Intelligent diagnosis method of electronic medical record based on sev-sdg-cnn

Abstract

The invention discloses an electronic medical record intelligent diagnosis method based on SEV-SDG-CNN, comprising: mining medical field vocabulary to construct a shallow semantic vocabulary dictionary; generating a shallow semantic vector of each word in a text corpus based on the shallow semantic vocabulary Express SEV; use the shallow semantic vocabulary dictionary to calculate the shallow semantic prediction mode SDG of each corpus; integrate SEV and SDG into the traditional CNN classification model construction, that is, splicing SEV and word embedding vector in the feature expression layer, and in the mode The output layer fuses SDG and CNN discriminative modes to achieve an improved SEV‑SDG‑CNN classification model. The invention is applied in the disease diagnosis of electronic medical records, can fully extract a large amount of rich semantic information contained in electronic medical records with strong professionalism and high knowledge density, and benefits from the advantage that CNN can automatically mine features from big data, so that The intelligent diagnosis method of electronic medical record has high performance.

Description

technical field [0001] The invention relates to the field of deep learning and intelligent diagnosis of electronic medical records, in particular to an intelligent diagnosis method of electronic medical records based on SEV-SDG-CNN. Background technique [0002] Electronic medical records have attracted the attention of the majority of researchers because of the integrity of data and the diversification of information. How to use the rapidly growing electronic medical record data to support biomedical research and clinical research is an important research content in medical informatics and translational medicine. In practice, doctors use the patient's chief complaint, current illness history, past history, family history, related examinations and other information to diagnose the patient's disease. However, the diagnostic accuracy of doctors often depends on individual medical knowledge and clinical experience. Different doctors may have different diagnoses for the same pa...

AI Technical Summary

Problems solved by technology

However, the diagnostic accuracy of doctors often depends on individual medical knowledge and clinical experience. Different doctors may have different diagnoses for the same patient, especially low-ranking doctors and doctors in remote areas have low diagnostic accuracy.

Images