The Mirai AI model significantly improves detection of interval breast cancers in negative screening mammograms.
Key Details
- 1Mirai risk model analyzed 134,217 screening mammograms, including 524 interval cancers.
- 2Top 20% risk group by Mirai captured 42.4% of interval cancers, corresponding to 1.7 additional detections per 1,000 exams.
- 3AUC values for interval cancer prediction ranged from 0.67 to 0.72 across time, age, and breast density subgroups.
- 4No significant performance variation across different age groups or breast densities was observed.
- 5Mirai has been validated on nearly 2 million mammograms across 21 countries.
- 6Editorial comments highlight progress but note limitations since interval cancer detection did not surpass 50%.
Why It Matters
Interval cancers are a challenge in breast screening, often missed during routine exams and harder to detect. AI-driven risk models like Mirai offer a promising path to personalized screening strategies, potentially catching cancers earlier and improving patient outcomes.
Source
AuntMinnie
Related News
•Radiology Business
AI Model Reduces Need for Gadolinium Contrast in Brain Tumor MRI
UCL researchers develop an AI tool to predict brain tumor MRI enhancement, reducing the need for gadolinium contrast agents.
•Radiology Business
Rad Partners Wins $1.29M FDA Grant for AI Radiology Report Evaluation Study
Cognita Imaging, part of Rad Partners, has received a $1.29M FDA grant to develop a new framework using large language models (LLMs) to evaluate AI-generated radiology reports.
•Radiology Business
Real-World Study: Radiology AI Best in Emergency and Inpatient Settings
A commercial AI tool for intracranial aneurysm detection outperformed in inpatient and emergency settings but yielded limited benefits for outpatients in a major health system study.