Maintaining hemodialysis patient management system and method based on artificial intelligence
Through the artificial intelligence-based hemodialysis patient management system, encrypted data transmission and human-computer interaction functions, the appointment and privacy issues in the management of hemodialysis patients are solved, the targetedness and patient autonomy of the treatment plan are improved, and information security is ensured. Reduced waiting time.
Patent Information
- Application Number
- CN202411097356.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-11-22
AI Technical Summary
In the existing hemodialysis patient management system, patients need to frequently travel to and from the hospital to make appointments for treatment. The treatment effect is limited by the untimely distribution of doctors and the lack of privacy protection for data transmission, resulting in a high risk of patient information leakage.
A maintenance hemodialysis patient management system based on artificial intelligence is designed, including patient terminals, doctor terminals, cloud storage units and comprehensive analysis units. Through wireless encrypted data transmission, a matching mechanism between patients and doctors is established to improve the targeted treatment plan. , and enhance patient autonomy through human-computer interaction function.
It improves the pertinence of treatment plans, enhances the convenience of interaction between patients and doctors, ensures the privacy and security of patient information, reduces waiting time and improves the quality of life of patients.
Smart Images

Figure CN119007960A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of hemodialysis, and in particular to an artificial intelligence-based maintenance hemodialysis patient management system and method. Background Art
[0002] Hemodialysis is one of the effective treatments for acute and chronic renal failure. Due to the shortage of kidney sources, maintenance hemodialysis is a renal replacement therapy for many patients with end-stage renal disease to remove metabolic waste from the body, maintain the function of important organs and maintain the stability of the internal environment.
[0003] At present, there are many patients who need long-term hemodialysis. They need to go to the hospital 2 to 3 times a week, and they need to make an appointment at the hospital each time. In most cases, they have to wait for a long time. In addition, due to the complex conditions of dialysis patients, the best treatment doctor cannot be assigned to the patient in time, which may have a certain impact on the treatment effect. In addition, in the current hemodialysis patient management system, privacy and possible leakage during data transmission are not considered, and patient information is not guaranteed. Summary of the invention
[0004] The purpose of the present invention is to provide a maintenance hemodialysis patient management system and method based on artificial intelligence, establish a patient-doctor matching mechanism, improve the pertinence of treatment plans, add a human-computer interaction function, facilitate multi-party use between patients and doctors, and have high autonomy and strong practicality.
[0005] To achieve the above-mentioned object, the present invention provides a maintenance hemodialysis patient management system based on artificial intelligence, comprising a patient terminal, a doctor terminal, a cloud storage unit, a comprehensive analysis unit and a hemodialysis machine; the patient unit, the doctor unit and the cloud storage unit are all connected to the comprehensive analysis unit via wireless; the data transmission between the patient terminal, the cloud storage unit, the doctor terminal and the comprehensive analysis unit is encrypted;
[0006] The cloud storage unit is used to store patient information and doctor information; the patient terminal includes a patient information registration module, a patient information collection module, a consultation module and a doctor's advice reminder module; the doctor's terminal includes a doctor's advice module and a patient information viewing module; the comprehensive analysis unit includes a doctor matching module, a condition analysis module and a treatment analysis module, and the doctor matching module includes a patient feature extraction submodule, a doctor feature extraction submodule, a matching submodule and an optimization submodule; the condition analysis module is used to generate a patient's condition report and divide the dialysis plan; the treatment analysis module is used to generate a patient's treatment plan and divide the dialysis adequacy;
[0007] The patient information registration module is used to register the patient's identity information; the patient information collection module is used to collect the patient's basic information and clinical symptom data; the clinical symptom data includes blood routine data, biochemical indicators, nutritional information, and dialysis complications; the doctor's order reminder module is used to receive the doctor's orders;
[0008] The medical order module includes long-term medical orders and temporary medical orders; the patient information viewing module is used to view patient information and treatment information, and doctors issue medical orders based on the patient information and treatment information;
[0009] The patient feature extraction submodule is used to extract and analyze the patient's basic information and clinical symptom data to obtain the patient feature vector; the doctor feature extraction submodule is used to extract and analyze the doctor's treatment plan and patient satisfaction to obtain the doctor feature vector; the matching submodule is used to calculate the similarity between the patient feature vector and the doctor feature vector, and match the doctor according to the similarity; the optimization submodule is used to optimize the patient feature extraction submodule and the doctor feature extraction submodule according to patient feedback and time factors.
