An AI algorithm helped detect previously missed intracranial aneurysms in routine contrast-enhanced neck CT scans.
Key Details
- 1Researchers tested the Aidoc AI aneurysm detection algorithm as a secondary reviewer on 631 routine neck CT exams.
- 2Routine neck CTs often capture portions of cerebral vasculature incidentally.
- 3The AI flagged three cases: two subtle aneurysms (<2 mm) missed by initial radiologist interpretation, and one 4.5-5 mm aneurysm already detected by the radiologist.
- 4AI prompts led to addenda in reports for findings not previously identified.
- 5The algorithm was not tested on dedicated CT angiography studies.
Why It Matters
Perceptual errors can result in missed diagnosis of incidental, clinically significant aneurysms on studies not primarily focused on the brain. Deploying AI as an additional safety net could improve detection rates and reporting quality in routine radiology workflow.
Source
AuntMinnie
Related News
•AuntMinnie
Deep Learning on PSMA-PET Outperforms Tools in Prostate Cancer Survival Prediction
Deep learning models using PSMA-PET scans predict survival in metastatic prostate cancer more accurately than established clinical tools.
•AuntMinnie
AI Boosts Sensitivity but Raises False Alarms in Microbleed Detection
AI significantly increases sensitivity for detecting cerebral microbleeds on brain MRI but at the cost of more false positives, highlighting the complementary strengths of AI and radiologists.
•AuntMinnie
AI on Chest X-Rays Matches PREVENT Score for Cardiac Risk
Deep learning models analyzing chest x-rays can predict cardiovascular risk on par with the PREVENT clinical score.