Statistical and Topological Summaries Aid Disease Detection for Segmented Retinal Vascular Images

TL;DR

This study evaluates statistical and topological summaries of segmented retinal images to improve machine learning detection of microvascular diseases like diabetic retinopathy, highlighting the effectiveness of specific descriptors across datasets.

Contribution

It compares statistical and topological descriptors for disease detection, identifying the most effective features and demonstrating their potential for automated retinal disease assessment.

Findings

Box-counting descriptor achieves high accuracy across datasets.

Topological Flooding descriptor is sensitive to annotation differences.

Merged datasets favor the Box-counting descriptor for detection.

Abstract

Disease complications can alter vascular network morphology and disrupt tissue functioning. Diabetic retinopathy, for example, is a complication of types 1 and 2 diabetes mellitus that can cause blindness. Microvascular diseases are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images. We apply our methods to three publicly-available datasets and find that, among the 13 total descriptor vectors we consider, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels on these datasets. We then created a fourth dataset by merging several datasets: the Box-counting vector outperforms all descriptors on this dataset, including the topological Flooding vector which is sensitive to differences in the annotation styles within the combined dataset. Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease as well as some of their current limitations. In the longer term, these methods could be incorporated into automated disease assessment tools.