[0010] Preferably, the patient terminal is any one of a mobile phone, a tablet computer and a laptop computer; the doctor terminal is any one of a mobile phone, a tablet computer and a laptop computer.
[0011] Preferably, the hemodialysis machine includes a temperature control system, a liquid preparation system, a degassing system, a conductivity monitoring system, an ultrafiltration monitoring system and a blood leakage monitoring system.
[0012] Preferably, the dialysis cycles are divided into 2 times a week, 3 times a week, and 5 times every two weeks.
[0013] Preferably, the dialysis adequacy effect assessment is divided into sufficient and inadequate.
[0014] The present invention also provides a maintenance hemodialysis patient management method based on artificial intelligence, comprising the following steps:
[0015] S1. When the patient confirms that he or she will undergo maintenance hemodialysis, the hospital provides the patient or his or her family with a patient terminal, and the patient fills in the patient information according to the prompts on the patient terminal;
[0016] S2, the patient terminal transmits the collected information to the cloud storage unit, and the cloud storage unit sends the information to the comprehensive analysis unit;
[0017] S3, the comprehensive analysis unit determines the patient's dialysis cycle based on the received information;
[0018] S4. When a patient needs hemodialysis, the patient feature extraction submodule extracts the patient feature vector, and the doctor feature extraction submodule extracts the doctor feature vectors of all doctors in the department. Then, the matching submodule calculates the cosine similarity between the patient feature vector and the doctor feature vector, sorts the corresponding doctors according to the cosine similarity, and finally recommends the patient to make an appointment for hemodialysis according to the sorting result and the doctor's schedule.
[0019] S5. The patient is in the dialysis room, and is undergoing hemodialysis through a hemodialysis machine. At this time, the doctor retrieves the patient's treatment status from the comprehensive analysis unit through the doctor's terminal; the patient asks the doctor about the treatment status through the consultation module of the patient terminal;
[0020] S6. The doctor issues medical advice to the patient through the medical advice module based on the patient information and treatment information. The medical advice reminder module of the patient terminal receives the issued medical advice and determines the next dialysis time based on the medical advice.
[0021] S7. After dialysis, optimize the patient feature extraction submodule based on patient feedback, and optimize the doctor feature extraction submodule based on time factors.
[0022] Preferably, in step S1, when the patient is undergoing dialysis for the first time, the patient's identity information, including name and address, is filled in through the patient information registration module in the patient terminal, and then the first dialysis screening is performed according to the prompt of the patient terminal, and the basic information of the patient and clinical symptom data obtained from the screening are input into the patient terminal; when the patient is not undergoing dialysis for the first time, the patient logs into the patient system by inputting the patient's identity information.
[0023] Preferably, in step S6, when the doctor's order is a long-term order, the patient's next dialysis treatment does not need to be adjusted; when the doctor's order is a temporary order, the patient's next dialysis treatment needs to be appropriately adjusted according to the patient's dry weight and condition.
[0024] Preferably, the data transmission between the patient terminal, the cloud storage unit, the doctor terminal, and the comprehensive analysis unit is encrypted, and the encryption method is as follows:
[0025] First, the data is encrypted using a symmetric encryption algorithm;
[0026] Then, the symmetric encryption key is encrypted using an asymmetric encryption algorithm;
[0027] After receiving the encrypted data and key, the receiver first uses the private key to decrypt the symmetrically encrypted key;
[0028] The decrypted key is then used to decrypt the data.
