Automated left atrial function analysis using AI is a stronger predictor of survival than physician-measured left ventricular ejection fraction
Motivation: Left ventricular function is a strong predictor of survival, but atrial function is less well understood. Goal(s): Assessing atrial function is time-consuming and labour-intensive, requiring manual tracing of the atrial borders frame-by-frame. Approach: We developed an artificial intelligence tool capable of automating this process and can run on scanners in real time. Results: We found that the proportion of blood ejected by the left atrium with each beat (left atrial ejection fraction) strongly predicts survival in a large cohort of patients, beyond and incremental to similar established left ventricular measures. Impact: Atrial function measured automatically using inline AI is a strong and incremental predictor of patient survival. This enables new biomarkers to be easily translated into clinical workflow for improved patient care.
Authors
- James Moon (ORCID: https://orcid.org/0000-0001-8071-1491)
- Peter Kellman (ORCID: https://orcid.org/0000-0002-9875-6070)
- Kelvin Chow (ORCID: https://orcid.org/0000-0003-0698-1746)
- Charlotte Manisty (ORCID: https://orcid.org/0000-0003-0245-7090)
- Hui Xue (ORCID: https://orcid.org/0000-0002-4561-5530)
- Kavitha Vimalesvaran (ORCID: https://orcid.org/0000-0003-2236-7279)
- Sameer Zaman (ORCID: https://orcid.org/0000-0002-1467-8176)
- Rhodri Davies (ORCID: https://orcid.org/0000-0001-7630-7517)
- James P. Howard (ORCID: https://orcid.org/0000-0001-9730-4606)
- Hoi Ching Cheung (ORCID: https://orcid.org/0000-0001-7651-4903)
- Graham Cole (ORCID: https://orcid.org/0000-0003-4802-2869)
Publication Details
- Journal
- Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
- Published
- 2025-09-16
- DOI
- https://doi.org/10.58530/2025/4273
- Primary Topic
- Cardiovascular Function and Risk Factors
- Type
- article
- Field-Weighted Citation Impact
- 0.00