Advancing medical imaging with language models: featuring a spotlight on ChatGPT

Area of Science:

• Artificial Intelligence
• Medical Imaging Analysis
• Natural Language Processing

Background:

• Language models (LMs) have evolved significantly, with large language models (LLMs) showing promise in various research domains.
• Medical imaging generates vast amounts of data, necessitating advanced analytical tools.

Purpose of the Study:

• To provide a comprehensive guide for implementing language models in medical imaging research.
• To review current applications and future potential of LLMs in medical imaging.

Main Methods:

• Literature review of existing research on language models in medical imaging.
• Detailed examination of LLM capabilities, including specific examples like ChatGPT.
• Discussion of the impact of LLMs on clinical workflow and diagnostic accuracy.

Main Results:

• LLMs are applied in medical imaging for tasks like image captioning, report generation, and classification.
• Applications include findings extraction, visual question answering, and interpretable diagnosis.
• LLMs offer significant advantages, including enhanced clinical efficiency and reduced diagnostic errors.

Conclusions:

• Effective integration of LLMs with medical imaging can inspire new research ideas and innovations.
• This review serves as a valuable resource for researchers exploring LLM applications in medical imaging.
• Continued investigation into LLMs holds potential for advancing medical imaging analysis and patient care.