AI-powered image enhancement significantly boosts the diagnostic quality of suboptimal chest CT and CTPA studies.
Key Details
- 1Researchers applied a CNN-based AI to 611 chest CT and CTPA exams (both optimal and suboptimal).
- 2318 suboptimal cases (identified by thoracic radiologists) showed marked image quality improvement after AI enhancement.
- 3Vascular attenuation (HU) increased significantly: Pulmonary artery from 192 to 293 HU, ascending aorta from 235 to 362 HU, segmentals from 187 to 282 HU (all p = 0.001).
- 4Noise reduction was notable: Main pulmonary artery noise dropped from 17.6 HU to 11.2 HU.
- 5Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) tripled post-AI (main pulmonary artery SNR from 20.8 to 59.7, CNR from 188.8 to 288.8).
- 6The work was presented at RSNA 2025 and conducted by Massachusetts General Hospital & Harvard Medical School researchers.
Why It Matters
AI image enhancement could reduce unnecessary repeat CT studies, lower patient radiation/contrast exposure, and support more confident diagnoses, especially in challenging cases with suboptimal image quality. This work may pave the way for integrating enhancement and diagnostic AI in standard radiology workflows.
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.