JMIR Publications

ABSTRACT

Background:

Large Language Models (LLMs) have fundamentally transformed approaches to Natural Language Processing (NLP) tasks across diverse domains. In healthcare, accurate and cost-efficient text classification is crucial for clinical notes analysis, diagnosis coding, and other tasks, where LLMs present promising potential. Text classification faces multiple challenges including manual annotation for training, handling imbalanced data, and developing scalable approaches. Healthcare adds additional challenges, particularly the critical need to preserve patients' data privacy and the complexity of medical terminology. Existing systematic reviews about LLMs either don't specialize in text classification or don't focus on the healthcare domain.

Objective:

This research synthesizes and critically evaluates the current evidence found in the literature regarding the use of LLMs for text classification in a healthcare setting.

Methods:

Major databases (e.g., Google Scholar, Scopus, PubMed, Science Direct) and other resources were queried, focusing on papers published between 2018 and 2024 within the framework of PRISMA guidelines. Studies were categorized by text classification type (e.g., binary classification, multi-label classification), application (e.g., clinical decision support, public health and opinion analysis), methodology, type of healthcare text, and metrics used for evaluation and validation.

Results:

The systematic review resulted in 65 eligible research articles that leveraged LLMs for automated healthcare text classification and contrasted results with existing machine learning-based methods in which embedding, annotation, and training are traditionally required.

Conclusions:

This review reveals existing gaps in the literature and suggests future research lines that can be investigated and explored regarding LLMs for healthcare text classification. Clinical Trial: N/A