doctor, and a doctor with a medical skill score of 86-90 is regarded as a second-level doctor.

[0064] The comprehensive diagnosis unit is used to determine whether the doctor's diagnosis results and the doctor's treatment plan are consistent with the autonomous treatment plan. If so, the actual treatment plan is confirmed, otherwise the consultation unit is activated. Among them, when judging the diagnosis results, the doctor can make independent choices by setting options to avoid misjudgments due to unclear text recognition. When judging the treatment plan, select the semantic duplication check method for comparison. In this embodiment, the semantic overlap is more than 80% to confirm consistency.

[0065] The consultation unit is used to match at least two doctors with the doctor who has made the diagnosis according to the disease type and condition to form a consultation group, jointly diagnose the patient remotely or offline, and record the diagnosis and treatment data after diagnosis. In order to prevent a single doctor from misjudging the patient's condition, when the comprehensive diagnosis is inconsistent, other doctors should be selected for diagnosis to avoid the patient's condition being aggravated due to incorrect diagnosis and treatment plans.

[0066] The consultation process of the consultation unit is as follows:

[0067] The matching unit outputs at least two doctors with the highest matching degree to form a consultation group with the doctor who has made the diagnosis. Generally speaking, the consultation group consists of three doctors of the same level. If necessary, doctors of the next higher level can be selected to form a consultation group.

[0068] The patient and the consultation team jointly confirm the online consultation time. In the remote diagnosis and treatment system, online diagnosis is still preferred to improve diagnosis efficiency and reduce diagnosis costs.

[0069] Doctors in the consultation group conduct online consultations with patients separately. If the consultation results are consistent, a common diagnosis and treatment plan is output. Otherwise, an offline consultation is conducted. In the online consultation, in order to prevent interference between doctors, each doctor makes a separate diagnosis, and then the comprehensive diagnosis unit determines whether the diagnosis results given by the three doctors are consistent. If they are consistent, the consultation group jointly confirms the diagnosis and treatment plan. If they are inconsistent, the diagnosis and treatment group is given priority for offline consultation.

[0070] The patient and the consultation team confirm the offline consultation time. If the offline consultation confirms a common diagnosis and treatment plan, the consultation data will be recorded and uploaded, and all doctors in the consultation team will jointly verify it. Otherwise, the patient will be treated as a patient with a difficult and complicated disease, and a higher-level consultation team will be formed to consult until the diagnosis and treatment plan is confirmed.

[0071] The delivery module is used to generate orders based on the actual treatment plan output, and deliver the corresponding drugs and medical equipment to the patient's reserved delivery address after the patient confirms the order and pays. The treatment plan generally includes drugs and dosage, auxiliary medical equipment, etc. The order should not only include drugs and medical equipment, but also the doctor's entire treatment plan.

[0072] The effect monitoring module is used to determine whether the patient's diagnosis and treatment is effective based on the health data fed back by the patient, and to record and upload the diagnosis and treatment data when the patient achieves the expected diagnosis and treatment effect. After the patient receives the diagnosis and treatment plan and undergoes the diagnosis and treatment, the patient's health data should be monitored in a timely manner to determine whether the patient's condition has improved to prevent the patient's condition from continuing to deteriorate.

[0073] The monitoring process of the effect monitoring module is as follows:

[0074] Construct a diagnosis and treatment effect prediction model based on deep learning. The prediction model includes an input layer, a feature extraction layer, a hidden layer, and an output layer. The input layer is used to receive the patient's real-time health data and diagnosis and treatment progress data. The feature extraction layer uses LSTM to extract valuable features from the input data. The hidden layer can perform nonlinear changes on the extracted features to capture deep patterns and relationships in the data. The output layer is used to output the diagnosis and treatment effect and diagnosis and treatment efficiency.

[0075] The patient health data, treatment plan and treatment progress in the diagnosis and treatment big data are taken as input, and the diagnosis and treatment effect and treatment efficiency are taken as output. The diagnosis and treatment effect prediction model is trained and the model parameters that meet the test accuracy are retained.

[0076] The real-time monitored treatment progress, patient health data and corresponding treatment plan are input into the treatment effect prediction model to obtain the treatment effect and treatment efficiency, and judge whether the treatment effect is consistent with the treatment plan. If it is consistent, the treatment plan is valid. If not, the treatment plan is invalid and the consultation unit is started.