[0029] Preferably, in step S4, the cosine similarity is calculated as follows:
[0030]
[0031] Among them, u is the patient feature vector; v is the doctor feature vector; ||u|| is the modulus of the patient feature vector u; ||v|| is the modulus of the doctor feature vector v.
[0032] Therefore, the present invention adopts the above-mentioned maintenance hemodialysis patient management system and method based on artificial intelligence, and the technical effects are as follows:
[0033] (1) Through deep learning technology, patient and doctor characteristics are extracted and a matching mechanism is established. Doctors can be matched more specifically according to the patient's condition, which improves the targetedness of the treatment plan and thus improves the patient's quality of life;
[0034] (2) During the data transmission process, an encryption mechanism is set up to effectively protect patient information and prevent patient information leakage;
[0035] (3) Setting up patient terminals and doctor terminals facilitates the interaction between patients and doctors and enhances patient autonomy. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 This is a schematic diagram of the structure of a maintenance hemodialysis patient management system based on artificial intelligence in the present invention;
[0037] Figure 2 is a schematic block diagram of a patient terminal;
[0038] Figure 3 is a schematic block diagram of a doctor terminal;
[0039] Figure 4 is a schematic block diagram of a comprehensive analysis unit;
[0040] Figure 5 This is a schematic block diagram of the doctor matching module. DETAILED DESCRIPTION
[0041] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.
[0042] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.
[0043] Embodiment 1 This application involves the collection and analysis of patient data, and all data collection and processing have been approved by the Ethics Committee. During the research process, the relevant provisions of the "Ethical Review Methods for Biomedical Research Involving Humans" and the "World Medical Association Declaration of Helsinki" were strictly followed to ensure that the ethics of the research and the rights and interests of patients are fully protected. All participants voluntarily signed the informed consent form on the basis of fully understanding the purpose, process and risks of the research.
[0044] like Figure 1 As shown, the present invention provides a structural schematic diagram of an artificial intelligence-based maintenance hemodialysis patient management system, including a patient terminal, a doctor terminal, a cloud storage unit, a comprehensive analysis unit and a hemodialysis machine.
[0045] The patient terminal and the doctor terminal are any of a mobile phone, a tablet computer and a laptop computer.
[0046] The hemodialysis machine includes a temperature control system, a liquid preparation system, a degassing system, a conductivity monitoring system, an ultrafiltration monitoring system and a blood leakage monitoring system.
[0047] The patient unit, doctor unit, and cloud storage unit are all connected to the comprehensive analysis unit via wireless; the data transmission between the patient terminal, cloud storage unit, doctor terminal, and comprehensive analysis unit is encrypted;
[0048] The cloud storage unit is used to store patient information and doctor information;
[0049] like Figure 2 As shown, the patient terminal includes a patient information registration module, a patient information collection module, a consultation module and a doctor's advice reminder module;
[0050] The patient information registration module is used to register the patient's identity information; the identity information includes the patient's name, ID number, and residential address;
[0051] The patient information collection module is used to collect the patient's basic information and clinical symptom data;
[0052] The patient's basic information includes height, weight, blood pressure, temperature, heart rate, etc.;
[0053] Clinical symptom data include blood routine data, biochemical indicators, nutritional information, dialysis complications, etc.;
[0054] Routine blood data include hemoglobin, white blood cell count, platelet count, etc.;
[0055] Biochemical indicators include renal function, liver function, electrolytes, blood calcium, blood phosphorus, parathyroid hormone, etc.;
[0056] Dialysis-related complications include cardiovascular complications (such as hypertension, hypotension, heart failure, cerebral infarction, cerebral hemorrhage, myocardial infarction, etc.), imbalance syndrome, allergic reactions, etc.:
[0057] The doctor's order reminder module is used to receive the doctor's orders;
[0058] Nutritional information can be determined by standardized protein catabolism rate, body mass index, etc.
