A deep learning model using echocardiography accurately detects cardiac amyloidosis, outperforming traditional methods.
Key Details
- 1AI model trained on echocardiography video clips from 2,612 patients across multiple sites and ethnic groups.
- 2External validation performed on 18 global sites; included 597 amyloidosis cases and 2,122 controls.
- 3Achieved AUROC of 0.93 (after excluding 13% uncertain predictions), sensitivity of 85%, specificity of 93%.
- 4Performance was consistent across amyloidosis subtypes and various subgroups.
- 5The AI model outperformed transthyretin cardiac amyloidosis (TTR-CA) score (AUROC = 0.73) and wall thickness scoring (AUROC = 0.8).
- 6Ultromics employees contributed and funded the study.
Why It Matters
This work demonstrates the robust, generalizable performance of AI to enhance echocardiographic detection of cardiac amyloidosis, surpassing current clinical scoring tools and suggesting real-world impact for earlier and more accurate diagnosis in routine practice.
Source
AuntMinnie
Related News
•Radiology Business
Meta Funds Ohio State’s Radiology AI Innovation Projects
Meta will fund Ohio State researchers to explore AI-driven advancements in radiology, including chest X-ray reporting and disease prediction.
•AuntMinnie
ChatGPT-5 Performs Poorly at Detecting Distal Radius Fractures on X-rays
ChatGPT-5 missed about half of distal radius fractures on x-rays, showing limited diagnostic capability in a real-world scenario.
•AuntMinnie
AI Triage Tool Speeds Up ICH Head CT Turnaround in ED
AI-based triage for intracranial hemorrhage on head CT reduced emergency department report turnaround by 13 minutes.