Prompt mechanisms in medical imaging: A comprehensive survey.
Authors
Affiliations (9)
Affiliations (9)
- Faculty of Applied Sciences, Macao Polytechnic University, Rua de Luis Gonzaga Gomes, Macao 999078, China.
- Department of Radiology, the Netherlands Cancer Institute, Plesmanlaan 121, Amsterdam 1066 CX, the Netherlands.
- Department of Radiology and Nuclear Medicine, Radboud University Medical Centre, Geert Grooteplein 10, Nijmegen 6525 GA, the Netherlands.
- College of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073, China.
- Medical Department of Breast Cancer, Hunan Cancer Hospital, Changsha 410013, China.
- Medical Department of Breast Cancer, the Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha 410013, China.
- Image-guided Surgery, Surgical Department, the Netherlands Cancer Institute, Plesmanlaan 121, Amsterdam 1066 CX, the Netherlands.
- Institute of Biomedical Engineering, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo 315300, China.
- Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China.
Abstract
Deep learning offers transformative potential in medical imaging, yet its clinical adoption is frequently hampered by challenges such as data scarcity, distribution shifts, and the need for robust task generalization. Prompt-based methodologies have emerged as a pivotal strategy to guide deep learning models, providing flexible, domain-specific adaptations that significantly enhance model performance and adaptability without extensive retraining. This systematic review critically examines the burgeoning landscape of prompt engineering in medical imaging. We dissect diverse prompt modalities, including textual instructions, visual prompts, and learnable embeddings, and analyze their integration for core tasks such as image generation, segmentation, and classification. Our survey shows that prompt mechanisms advance medical AI on two fronts. At a performance level, they enhance accuracy, robustness, and data efficiency. Methodologically, they circumvent the need for manual feature engineering. This model guidance has the potential to enhance the interpretability of model behavior by making task guidance more explicit. Despite substantial advancements, we identify persistent challenges, particularly in prompt design optimization, data heterogeneity, and ensuring scalability for clinical deployment. Finally, this review outlines promising future trajectories, including advanced multimodal prompting and robust clinical integration, underscoring the critical role of prompt-driven AI in accelerating the revolution of diagnostics and personalized treatment planning in medicine.