[0059] The normalized protein catabolism rate was calculated as follows:
[0060] nPCR=(α-β)×(0.045 / 2d)
[0061] Among them, nPCR is an indicator of protein intake and dialysis adequacy; α is the anterior blood urea nitrogen; β is the posterior blood urea nitrogen; d is the number of days between blood samples;
[0062] like Figure 3 As shown, the doctor terminal includes a doctor's order module and a patient information viewing module;
[0063] The medical order module includes long-term medical orders and temporary medical orders;
[0064] The patient information viewing module is used to view patient information and treatment information, and doctors issue medical advice based on the patient information and treatment information;
[0065] like Figure 4 As shown, the comprehensive analysis unit includes a doctor matching module, a condition analysis module, and a treatment analysis module;
[0066] The condition analysis module is used to generate patient condition reports and divide dialysis cycles; dialysis cycles are divided into 2 times a week, 3 times a week, and 5 times every two weeks.
[0067] The treatment analysis module is used to generate patient treatment plans and classify dialysis adequacy; the dialysis adequacy effect evaluation is divided into sufficient and inadequate.
[0068] like Figure 5 As shown, the doctor matching module includes a patient feature extraction submodule, a doctor feature extraction submodule, a matching submodule and an optimization submodule;
[0069] The patient feature extraction submodule is used to extract and analyze the patient's basic information and clinical symptom data to obtain the patient feature vector;
[0070] The doctor feature extraction submodule is used to extract and analyze the features of the doctor's treatment plan and patient satisfaction to obtain the doctor's feature vector;
[0071] The matching submodule is used to calculate the similarity between the patient feature vector and the doctor feature vector, and match the doctor based on the similarity;
[0072] The optimization submodule is used to optimize the patient feature extraction submodule and the doctor feature extraction submodule according to patient feedback and time factors respectively.
[0073] Embodiment 2
[0074] The present invention also provides a maintenance hemodialysis patient management method based on artificial intelligence, comprising the following steps:
[0075] S1. When the patient confirms that he or she will undergo maintenance hemodialysis, the hospital provides the patient or his or her family with a patient terminal, and the patient fills in the patient information according to the prompts on the patient terminal;
[0076] When the patient is undergoing dialysis for the first time, the patient's identity information, including name and address, is filled in through the patient information registration module in the patient terminal. Then, the first dialysis screening is performed according to the prompts of the patient terminal, and the basic information and clinical symptom data of the patient obtained through the screening are input into the patient terminal;
[0077] When the patient is not undergoing dialysis for the first time, he / she logs into the patient system by entering the patient's identity information.
[0078] S2, the patient terminal transmits the collected information to the cloud storage unit, and the cloud storage unit sends the information to the comprehensive analysis unit;
[0079] S3, the comprehensive analysis unit determines the patient's dialysis cycle based on the received information;
[0080] S4. When a patient needs hemodialysis, the patient feature extraction submodule extracts the patient feature vector, and the doctor feature extraction submodule extracts the doctor feature vectors of all doctors in the department. Then, the matching submodule calculates the cosine similarity between the patient feature vector and the doctor feature vector, sorts the corresponding doctors according to the cosine similarity, and finally recommends the patient to make an appointment for hemodialysis according to the sorting result and the doctor's schedule.
[0081] The cosine similarity is calculated as follows:
[0082]
[0083] Among them, u is the patient feature vector; v is the doctor feature vector, including the doctor's professional field, diseases he is good at, treatment experience, patient satisfaction, treatment effect, etc.; ||u|| is the modulus of the patient feature vector u; ||v|| is the modulus of the doctor feature vector v.
[0084] The patient feature vector and doctor feature vector are extracted through a deep learning model.
