Joint Optic Disc and Cup Segmentation Based on Multi-label Deep Network and Polar Transformation

TL;DR

This paper introduces M-Net, a multi-label deep learning system that jointly segments optic disc and cup in fundus images, utilizing multi-scale input, U-shape architecture, and polar transformation to improve glaucoma diagnosis accuracy.

Contribution

The paper presents a novel joint segmentation model with multi-scale input, multi-label loss, and polar transformation, advancing the accuracy and efficiency of glaucoma-related optic structure segmentation.

Findings

Achieved state-of-the-art segmentation results on ORIGA dataset.

Provided satisfactory glaucoma screening performance using CDR calculation.

Demonstrated the effectiveness of polar transformation in segmentation accuracy.

Abstract

Glaucoma is a chronic eye disease that leads to irreversible vision loss. The cup to disc ratio (CDR) plays an important role in the screening and diagnosis of glaucoma. Thus, the accurate and automatic segmentation of optic disc (OD) and optic cup (OC) from fundus images is a fundamental task. Most existing methods segment them separately, and rely on hand-crafted visual feature from fundus images. In this paper, we propose a deep learning architecture, named M-Net, which solves the OD and OC segmentation jointly in a one-stage multi-label system. The proposed M-Net mainly consists of multi-scale input layer, U-shape convolutional network, side-output layer, and multi-label loss function. The multi-scale input layer constructs an image pyramid to achieve multiple level receptive field sizes. The U-shape convolutional network is employed as the main body network structure to learn the rich hierarchical representation, while the side-output layer acts as an early classifier that produces a companion local prediction map for different scale layers. Finally, a multi-label loss function is proposed to generate the final segmentation map. For improving the segmentation performance further, we also introduce the polar transformation, which provides the representation of the original image in the polar coordinate system. The experiments show that our M-Net system achieves state-of-the-art OD and OC segmentation result on ORIGA dataset. Simultaneously, the proposed method also obtains the satisfactory glaucoma screening performances with calculated CDR value on both ORIGA and SCES datasets.