Deep Learning-Based Chatbots for Patient Queries
Priya Vijay, K. Jayashree, R. M. Suresh Babu, K. Vijay · 2023
Smart and intelligent healthcare systems are thriving now by integrating machine learning or deep learning algorithms of the IT industry to impact day-to-day vast medical needs. Over the last few decades, chatbots, or automatic FAQ answering machines, have become quite popular for answering customer queries without any human intervention. In recent years, the Covid pandemic situation made chatbots still more valuable, since people were not able to move around freely. The chatbot is an e-answering system that uses natural language processing algorithms which automatically understand the questions of the customer and answer them accordingly. Nevertheless, the chatbot with only a text answering system may not be sufficient for all diseases, which need complex dialogue and response management with images or videos. For some diseases, chatbots need the support of X-ray analysis or patient scan reports to diagnose and answer patient queries. Deep learning techniques have a wide range of algorithms to support computer vision-grounded image and video analytics pattern-matching and analysis. The scan reports or X-ray images are analyzed with deep learning techniques and the question of the severity of disease or precautions or medications to be taken can be answered automatically by integrating chatbots with deep learning techniques. This chapter discusses the role of various deep learning algorithms to train and test models of X-ray and scan reports and integrate with medical chatbots to provide resourceful answers for appropriate remedial care and accurate diagnosis of the report to prescribe drugs or precautions as a response to patient queries. It also deliberates on future challenges, limitations, regulatory standard issues, ethical problems, security glitches, and the scope of research in the field of deep learning chatbots.