[0085] S5. The patient is in the dialysis room, and is undergoing hemodialysis through a hemodialysis machine. At this time, the doctor retrieves the patient's treatment status from the comprehensive analysis unit through the doctor's terminal; the patient asks the doctor about the treatment status through the consultation module of the patient terminal;
[0086] S6. The doctor issues medical advice to the patient through the medical advice module based on the patient information and treatment information. The medical advice reminder module of the patient terminal receives the issued medical advice and determines the next dialysis time based on the medical advice.
[0087] When the doctor's order is a long-term order, the patient's next hemodialysis treatment does not need to be adjusted; when the doctor's order is a temporary order, the patient's next hemodialysis treatment needs to be appropriately adjusted according to the patient's dry weight and condition. By issuing a doctor's order, the patient's waiting time for the next hemodialysis can be saved according to the actual situation.
[0088] S7. After dialysis, optimize the patient feature extraction submodule based on patient feedback (patient satisfaction, treatment plan), and optimize the doctor feature extraction submodule based on time factors (doctor's latest research results, etc.).
[0089] The data transmission between the patient terminal, cloud storage unit, doctor terminal and comprehensive analysis unit is encrypted as follows:
[0090] First, the data is encrypted using a symmetric encryption algorithm;
[0091] Then, the symmetric encryption key is encrypted using an asymmetric encryption algorithm;
[0092] After receiving the encrypted data and key, the receiver first uses the private key to decrypt the symmetrically encrypted key;
[0093] The decrypted key is then used to decrypt the data.
[0094] Therefore, the present invention adopts the above-mentioned artificial intelligence-based maintenance hemodialysis patient management system and method, establishes a patient-doctor matching mechanism, improves the pertinence of the treatment plan, adds a human-computer interaction function, facilitates multi-party use between patients and doctors, has high autonomy, and is highly practical.
[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.
Claims
1. A maintenance hemodialysis patient management system based on artificial intelligence, characterized in that: It includes a patient terminal, a doctor terminal, a cloud storage unit, a comprehensive analysis unit and a hemodialysis machine; the patient unit, the doctor unit and the cloud storage unit are all connected to the comprehensive analysis unit via wireless; the data transmission between the patient terminal, the cloud storage unit, the doctor terminal and the comprehensive analysis unit is encrypted; The cloud storage unit is used to store patient information and doctor information; the patient terminal includes a patient information registration module, a patient information collection module, a consultation module and a doctor's advice reminder module; The doctor terminal includes a doctor's order module and a patient information viewing module; the comprehensive analysis unit includes a doctor matching module, a condition analysis module, and a treatment analysis module. The doctor matching module includes a patient feature extraction submodule, a doctor feature extraction submodule, a matching submodule, and an optimization submodule; the condition analysis module is used to generate a patient's condition report and divide the dialysis plan; the treatment analysis module is used to generate a patient treatment plan and divide the dialysis adequacy; The patient information registration module is used to register the patient's identity information; The patient information collection module is used to collect the patient's basic information and clinical symptom data; the clinical symptom data includes blood routine data, biochemical indicators, nutritional information, and dialysis complications; the doctor's order reminder module is used to receive the doctor's orders; The medical order module includes long-term medical orders and temporary medical orders; The patient information viewing module is used to view patient information and treatment information, and doctors issue medical advice based on the patient information and treatment information; The patient feature extraction submodule is used to extract and analyze the patient's basic information and clinical symptom data to obtain the patient feature vector; The doctor feature extraction submodule is used to extract and analyze the features of the doctor's treatment plan and patient satisfaction to obtain the doctor feature vector; the matching submodule is used to calculate the similarity between the patient feature vector and the doctor feature vector, and match the doctor according to the similarity; the optimization submodule is used to optimize the patient feature extraction submodule and the doctor feature extraction submodule according to patient feedback and time factors.
2. The maintenance hemodialysis patient management system based on artificial intelligence according to claim 1, characterized in that: The patient terminal is any one of a mobile phone, a tablet computer and a laptop computer; the doctor terminal is any one of a mobile phone, a tablet computer and a laptop computer.
