Abstract

Patient feedback plays a crucial role in improving the quality, responsiveness, and patient-centric approach
of healthcare services. This paper presents a comprehensive review of both traditional and digital methods
used to collect patient feedback, emphasizing their value in improving healthcare delivery, examines the
tools and channels used, including surveys, interviews, and multichannel digital platforms. The review
further explores sentiment analysis techniques applied to patient feedback, focusing on how machine
learning, deep learning, and large language models are used to interpret and categorize unstructured text.
The recent literature is systematically analyzed, with comparative tables that highlight feature extraction
methods, classification algorithms, and performance metrics reported in various studies. Additionally, the
paper addresses key challenges in feedback collection and sentiment analysis. Future research directions
are proposed, such as automating feedback systems and incorporating patient perspectives into quality
improvement frameworks. This review is intended to assist Healthcare IT Professionals and medical
Data Scientists who deal with healthcare delivery and computational analysis, whose target is to extract
actionable insights from patient feedback using modern AI techniques.

Key words: Patient Feedback, Sentiment Analysis, Lexicon, Machine Learning, Deep Learning, Generative AI