Researchers have developed AMIR-GPT, an AI model trained to recommend appropriate imaging exams per clinical guidelines.
Key Details
- 1AMIR-GPT is a generative pre-trained transformer similar to ChatGPT.
- 2It was trained using over 1,000 question-and-answer pairs from 26 American College of Radiology Appropriateness Criteria guidelines.
- 3The model assists providers in selecting the most suitable imaging exam for specific clinical scenarios.
- 4AMIR-GPT's performance was benchmarked against GPT-3.5, GPT-4, and Gemini using a curated test set.
- 5Responses were evaluated with a 1–5 scale and weighted Cohen’s kappa for agreement.
Why It Matters
Addressing imaging overutilization is crucial for reducing unnecessary costs and aligning clinical practice with evidence-based guidelines. This domain-specific AI could improve decision-making and efficiency in radiologic ordering, benefiting both practitioners and patients.
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.