3. The maintenance hemodialysis patient management system based on artificial intelligence according to claim 1, characterized in that: The hemodialysis machine includes a temperature control system, a liquid preparation system, a degassing system, a conductivity monitoring system, an ultrafiltration monitoring system and a blood leakage monitoring system.
4. The maintenance hemodialysis patient management system based on artificial intelligence according to claim 1, characterized in that: The dialysis cycles were divided into 2 times a week, 3 times a week, and 5 times every two weeks.
5. The maintenance hemodialysis patient management system based on artificial intelligence according to claim 1, characterized in that: The dialysis adequacy effect was evaluated as sufficient or inadequate.
6. A maintenance hemodialysis patient management method based on artificial intelligence, characterized in that: The following steps are involved: S1. When the patient confirms that he or she will undergo maintenance hemodialysis, the hospital provides the patient or his or her family with a patient terminal, and the patient fills in the patient information according to the prompts on the patient terminal; S2, the patient terminal transmits the collected information to the cloud storage unit, and the cloud storage unit sends the information to the comprehensive analysis unit; S3, the comprehensive analysis unit determines the patient's dialysis cycle based on the received information; S4. When a patient needs hemodialysis, the patient feature extraction submodule extracts the patient feature vector, and the doctor feature extraction submodule extracts the doctor feature vectors of all doctors in the department. Then, the matching submodule calculates the cosine similarity between the patient feature vector and the doctor feature vector, sorts the corresponding doctors according to the cosine similarity, and finally recommends the patient to make an appointment for hemodialysis according to the sorting result and the doctor's schedule. S5. The patient is in the dialysis room, and is undergoing hemodialysis through a hemodialysis machine. At this time, the doctor retrieves the patient's treatment status from the comprehensive analysis unit through the doctor's terminal; the patient asks the doctor about the treatment status through the consultation module of the patient terminal; S6. The doctor issues medical advice to the patient through the medical advice module based on the patient information and treatment information. The medical advice reminder module of the patient terminal receives the issued medical advice and determines the next dialysis time based on the medical advice. S7. After dialysis, optimize the patient feature extraction submodule based on patient feedback, and optimize the doctor feature extraction submodule based on time factors.
7. The maintenance hemodialysis patient management method based on artificial intelligence according to claim 6, characterized in that: In step S1, when the patient is undergoing dialysis for the first time, the patient's identity information, including name and address, is filled in through the patient information registration module in the patient terminal, and then the first dialysis screening is performed according to the prompt of the patient terminal, and the basic information and clinical symptom data of the patient obtained from the screening are input into the patient terminal; when the patient is not undergoing dialysis for the first time, the patient logs into the patient system by inputting the patient's identity information.
8. The maintenance hemodialysis patient management method based on artificial intelligence according to claim 7, characterized in that: In step S6, when the doctor's order is a long-term order, the patient's next dialysis treatment does not need to be adjusted; when the doctor's order is a temporary order, the patient's next dialysis treatment needs to be appropriately adjusted according to the patient's dry weight and condition.
9. The maintenance hemodialysis patient management method based on artificial intelligence according to claim 8, characterized in that: The data transmission between the patient terminal, cloud storage unit, doctor terminal and comprehensive analysis unit is encrypted as follows: First, the data is encrypted using a symmetric encryption algorithm; Then, the symmetric encryption key is encrypted using an asymmetric encryption algorithm; After receiving the encrypted data and key, the receiver first uses the private key to decrypt the symmetrically encrypted key; The decrypted key is then used to decrypt the data.
10. The maintenance hemodialysis patient management method based on artificial intelligence according to claim 9, characterized in that: In step S4, the cosine similarity is calculated as follows: Among them, u is the patient feature vector; v is the doctor feature vector; ||u|| is the modulus of the patient feature vector u; ||v|| is the modulus of the doctor feature vector v.