[0077] In this embodiment, the remote diagnosis and treatment system is implemented through an intelligent diagnosis and treatment platform constructed by the network, and the intelligent diagnosis and treatment platform includes a user end, a doctor end and a server. Among them, the user end is the patient login end, which can upload the patient's basic information and health data. The doctor end is the doctor login end, which can upload the doctor's basic information, diagnosis and treatment experience data, and the diagnosis results and treatment plans of each diagnosis and treatment. The server can realize the general functions of the matching module, diagnosis module, distribution module and effect monitoring module. After the patient completes the information registration, the server authenticates the patient's basic information, and then uploads the health data uploaded by the patient on his own, or the health data monitored by the patient in real time, and generates a diagnosis order after server authentication, or the patient chooses to generate a diagnosis order on his own. After the server identifies the patient's disease and condition level, it matches the doctor in the relevant field, and inquires the corresponding doctor in turn according to the matching degree. After the doctor confirms the order, the relevant information is sent to the doctor, and the doctor uploads the diagnosis result. After the diagnosis result and treatment plan are confirmed, a distribution order is generated and the relevant medicines are sent to the patient. During the treatment process, the patient uploads real-time health data according to the frequency of the treatment plan, and the server determines whether the treatment plan is effective.

[0078] The remote diagnosis and treatment system based on the Internet of Things in this embodiment not only realizes remote diagnosis and treatment between doctors and patients, but also solves the high cost and low efficiency problems in offline diagnosis and treatment. At the same time, according to the patient's disease type and condition level independently identified by the system, it effectively prevents doctors from misjudging and improves the accuracy of remote diagnosis. In addition, by evaluating the doctors' medical skills, doctors can match suitable patients in the medical field they are good at, which can not only improve the success rate and efficiency of diagnosis and treatment, but also avoid the waste of doctor resources, and has high timeliness, accuracy and robustness. By monitoring the health status of patients during treatment, it is possible to timely judge whether the patient's diagnosis and treatment is effective, avoid or reduce the deterioration of the patient's condition due to misdiagnosis, and improve the safety of remote diagnosis and treatment.

[0079] Example 2: Please refer to Figure 2 This embodiment provides a remote diagnosis and treatment method based on the Internet of Things, which can be applied to the remote diagnosis and treatment system based on the Internet of Things in Example 1, and includes the following steps:

[0080] Patients actively upload real-time health data and make treatment requests. First, a remote diagnosis and treatment platform based on the Internet of Things is built. After completing registration, patients actively make treatment requests, and then generate treatment orders on the remote diagnosis and treatment platform.

[0081] The constructed disease evaluation model is used to identify the type of disease and the level of the disease. The corresponding doctor is matched according to the type of disease and the level of the disease. After the doctor completes the registration, the remote diagnosis and treatment platform authenticates the doctor and scores the doctor based on the doctor's diagnosis and treatment experience data. After identifying the type of disease and the level of the disease, the remote diagnosis and treatment platform identifies the corresponding diagnosis and treatment field and outputs the best matching score. The doctor with the corresponding score is selected according to the matching score.

[0082] The corresponding autonomous diagnosis and treatment plan is output according to the disease type and disease level output by the disease evaluation model. The doctor conducts remote diagnosis and treatment and uploads the doctor's diagnosis and treatment plan. The remote diagnosis and treatment platform integrates the diagnosis and treatment big data to obtain the diagnosis and treatment plan corresponding to each disease type and disease level, and makes fine adjustments based on the patient's age, gender, etc., and then directly searches and outputs the corresponding autonomous diagnosis and treatment plan after identifying the patient's disease type and disease level. The doctor can directly diagnose based on the health data uploaded by the patient, or conduct remote video call diagnosis or offline diagnosis, and output the diagnosis results and the doctor's diagnosis and treatment plan.

[0083] Determine whether the self-diagnosis and treatment plan is consistent with the doctor's diagnosis and treatment plan. If yes, confirm the diagnosis and treatment plan. Otherwise, consult with the patient. Upload the diagnosis and treatment plan directly after the consultation. If the diagnosis and treatment plan output by the doctor and the remote diagnosis and treatment platform are consistent, it can be basically judged that the diagnosis and treatment is correct. If there is a disagreement, more doctors should jointly confirm to avoid misdiagnosis.

