Multimodal Foundation Models in Radiology

Artificial intelligence has entered a transformative era with the advent of large foundation models, including large language models (LLMs), large vision models (LVMs), and large multimodal models (LMMs). These models have demonstrated exceptional capabilities in enhancing natural language understanding, computer vision, and multimodal tasks. Radiology, a field facing increasing demands due to the growing volume and complexity of medical imaging, stands to benefit significantly from these advancements. Foundation models offer the possibility to aid radiologists in triaging cases, generating accurate reports, and identifying potential abnormalities. Establishing effective, radiology-specific models could revolutionize diagnostic processes and improve overall clinical outcomes.

This Research Topic aims to facilitate the development and application of foundation models in the radiological domain. The goal is to present innovative research exploring the adaptation and implementation of LLMs, LVMs, and LMMs in radiology. By focusing on the specific challenges inherent to radiology, such as the sheer volume of imaging data and task complexity, this research topic seeks to advance our understanding of how these models can enhance clinical workflows, improve diagnostic precision, and ultimately elevate patient care standards. Contributions that display the potential of these technologies to successfully integrate into radiological practices are especially encouraged.

To gather further insights into the integration and impact of foundation models in radiology, we welcome articles addressing, but not limited to, the following themes:

• New neural network architecture design for radiology-specific foundation models.
• Training, fine-tuning, and adaptation methods tailored for radiology.
• Algorithms for multimodal alignment within radiology-specific models.
• Paradigms for incorporating human feedback into model training.
• Optimization, acceleration, and compression of radiology-specific models.
• Frameworks for integrating multiple models in radiology applications.
• Privacy-preserving training techniques for radiology models.
• Interpretability and explainability considerations for foundation models.
• Trustworthiness, including ethical considerations and bias mitigation.
• Evaluation metrics, open datasets, and benchmarking for radiological applications.

In addition to original research articles, we invite systematic reviews that shed light on the latest advancements, challenges, and potential directions in the field of radiology-specific foundation models, appealing to researchers working at the nexus of AI, machine learning, and applied radiology.

Article types and fees

This Research Topic accepts the following article types, unless otherwise specified in the Research Topic description:

• Brief Research Report
• Case Report
• Data Report
• Editorial
• FAIR² Data
• FAIR² DATA Direct Submission
• General Commentary
• Hypothesis and Theory
• Methods
• ... View all formats

Articles that are accepted for publication by our external editors following rigorous peer review incur a publishing fee charged to Authors, institutions, or funders.

Article types

This Research Topic accepts the following article types, unless otherwise specified in the Research Topic description:

• Brief Research Report
• Case Report
• Data Report
• Editorial
• FAIR² Data
• FAIR² DATA Direct Submission
• General Commentary
• Hypothesis and Theory
• Methods
• Mini Review
• Opinion
• Original Research
• Perspective
• Policy and Practice Reviews
• Policy Brief
• Review
• Systematic Review

Keywords: foundation models, radiology, radiology-specific foundation models, the latest breakthroughs, challenges, future directions

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.