A custom large language model significantly improved the identification of patients needing follow-up imaging by analyzing radiologists’ notes.
Key Details
- 1A new LLM-based tool was developed at Parkland Health to flag patients who require follow-up imaging.
- 2Traditional EHR macros and structured notes failed to efficiently capture needed follow-up recommendations.
- 3The AI model reads clinical impressions from radiologist notes to extract and standardize follow-up indications.
- 4More than 500,000 radiology studies are performed annually at the health system, emphasizing the scale and need for automation.
- 5Integration into the EHR enables real-time flagging and streamlined workflow for ensuring follow-up imaging.
Why It Matters
Missed follow-ups are a key source of diagnostic errors in radiology. This LLM-driven tool helps close communication gaps, ensuring at-risk patients receive necessary imaging, which could improve outcomes and reduce diagnostic delays.
Source
Radiology Business
Related News
•AuntMinnie
Study: Computer Vision Models Best LLMs in Chest CT Breast Abnormality Detection
Computer vision models (CVMs) surpass large language models (LLMs) in accurately labeling incidental breast abnormalities on chest CT scans.
•AuntMinnie
Deep Learning Models Rival Radiologists for Pancreatic Cancer Detection on CT
Deep-learning models achieved comparable or superior accuracy to experienced radiologists in detecting pancreatic cancer on CT scans, especially for small tumors.
•Radiology Business
Radiology AI Devices at Elevated Risk for FDA Recalls, Study Finds
Radiology AI devices are more likely to face FDA recalls, largely due to deviations from intended use and incomplete clinical data.