Fundus Image-based Visual Acuity Assessment with PAC-Guarantees
This work addresses the need for robust and reliable visual acuity assessment in eye health management, offering a solution that could reduce clinician workload, though it is incremental as it applies existing PAC techniques to a new task.
The paper tackles the problem of predicting visual acuity from fundus images by developing a method that provides prediction intervals with PAC guarantees, ensuring reliability despite model uncertainties. Experimental results show that the method maintains these guarantees and performs comparably or better than prior works without such assurances.
Timely detection and treatment are essential for maintaining eye health. Visual acuity (VA), which measures the clarity of vision at a distance, is a crucial metric for managing eye health. Machine learning (ML) techniques have been introduced to assist in VA measurement, potentially alleviating clinicians' workloads. However, the inherent uncertainties in ML models make relying solely on them for VA prediction less than ideal. The VA prediction task involves multiple sources of uncertainty, requiring more robust approaches. A promising method is to build prediction sets or intervals rather than point estimates, offering coverage guarantees through techniques like conformal prediction and Probably Approximately Correct (PAC) prediction sets. Despite the potential, to date, these approaches have not been applied to the VA prediction task.To address this, we propose a method for deriving prediction intervals for estimating visual acuity from fundus images with a PAC guarantee. Our experimental results demonstrate that the PAC guarantees are upheld, with performance comparable to or better than that of two prior works that do not provide such guarantees.