HealthCare Text Analytics Using Recent ML Techniques

Abstract

A natural language processing (NLP) technique known as automatic clinical text classification opens up information contained in clinical descriptions. By gaining knowledge of clinical outcomes recorded in the medical literature, classification of clinical texts has a significant impact on disease diagnosis, medical research, and automated development of disease ontologies. Because they contain terms that describe medical concepts and terminology, clinical texts are difficult to classify. The process of extracting information from clinical descriptions is called clinical text classification. An audio file, lecture notes, or any other spoken word provided by a physician is a clinical narrative. The clinical narrative, which provides a personalized account of the patient’s history and evaluations as well as extensive data for clinical decision-making, is the main form of communication in the medical field. The development of computer technology has allowed data mining and machine learning (ML) technologies to grow dramatically in recent years and several applications for text mining and information extraction (IE) have emerged as a result of the rapid growth of healthcare data. NLP has become a hot topic in AI research and applications as machine learning and deep learning algorithms have advanced because text, such as English sentences, is a significant type of natural language data. The aim of the study is to make disease diagnoses based on medical records using ML algorithms. The proposed clinical text classification model uses weak monitoring to develop the application of ML models to clinical text classification and reduces the amount of human effort required to create labeled training data and feature engineering. This categorization may be used in various initiatives when examining the clinical text data on diagnosis and medical procedures. The primary objective of this paper is to contrast a logistic multi-class regression model with a support vector classifier model used to classify medical records’ clinical text.