Unsupervised Approaches for the Segmentation of Dry ARMD Lesions in Eye Fundus cSLO Images.

Clément Royer,Florence Rossant,Jérémie Sublime,Michel Paques

doi:10.3390/jimaging7080143

Abstract

Age-related macular degeneration (ARMD), a major cause of sight impairment for elderly people, is still not well understood despite intensive research. Measuring the size of the lesions in the fundus is the main biomarker of the severity of the disease and as such is widely used in clinical trials yet only relies on manual segmentation. Artificial intelligence, in particular automatic image analysis based on neural networks, has a major role to play in better understanding the disease, by analyzing the intrinsic optical properties of dry ARMD lesions from patient images. In this paper, we propose a comparison of automatic segmentation methods (classical computer vision method, machine learning method and deep learning method) in an unsupervised context applied on cSLO IR images. Among the methods compared, we propose an adaptation of a fully convolutional network, called W-net, as an efficient method for the segmentation of ARMD lesions. Unlike supervised segmentation methods, our algorithm does not require annotated data which are very difficult to obtain in this application. Our method was tested on a dataset of 328 images and has shown to reach higher quality results than other compared unsupervised methods with a F1 score of 0.87, while having a more stable model, even though in some specific cases, texture/edges-based methods can produce relevant results.

Highlights

Age-related macular degeneration (ARMD) is a degenerative disease that affects the retina, and a leading cause of visual loss.In this paper, we focus on the dry form of this pathology which currently does not have any treatments
In this work, we propose the first fully unsupervised application of automatic segmentation of geographic atrophy (GA) using W-net [2] on confocal Scanning Laser Ophtalmoscopy (cSLO) IR acquired images (Section 2) and to assess how well it performs compared with other state of the art unsupervised methods
We focus on unsupervised algorithms to segment the GA in IR cSLO eye fundus images with dry-ARMD

Summary

Introduction

We focus on the dry form of this pathology which currently does not have any treatments It is characterized by a progressive loss of pigment epithelium which engenders a lesion located in the macula, growing slowly and impeding more and more the patient central vision. The reliability of manual delineations is an issue as even experts tends to disagree on their segmentations [1] To solve this problem, in this work, we propose the first fully unsupervised application of automatic segmentation of GA using W-net [2] on cSLO IR acquired images (Section 2) and to assess how well it performs compared with other state of the art unsupervised methods. Our contribution is three-fold: First, we propose a successful adaptation of the original developed by Xia et al [2]

Objectives

Methods

Results

Conclusion