Longitudinal Validation of a Deep Learning Index for Aortic Stenosis Progression
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Aims
Aortic stenosis (AS) is a progressive disease requiring timely monitoring and intervention. Transthoracic echocardiography (TTE) remains the gold standard for AS assessment, while deep learning (DL)-based approaches have emerged as promising tools for enhanced evaluation. This study examined the longitudinal changes in a previously developed DL-derived index for AS continuum (DLi-ASc; scaled 0 to 100) and its association with AS progression. Additionally, we evaluated its utility in identifying high-risk patients prone to progression to severe AS.
Methods and Results
A total of 2,563 patients with follow-up TTE (8,053 TTEs) from two tertiary hospitals were included. DLi-ASc, generated from limited TTE views, was tracked over time. The longitudinal trends in AS severity showed a progressive shift toward advanced stages, with a corresponding increase in DLi-ASc (p for trend <0.001). DLi-ASc demonstrated positive correlations with conventional AS parameters, including AV maximal velocity (V max ) (Pearson correlation coefficients [PCC] = 0.68) and mean pressure gradient (mPG) (PCC = 0.69). Higher baseline DLi-ASc was associated with a faster AS progression rate (p for trend <0.001). Furthermore, baseline DLi-ASc was an independent predictor of progression to severe AS, with hazard ratios of 2.92, 2.45, and 2.67 per 10-score increase within 1-, 2-, and 3-year follow-up windows, respectively. A DLi-ASc threshold of 50 was identified as a threshold for rapid progression to severe AS.
Conclusion
DLi-ASc increased in parallel with AS progression and independently predicted severe AS progression. These findings support its role as a non-invasive imaging-based digital marker for longitudinal AS monitoring and risk stratification.
