Pathological myopia classification with simultaneous lesion segmentation using deep learning. (February 2021)

• Record Type: Journal Article
• Authors: Hemelings, Ruben; Elen, Bart; Blaschko, Matthew B.; Jacob, Julie; Stalmans, Ingeborg; De Boever, Patrick
• Is Part Of: Computer methods and programs in biomedicine. Volume 199(2021)
• Publication Date: 2021-02
• Subjects: Pathological myopia -- fovea localization -- peripapillary atrophy -- retinal detachment -- convolutional neural network -- fundus image -- glaucoma; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2020.105920
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Reports on methods and results, along with ablation study, for the classification of (non-)pathological myopia, and semantic segmentation of retinal atrophy and detachment from fundus images Presents a convolutional neural network bundling disease classification and segmentation of associated lesions, that generalizes well to other data sets Introduces a new approach to obtain fovea localization in fundus images through the reformulation as a segmentation problem A novel Optic Nerve Head (ONH)-based prediction enhancement for the segmentation of retinal atrophy and fovea is proposed This work was acknowledged with an award in the context of the "PAthoLogical Myopia detection from retinal images" challenge held during the IEEE International Symposium on Biomedical Imaging (April 2019) Abstract: Background and Objectives: Pathological myopia (PM) is the seventh leading cause of blindness, with a reported global prevalence up to 3%. Early and automated PM detection from fundus images could aid to prevent blindness in a world population that is characterized by a rising myopia prevalence. We aim to assess the use of convolutional neural networks (CNNs) for the detection of PM and semantic segmentation of myopia-induced lesions from fundus images on a recently introduced reference data set. Methods: This investigation reports on the results of CNNs developed for the recently introduced Pathological Myopia (PALM) dataset, which consists of 1200 images. Our CNN bundles lesion segmentation and PM classification, as the two tasks are heavily intertwined. Domain knowledge is also inserted through the introduction of a new Optic Nerve Head (ONH)-based prediction enhancement for the segmentation of atrophy and fovea localization. Finally, we are the first to approach fovea localization using segmentation instead of detection or regression models. Evaluation metrics include area under the receiver operating characteristic curve (AUC) for PM detection, Euclidean distance for fovea localization, and Dice and F1 metrics for the semantic segmentation tasks (optic disc, retinal atrophy and retinal detachment). Results: Models trained with 400 available training images achieved an AUC of 0.9867 for PM detection, and a Euclidean distance of 58.27 pixels on the fovea localization task, evaluated on a test set of 400 images. Dice and F1 metrics for semantic segmentation of lesions scored 0.9303 and 0.9869 on optic disc, 0.8001 and 0.9135 on retinal atrophy, and 0.8073 and 0.7059 on retinal detachment, respectively. Conclusions: We report a successful approach for a simultaneous classification of pathological myopia and segmentation of associated lesions. Our work was acknowledged with an award in the context of the "Pathological Myopia detection from retinal images" challenge held during the IEEE International Symposium on Biomedical Imaging (April 2019). Considering that (pathological) myopia cases are often identified as false positives and negatives in glaucoma deep learning models, we envisage that the current work could aid in future research to discriminate between glaucomatous and highly-myopic eyes, complemented by the localization and segmentation of landmarks such as fovea, optic disc and atrophy.