Natural language generation in healthcare: A review of methods and applications

Clinical Snapshot

Key Findings

• P Value: Not applicable for descriptive systematic review

• Effect Size: Text-to-text generation dominant at 65.5% of applications

• Primary Outcome: Growth in healthcare NLG publications from 2 (2018) to 40 (2024) studies annually

• Nnt Or Sensitivity: Transformer encoder-decoder models used in 61.6% of studies; ROUGE evaluation in 81.4%

• Confidence Interval: Not applicable for descriptive systematic review

Clinical Application

Limited by identified challenges in factual consistency, explainability, and AI safety requiring further development before widespread clinical adoption Relevant to Australian healthcare digitization efforts, particularly for clinical documentation efficiency and radiology reporting, but requires TGA consideration for AI medical devices and alignment with RACGP digital health guidelines Healthcare organizations implementing clinical documentation, radiology reporting, discharge summaries, and medical dialogue systems

Abstract

OBJECTIVE: This study presents a systematic review of natural language generation (NLG) methods and applications in the medical domain, providing quantitative and qualitative analyses to answer four key research questions regarding methods, evaluation, applications, and challenges of NLG in healthcare. METHODS: We searched PubMed, ACM Digital Library, Web of Science, Science Direct, Scopus, Embase, and ACL Anthology for NLG-related studies in healthcare from 2018 to 2024. Out of 3,988 research articles, 113 met the inclusion criteria and were analyzed across data modality, model architecture, evaluation metrics, and application domain. RESULTS AND CONCLUSION: NLG in healthcare has grown substantially, with annual publications increasing from 2 in 2018 to 40 in 2024. Of the 113 included studies, text-to-text generation was the most common data modality (65.5%), followed by image-to-text (19.5%) and multimodal-to-text (15.0%). Transformer-based architectures were dominant, especially encoder-decoder models (61.6%). Automatic evaluation metrics such as ROUGE (81.4%) and BLEU (57.5%) were widely used. Human evaluation metrics, such as Likert scales (31.9%), were increasingly adopted. The four most prevalent application domains include summarization (e.g., discharge summaries, radiology reports), clinical documentation, medical dialogue, and data augmentation. The transformer-based large language models (LLMs) and the accumulation of large-scale multimodal clinical datasets have remarkably advanced NLG in healthcare. However, challenges remain in factual consistency, explainability, evaluation robustness, and AI safety. Addressing these challenges is essential for the adoption of NLG in various healthcare applications.

References

1. 1.Lyu, M., Li, X., Chen, Z., Pan, J., Peng, C., Talankar, S., & Wu, Y. (2026). Natural language generation in healthcare: A review of methods and applications. Journal of Biomedical Informatics. https://doi.org/10.21203/rs.3.rs-2566942/v1