Unifying Structure Analysis and Surrogate-Driven Function Regression for Glaucoma OCT Image Screening

Abstract

Optical Coherence Tomography (OCT) imaging plays an important role in glaucoma diagnosis in clinical practice. Early detection and timely treatment can prevent glaucoma patients from permanent vision loss. However, only a dearth of automated methods has been developed based on OCT images for glaucoma study. In this paper, we present a novel framework to effectively classify glaucoma OCT images from normal ones. A semi-supervised learning strategy with smoothness assumption is applied for surrogate assignment of missing function regression labels. Besides, the proposed multi-task learning network is capable of exploring the structure and function relationship from the OCT image and visual field measurement simultaneously, which contributes to classification performance boosting. Essentially, we are the first to unify the structure analysis and function regression for glaucoma screening. It is also worth noting that we build the largest glaucoma OCT image dataset involving 4877 volumes to develop and evaluate the proposed method. Extensive experiments demonstrate that our framework outperforms the baseline methods and two glaucoma experts by a large margin, achieving 93.2%, 93.2% and 97.8% on accuracy, F1 score and AUC, respectively.

Circumpapillary OCT-focused hybrid learning for glaucoma grading using tailored prototypical neural networks

“…The main outcomes revealed that the deep learning approach outperformed conventional SD-OCT parameters to discriminate glaucomatous from normal samples. At this point, it should be noted that there are other studies which addressed the binary classification between healthy and glaucomatous cases by applying deep learning algorithms on spectral-domain OCT volumes [38][39][40], including our previous work [41].…”

“…Most of the previous studies centred on B-scans are designed to discriminate glaucoma from healthy samples [36,[46][47][48][49][50]. Other state-of-the-art studies also pursued the classification of healthy and glaucomatous cases, but using the OCT volumes as an input to their models [37][38][39][40][41]. Additional glaucoma-related studies were addressed from B-scans to accomplish different discrimination tasks such as preperimetric vs perimetric glaucoma [51,52], progressing vs non-progressing glaucoma [53,54] and close angle vs open-angle glaucoma [55,56], among others.…”