A personalized diagnosis and treatment plan recommendation system for general practice patients based on cognitive graph

Abstract

The invention discloses a system for recommending personalized diagnosis and treatment plans for general practitioners based on cognitive maps. The system includes a data acquisition module, a data preprocessing module, a data analysis and reasoning module, and a recommendation result display module. In the data analysis and reasoning module, first Build a general practice knowledge map, and then build a patient's personalized cognitive map of the disease development trajectory based on the patient's disease, symptoms, medication and other information and the constructed general practice knowledge map, and then recommend the patient's personalized diagnosis and treatment plan . The present invention uses the reasoning method based on the cognitive graph, so that the system can truly simulate the clinician's diagnosis and treatment ideas, and provides the clinician with an interpretable and highly acceptable clinical auxiliary decision-making tool; the present invention starts from the symptoms and formulates the personality for the patient. The chemical diagnosis and treatment plan can help patients find the cause early and receive targeted treatment. At the same time, it can also realize the early screening of dangerous diseases and prompt patients to be referred to specialist treatment in time.

Description

technical field [0001] The invention belongs to the technical field of medical and health information, and in particular relates to a system for recommending a personalized diagnosis and treatment plan for general practitioners based on a cognitive map. Background technique [0002] In my country, general medicine is a comprehensive medical discipline oriented to communities and families, integrating clinical medicine, preventive medicine, rehabilitation medicine, and humanities and social sciences. It was established in the 1990s and covers various age, gender, organ system, and health problems. The main service area of ​​general medicine is primary health care, with family and community as the background, mainly dealing with common problems, and a large number of health problems in the undifferentiated stage of disease. General undifferentiated disease refers to medically unexplained somatic symptoms or disease that has not been clearly attributed to a system in the early ...

AI Technical Summary

Problems solved by technology

[0004] 1. Existing technical solutions seldom analyze different types or different systems of diseases that patients may suffer from based on symptoms, and most of them focus on one or one type of disease that has been identified, such as diabetes and autism , infectious diseases, etc., but in general practice clinics, doctors are often faced with undifferentiated diseases, and patients with the same symptoms may need to be referred to different departments for follow-up treatment, so the current technical solutions cannot solve the problem well Current problems encountered in the field of general practice;

[0005] 2. The existing technical solutions are mainly used in the guidance or pre-interrogation process before the patient and the doctor face-to-face consultation and the self-monitoring of the patient at home. They are not embedded in the real-time communication activities with the doctor and cannot be truly applied to clinical scenarios. middle;

[0006] 3. Existing technical solutions basically use data-driven methods to extract a large number of qualified patient cohorts, and then based on the clinical data of patients before diagnosis, use traditional machine learning or deep learning methods to analyze and assist doctors in clinical decision support , the whole process is not integrated into clinical guidelines, expert consensus, etc., and lacks knowledge drive;

[0007] 4. Existing technical solutions are often based on the big data of a certain group of people for model training and data analysis. After changing the group of people, the effect of the model will drop a lot, that is, the generalization performance of the model is poor. At the same time, there is a "black box" problem in deep learning. The model Not interpretable or poorly interpretable

Images