[0084] Generate a delivery order based on the diagnosis and treatment plan, and send the diagnosis and treatment plan and the required medicines and medical devices to the delivery address provided by the patient. The remote diagnosis and treatment platform based on the Internet of Things can directly generate delivery orders and complete the delivery, allowing patients to complete the diagnosis and treatment process without leaving home, which not only improves the efficiency of diagnosis and treatment, but also avoids the deterioration of the condition caused by long-distance travel.

[0085] Monitor the patient's treatment process, determine whether the treatment plan is effective, and record and upload the treatment data when the patient achieves the expected treatment effect. Even if the treatment plan is confirmed, the correctness of the treatment effect cannot be guaranteed. Therefore, it is extremely necessary to monitor the patient in a timely manner so that the patient can achieve the expected treatment effect through stable treatment. At the same time, the treatment plan can be terminated in time when an abnormality occurs to avoid causing greater harm to the patient.

[0086] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0087] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A remote diagnosis and treatment system based on the Internet of Things, characterized in that: include: Information collection module, including a collection unit and an authentication unit; The collection unit is used to collect patient information and doctor information; The authentication unit is used to confirm the reliability of the doctor information and the patient information; The collection unit includes a patient collection terminal, a doctor collection terminal, and a big data collection terminal; the patient collection terminal is used to collect the patient's health data; the doctor collection terminal is used to collect the doctor's medical data; the big data collection terminal is used to collect the diagnosis and treatment big data based on the doctor and the patient; The matching module includes a condition evaluation unit, a doctor evaluation unit and a matching unit; the condition evaluation unit is used to determine the type of disease and the level of the disease of the patient based on the collected patient information; The doctor evaluation unit is used to determine the doctor's diagnosis and treatment scope and ability score based on the collected doctor information, and update it according to the doctor's diagnosis and treatment data; the matching unit is used to recommend the corresponding doctor according to the patient's disease type and condition level, and select the doctor with the highest matching degree to perform remote diagnosis and treatment on the patient according to the doctor's reply or work schedule; the matching degree is queried in a preset score matching table based on the patient's disease type and condition level output by the condition evaluation unit; a doctor with a corresponding diagnosis and treatment scope and a suitable score is selected; wherein the score matching table is used to characterize the mapping relationship between the disease type and condition level and the medical skill score; The judgment process of the condition evaluation unit is as follows: Construct a disease assessment model based on convolutional neural network; The patient health data in the diagnosis and treatment big data is used as input, and the disease type and condition level are used as output to train the condition evaluation model, and retain the model parameters that meet the test accuracy; Input the health data into the disease assessment model to obtain the corresponding disease type and disease level; The evaluation process of the doctor evaluation unit is as follows: Construct a medical skill scoring model based on machine learning; Using the medical skill data in the diagnosis and treatment big data as input and the doctor's score as output, the medical skill scoring model is trained, and the model parameters that meet the test accuracy are retained; Inputting the medical skill data into the medical skill scoring model to obtain a corresponding score; According to the disease types and corresponding scores selected by the doctors, the disease types with scores higher than the preset minimum score will be retained and displayed in the diagnosis and treatment system; Set experience weights and update doctor scores based on the diagnosis and treatment data completed by the doctor in the diagnosis and treatment system; The diagnosis module is used to output an autonomous diagnosis and treatment plan according to the disease type and condition level, and compare it with the doctor's diagnosis and treatment plan uploaded by the doctor. If the plans are consistent, the doctor's diagnosis and treatment plan is output; otherwise, a consultation is initiated and the consultation diagnosis and treatment plan is output; the diagnosis module includes an intelligent diagnosis unit, an inquiry unit, a comprehensive diagnosis unit and a consultation unit; the intelligent diagnosis unit is used to output the corresponding autonomous diagnosis and treatment plan according to the disease type and condition level output by the condition evaluation unit; the inquiry unit is used to output the doctor's diagnosis result and the corresponding doctor's diagnosis and treatment plan; the comprehensive diagnosis unit is used to determine whether the doctor's diagnosis result and the doctor's diagnosis and treatment plan are consistent with the autonomous diagnosis and treatment plan, and if so, the actual diagnosis and treatment plan is confirmed, otherwise the consultation unit is initiated; the consultation unit is used to match at least two doctors with the diagnosed doctor to form a consultation group according to the disease type and condition, and jointly diagnose the patient remotely or offline, and record the diagnosis and treatment data after the diagnosis; The diagnostic process of the intelligent diagnostic unit is as follows: Construct an intelligent diagnosis model based on convolutional neural network; Input the disease type, disease level and basic information of the patient output by the disease evaluation unit into the intelligent diagnosis model to obtain a corresponding diagnosis and treatment plan; The consultation process of the consultation unit is as follows: The matching unit outputs at least two doctors with the highest matching degree, and they form a consultation group with the doctor who has made the diagnosis; The patient and the consultation team jointly confirm the online consultation time; Doctors in the consultation group conduct online consultations with patients separately; if the consultation results are consistent, a common diagnosis and treatment plan is output; otherwise, an offline consultation is conducted; The patient and the consultation team confirm the offline consultation time; if the offline consultation confirms a common diagnosis and treatment plan, the consultation data will be recorded and uploaded, and jointly certified by all doctors in the consultation team; otherwise, the patient will be treated as a patient with a difficult and complicated disease, and a higher-level consultation team will be formed to consult until the diagnosis and treatment plan is confirmed; The delivery module is used to generate an order based on the actual diagnosis and treatment plan output, and deliver the corresponding medicines and medical equipment to the patient's reserved delivery address after the patient confirms the order and pays; The effect monitoring module is used to determine whether the patient's diagnosis and treatment is effective based on the health data fed back by the patient, and to record and upload the diagnosis and treatment data when the patient achieves the expected diagnosis and treatment effect.

