A large radiology practice found a generative AI model beneficial for chest x-ray worklist prioritization and quality assurance.
Key Details
- 1The AI model generated text-based clinical reports for 34,680 chest x-ray studies over two weeks.
- 2An NLP model mapped reports to 155 chest x-ray findings, comparing AI and radiologist results.
- 3Sensitivity and specificity for pneumothorax detection were 62.4% and 99.3%, respectively.
- 4Studies with positive pneumothorax findings were flagged for urgent review; 36 cases with discrepant findings went to secondary review.
- 525% of secondary-reviewed cases revealed missed pneumothorax by radiologists.
- 644% of radiologists rated AI-generated reports as equivalent in quality to their own.
Why It Matters
This demonstrates tangible workflow improvements and diagnostic support from generative AI in radiology, suggesting the potential to reduce radiologist workload and aid in addressing staffing shortages.
Source
AuntMinnie
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.
•Cardiovascular Business
Radiology Maintains Lead in FDA-Cleared AI Algorithms, Cardiology Follows
Radiology remains the top specialty for FDA-cleared AI, with cardiology as a strong second, particularly in cardiovascular imaging.
•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.