DOI 10.1109/embc58623.2025.11254657
ReLAC: A Transformation Methodology for Automated Optic Disc Tilt and Torsion Correction and Circularity Improvement in Fundus Images.
Аннотация
The Retinal Landmark Alignment and Correction (ReLAC) algorithm is a novel method to mitigate confounding optic disc tilt and torsion in fundus image analysis. It utilizes automated optic disc segmentation and refined fovea localization based on pixel-wise distance regression. ReLAC quantitatively assesses optic disc tilt using the Ovality Index (OI) and rotation angle. Subsequently, a polar coordinate-based realignment algorithm rectifies these distortions to standardize image orientation. Validation on the MyoCUTU-DATA dataset confirmed ReLAC enhances optic disc circularity across varying myopia categories. Following realignment, mean OI consistently decreased across all groups from 1.138 to 1.083 in High Myopia, 1.092 to 1.058 in Myopia, and 1.087 to 1.043 in Non-myopia cohorts. Consequently, the proportion of images falling into the near-circular OI range of 1.0 to 1.1 increased by 36%, while the proportion in the highly elliptical OI range above 1.2 decreased by 80%. ReLAC provides a robust method for refining fundus image analysis by ensuring a more consistently circular optic nerve head representation. This improved circularity enhances the reliability of quantitative image features, offering significant advantages for diverse ophthalmological applications, particularly glaucoma assessment and myopia management.
Предметные рубрики MeSH
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.