Using Natural Language Processing in Electronic Health Records to Develop Computer Modules for Deep Learning

Natural Language Processing (NLP) is a theory-driven computational technique thatautomatically analyzes and represents human language, related to computer science, artificialintelligence, and computational linguistics. Currently, NLP is mainly used for text screening, voicerecognition, sentiment analysis, and interpretation detection. Since NLP is operated in computer andhuman natural language, NLP is widely used in artificial intelligence and machine learning. In 2009," Health Information Technology for Economic and Clinical Health Act " was passed in the UnitedStates to encourage medical institutions to apply Electronic Health Records (EHR), revealing the needfor collaboration between medicine, clinical informatics, and computer science. EHR contain richlongitudinal data on large populations and can be linked to contextual data in complex networks ofcausation. Such complex data can be parsed into structured data through NLP, and the correlation inmedical record data can be found from it. Therefore, this project will use the latest natural TheSentence-BERT module for language processing uses sentiment analysis classifiers and sentencesimilarity calculation models trained on actual clinical EHR text reports (free-text clinical diagnosis notes) to automatically label and analyze various medical records. To facilitate the development of"medical service artificial intelligence system".