A study finds mammography acquisition settings influence both AI and radiologist performance in breast cancer detection.
Key Details
- 1Seven acquisition parameters studied: machine type, kVp, x-ray exposure, relative exposure, paddle size, compression force, tissue thickness.
- 2Dataset: 28,278 2D mammograms from 22,626 women; 324 with cancer diagnosis within a year.
- 3Radiologists: Sensitivity 79.3%, specificity 88.7%; AI: Sensitivity 76.9%, specificity 76.9%.
- 4Increased x-ray exposure reduced specificity for AI but not radiologists; increased compression reduced specificity for radiologists but not AI.
- 5Trends for kVp: little effect on sensitivity, slight increase in specificity for both AI and radiologists.
Why It Matters
Understanding how technical factors affect AI versus human interpretation is vital for optimizing AI deployment in mammography. These insights could inform guidelines, quality control, and trust calibration when integrating AI into clinical screening workflows.
Source
AuntMinnie
Related News
•HealthExec
AI's Expanding Role in Healthcare and Implications for Radiology
A series of thought leaders and institutions weigh in on AI's transformative potential in healthcare, with emphasis on radiology adoption and responsible use.
•Radiology Business
Most Radiology AI Users Lack Clear Evidence of Financial ROI
Survey finds over 75% of radiology organizations using AI lack clear, quantified ROI data.
•AuntMinnie
Toronto Study: LLMs Must Cite Sources for Radiology Decision Support
University of Toronto researchers found that large language models (LLMs) such as DeepSeek V3 and GPT-4o offer promising support for radiology decision-making in pancreatic cancer when their recommendations cite guideline sources.