Medical Text and Image Processing: Applications, Issues and Challenges

Abstract

AbstractText and image analysis are playing very important role in healthcare and medical domain. The whole clinical process is getting affected positively by text and image processing. Many datasets, algorithms, models and tools are available for extracting useful information and for applying natural language processing, machine learning and deep learning algorithms. But there exist many challenges in healthcare data for successful implementation of text and image based machine learning models, which include: (i) storage and retrieval of high resolution images, (ii) scarcity of data (iii) dataset generation and validation, (iv) appropriate algorithms and models for extracting hidden information from images and texts, (v) use of modern concepts like deep neural networks, recurrent neural network, (vi) data wrangling and (vii) processing capacity of processors. This chapter will: (i) establish a background for the research work in the area of machine learning and deep learning, (ii) provide the brief about various types of medical texts and images, (iii) discuss the various existing models and applications of natural language processing, machine learning, computer vision and deep learning in medical domain and (iv) discuss various issues and challenges on applying natural language processing, machine learning and deep learning on medical data.KeywordsMachine learningDeep learningMedical imagingInformation extractionMedical textElectronic health records