# Recent Advances in AI and GenAI for Health Informatics

## TL;DR

This paper reviews recent AI and GenAI applications in health informatics, highlighting opportunities and challenges across five key healthcare areas.

## Contribution

The paper provides a broad, comprehensive review of AI in health informatics, covering five major topics and identifying common concerns and future directions.

## Key findings

- Clinical decision support, patient care, and electronic health records are the most mentioned AI applications in health informatics.
- Common concerns include patient privacy, cybersecurity, ethics, and lack of explainability in AI systems.
- The review emphasizes the need for benchmarks, standardization, and engaging health professionals in AI adoption.

## Abstract

The emergence of large language models (LLMs) and generative artificial intelligence (GenAI) has marked a turning point in health informatics. AI has become a very helpful tool for health informatics applications, with numerous AI applications in health informatics being reported in the last years. The objective of this paper is to synthesize the common concerns and opportunities raised by recent popular reviews on AI and health informatics. The main methodological topics covered in this up-to-date review include traditional AI, GenAI, and LLMs. The literature search was conducted through the popular academic database Scopus, which covers over one hundred million records, including both computer science and healthcare. Among these popular reviews (measured by the number of citations that each one received), clinical decision support, patient care, electronic health records, hospital management, and remote patient monitoring are the most mentioned healthcare topics. Different from the majority of the existing reviews that narrowly cover on one to a few topics in healthcare, our review is designed with the objective to provide a broad coverage, such that practitioners may benefit from comprehensive insights covering the above mentioned five popular topics in AI health informatics applications. Based on an in-depth analysis of these reviews by human experts, the main AI tools used, their main challenges, and some future directions have been identified in our investigation. Patient privacy, cybersecurity, ethics, clinical accountability, engaging health professionals, benchmarks and standardization, as well as lack of explainability are the common concerns identified from the literature covered in this review.