Reem AlAhmadi - research works

Data-centric optimisation significantly influences DL performance for UM detection.

Three principles emerge: healthy class addition improves specificity and anatomical feature learning; data quality may outweigh quantity; and contextual input tuning is a key model parameter.

These findings offer a practical framework for developing clinically robust, physician-supervised AI tools to support early UM triage and reduce diagnostic variability.

A standard, operationalized segmentation protocol enables reproducible boundary delineation on ophthalmic ultrasound and provides a foundation for consistent tumor measurement and future computational automation.

In this study, early retreatment was low (8.8%), with most eyes initially treated with laser (21.5%) rather than bevacizumab (1.7%).

Aggressive ROP was associated with earlier retreatment, highlighting its role as a marker of more severe disease.

Compared to laser, bevacizumab was associated with lower overall and early retreatment, and delayed need for additional intervention when necessary.