Research Area:  Machine Learning

Abstract:

Clinical named entity recognition plays an important role in the field of clinical research based on clinical information mining. The objective of clinical named entity recognition is to analyze and categorize medical conditions, namely symptoms, treatments, diseases, and body conditions in the Electronic Medical Records (EMRs). In recent years, deep neural networks have gained considerable achievement in several languages handling tasks and named entity recognition. Many algorithms are trained to learn the text features from the big scale labeled datasets. However, these data-driven techniques do not handle rare and unseen cases. Most of the existing methods have shown that human knowledge offered important information for managing rare and unobserved entities. However, there exist many issues in the medical records based on the clinical named entity recognition because of the various natural language text features and the exceptional clinical conditions in EMRs. Therefore, it is necessary to enhance the natural language text features of the model. Hence, this survey analyzes various methods of clinical named entity recognition. The main aim of this survey is to study the existing clinical named entity recognition techniques and classifies them under various categories. Accordingly, this paper gives a detailed survey of 25 research papers and classifies them under different categories, such as machine-learning-based methods, deep-learning-based methods, namely Neural Networks (NN), Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and active learning methods. Also, the analysis is promoted in the survey based on the publication year, research techniques, performance measures, and achievement of the research methodologies. Moreover, the problems in the methods are explained in the research gaps and issues. Furthermore, the future extent of this research work is provided based on the limitations identified from the existing research methods.

Keywords:  
Clinical
Named Entity Recognition
Electronic Medical Records (EMRs)
Deep Learning
Machine Learning

Author(s) Name:  Naveen S. Pagad & N. Pradeep

Publisher name:  Springer

DOI:  10.1007/978-981-16-2597-8_13

Volume Information:  pp 151–165

Paper Link:   https://link.springer.com/chapter/10.1007/978-981-16-2597-8_13