Research Challenges in Deep Generative Models for Healthcare and Medical Applications

Kumar, Himanshu, Aruldoss, Martin, Wynn, Martin G ORCID: https://orcid.org/0000-0001-7619-6079 and Lakshmi, Miranda (2025) Research Challenges in Deep Generative Models for Healthcare and Medical Applications. In: Advances in Deep Generative Models for Healthcare and Medical Applications. Routledge/Taylor & Francis Group, Boca Raton, pp. 1-21. ISBN 9781003602088

Text 15668 Himanshu, K et al. (2025) Research Challenges in Deep Generative Models for Healthcare and Medical Applications.pdf - Accepted Version Restricted to Repository staff only until 13 June 2027. Available under License Creative Commons Attribution Non-commercial No Derivatives 4.0. Download (476kB)

Abstract

Deep generative models are a developing branch of generative artificial intelligence (AI), which is growing rapidly across various sectors due to current advancements. Its application has now extended to the medical and healthcare domains. Deep generative model technology is bringing new possibilities in medical and healthcare applications. This technology is driving revolutionary advancements across a wide range of medical facilities, specializations, and innovative research. Deep generative models and related tools aid in generating and supporting medical and healthcare applications by facilitating innovative research, in medical reporting and analysis, medical images, disease and risk prediction, and associated drug discovery. Nevertheless, these models face serious obstacles in their implementation in real time applications, including challenges related to accuracy, cost-effectiveness, privacy, security, and authentication, as well as ethical concerns. This chapter provides a detailed discussion of the challenges involved in building the robust deep generative models, focusing on aspects such as handling the clinical trials, clinical relevance, working with heterogeneous datasets, addressing dataset scarcity, and overcoming biological complexity.

Item Type:	Book Section
Uncontrolled Keywords:	Artificial intelligence; Deep generative models; Health record; Generative AI; Healthcare; Medical image Dataset; Medical applications; Patient privacy; Research challenges
Related URLs:	https://www.routledge.com/Advances-in-De...
Subjects:	T Technology > T Technology (General)
Divisions:	Schools and Research Institutes > School of Business, Computing and Social Sciences
Depositing User:	Martin Wynn
Date Deposited:	16 Dec 2025 13:17
Last Modified:	16 Dec 2025 13:30
URI:	https://eprints.glos.ac.uk/id/eprint/15668

University Staff: Request a correction | Repository Editors: Update this record