Disease classification intelligent service method based on patient symptom description text

Abstract

The invention provides a disease classification intelligent service method based on a patient symptom description text, and belongs to the field of artificial intelligence, and the method specifically comprises the steps: firstly, extracting an entity from existing data according to a description statement of a patient symptom, marking a type, and storing the entity as a sample in a graphic database; then, through a generator, new samples of the same type of each sample are generated and marked, and for each new sample, an inspector is utilized to judge that an available sample is added into a marked data set; sequentially constructing entity identification micro-services, inputting a symptom description text of an actual patient, and outputting an entity identification result; constructing a disease query micro-service, querying diseases related to the identification result of the entity in a graphic database, sorting the diseases, and outputting the diseases as candidate disease results; and finally, constructing a user interaction micro-service, and packaging and returning an entity recognition result, a candidate disease result and historical health data to the patient. The operation and maintenance cost is reduced, and subsequent updating and data perfection are facilitated.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a disease classification intelligent service method based on patient symptom description text. Background technique [0002] With the intensification of population aging in my country, more and more people pay more attention to their own health; however, the allocation of medical resources in my country is not balanced, and many patients usually need to spend a lot of money in order to determine their own diseases. [0003] At present, most doctors determine the possible diseases of the patient by asking about the clinical symptoms of the patient, and combining professional knowledge, clinical experience and medical device diagnosis; when the doctor in the department is not good at the field of clinical symptoms, the patient needs Re-link the number of other departments or hospitals. Therefore, automatically inferring the patient's disease based on the text of the symptom ...

AI Technical Summary

Problems solved by technology

However, training a deep neural network model requires a large amount of labeled data, which takes a lot of time and manpower

[0008] A new way of thinking is to gradually accumulate data in practical applications, and gradually iteratively upgrade the model. Because of the high degree of coupling between the model, the knowledge graph, and the data set, the iterative deployment and operation costs are high, which hinders its practical application.

Images