SUPER-LARGE-SCALE DATA ANALYSIS FOR ELECTRONIC HEALTH RECORD WITH ECML

Abstract

With the deepening of hospital informatization construction, the electronic health record (EHR) system has been widely used in the clinical diagnosis and treatment process, resulting in a large amount of medical data. Electronic medical records contain a large amount of rich medical information, which is an important resource for disease prediction, personalized information recommendation, and drug mining. However, the medical information contained in electronic medical records cannot be automatically acquired, analyzed and utilized by computers. In this paper, we utilize machine learning algorithms for intelligent analysis of large-scale electronic medical records to explore and develop general methods and tools suitable for electronic medical record analysis in medical databases. This is of great value for summarizing the therapeutic effects of various diagnosis and treatment programs, disease diagnosis, treatment, and medical research. We propose an ECML-based intelligent analysis method for electronic medical records. First, we perform data preprocessing on the electronic medical record. Second, we design an intelligent analysis method for electronic medical records based on a deep learning model. Third, we design a model hyperparameter optimization method based on evolutionary algorithms. Finally, we compare and analyze the performance of the proposed model through experiments, and the experimental results show that the model proposed in this paper has good performance.