2. According to the remote diagnosis and treatment system based on the Internet of Things according to claim 1, it is characterized in that: The health data includes real-time health data and historical health data; the real-time health data is acquired and uploaded by four diagnostic instruments; the historical health data is screened by the diagnosis and treatment big data or uploaded by the patient and then confirmed by the authentication unit and uploaded; the medical data includes basic information of the doctor and historical diagnosis and treatment data.

3. The remote diagnosis and treatment system based on the Internet of Things according to claim 1, characterized in that: The monitoring process of the effect monitoring module is as follows: Construct a diagnosis and treatment effect prediction model based on deep learning; Taking the patient health data, diagnosis and treatment plan and diagnosis and treatment progress in the diagnosis and treatment big data as input, and the diagnosis and treatment effect and diagnosis and treatment efficiency as output, training the diagnosis and treatment effect prediction model, and retaining the model parameters that meet the test accuracy; The real-time monitored treatment progress, patient health data and corresponding treatment plan are input into the treatment effect prediction model to obtain the treatment effect and treatment efficiency, and judge whether the treatment effect is in line with the treatment plan. If so, the treatment plan is valid; if not, the treatment plan is invalid and the consultation unit is started.

4. A remote diagnosis and treatment method based on the Internet of Things, which is applied to a remote diagnosis and treatment system based on the Internet of Things as claimed in any one of claims 1 to 3, characterized in that: The steps include: Patients actively upload real-time health data and send out diagnosis and treatment requests; Use the established disease assessment model to identify the patient's disease type and disease level; match the corresponding doctor according to the patient's disease type and disease level; Outputting corresponding autonomous diagnosis and treatment plans according to the disease types and disease levels output by the disease evaluation model; doctors conduct remote diagnosis and treatment and upload the doctor's diagnosis and treatment plans; Determine whether the self-diagnosis and treatment plan is consistent with the doctor's diagnosis and treatment plan, and if so, confirm the diagnosis and treatment plan; otherwise, consult with the patient; directly upload the diagnosis and treatment plan after the consultation; Generate a delivery order based on the diagnosis and treatment plan, and send the diagnosis and treatment plan and the required drugs and medical devices to the delivery address provided by the patient; Monitor the patient's treatment process to determine whether the treatment plan is effective, and record and upload treatment data when the patient achieves the expected treatment effect.

Images