Prompt engineering for healthcare: Methodologies and applications
Jiaqi Wang, Enze Shi, Sigang Yu, Zihao Wu, Huawen Hu, Chong Ma et al. · 2025 · 137 citationsRead the paper
Prompt engineering is a critical technique in the field of natural language processing (NLP) that involves designing and optimizing the prompts used to input information into models, aiming to enhance their performance on specific tasks. With the recent advancements in large language models (LLM), prompt engineering has shown significant superiority across various domains and has become increasingly important in the healthcare domain. However, there is a lack of comprehensive reviews specifically focusing on prompt engineering in the medical field. This review will introduce the latest advances in prompt engineering in the field of NLP for the medical field. First, we will provide the development of prompt engineering and emphasize its significant contributions to healthcare NLP applications such as question-answering systems, text summarization, and machine translation. With the continuous improvement of general LLM, the importance of prompt engineering in the healthcare domain is becoming increasingly prominent. The aim of this article is to provide useful resources and bridges for healthcare NLP researchers to better explore the application of prompt engineering in this field. We hope that this review can provide new ideas and inspire for research and application in medical NLP.
What the full text shows
Working prototype- Continuous prompting (soft prompting) serves as an effective alternative to standard model fine-tuning in scenarios with highly imbalanced data. (Downstream tasks with highly imbalanced data)
- PromptFuse and BlindPrompt align different modalities in a parameter-efficient manner, performing comparably to multi-modal fusion methods. (Low-resource scenarios)
- Adaptive PromptNet achieves competitive glioma grading performance using only non-enhanced MRI data by utilizing enhanced features as prompts during training. (Glioma grading on NE-MRI data)
- CPP improves class separation and outperforms existing methods under a light memory budget using task-specific prompts optimized with contrastive prototypical loss. (Task-specific information training)
- PrefixMol allows for more precise and efficient control over output compared to traditional generative models by using a learnable prompt token as prefix embedding. (Targeted pocket circumstances and properties)
- Setup
- Review of existing literature and specific model implementations (e.g., PromptFuse, BlindPrompt, PrefixMol, PromptNet, CPP).
- How
- The paper is a comprehensive review that analyzes various prompt engineering methodologies, including discrete (manual and automated) and continuous (soft) prompting. It evaluates these techniques across medical NLP tasks such as classification, generation, detection, and data augmentation.
- Limitations
- Effectiveness of automated prompts depends heavily on search algorithm quality and model representation power. · Manual prompt design is time-consuming and costly.
Summarized by Gemma 4 from the open-access full text · 13 Oct 2026. Read it