Finely tuned, pre-trained large language models are beginning to reliably translate image content into text, but are they ready to take on medical images? 

The combination of AI and CT helped Nehal Mehta, MD, and colleagues track the performance of a new drug designed to target coronary inflammation. 

Machine learning models could help create a more standardized, reproducible and efficient way of grading Crohn’s disease severity in the small bowel based on CT imaging. 

The model was developed and validated specifically for liver, spleen and pancreas segmentation, and outperformed a publicly available segmentation model already in use.

Although these tools have proven themselves valuable in numerous settings, they must be used with caution, especially by patients and nonradiologist providers who may be seeking clarification on imaging reports. 

Image-based questions were significantly more challenging for the large language model to answer, despite the latest version now being capable of accepting image prompts.

Automated AI-generated measurements combined with annotated CT images can improve treatment planning and help referring physicians and patients better understand their disease, explained Sarah Jane Rinehart, MD, director of cardiac imaging with Charleston Area Medical Center.

Finding such discrepancies is critical to the continuity of patient care, as medical records and reports are often utilized across multiple providers and facilities.