A Deep Convolutional Neural Network-Based Novel Class Balancing for Imbalance Data Segmentation

TL;DR

This paper introduces BLCB-CNN, a deep learning approach with bi-level class balancing and preprocessing techniques to improve retinal vessel segmentation in imbalanced fundus images, achieving high accuracy and generalization.

Contribution

The paper presents a novel bi-level class balancing scheme integrated with CNN for retinal vessel segmentation, addressing class imbalance and vessel thickness variation.

Findings

Achieved 98.23% AUC in vessel segmentation

Attained 96.22% accuracy on standard datasets

Demonstrated strong generalization on external data

Abstract

Retinal fundus images provide valuable insights into the human eye's interior structure and crucial features, such as blood vessels, optic disk, macula, and fovea. However, accurate segmentation of retinal blood vessels can be challenging due to imbalanced data distribution and varying vessel thickness. In this paper, we propose BLCB-CNN, a novel pipeline based on deep learning and bi-level class balancing scheme to achieve vessel segmentation in retinal fundus images. The BLCB-CNN scheme uses a Convolutional Neural Network (CNN) architecture and an empirical approach to balance the distribution of pixels across vessel and non-vessel classes and within thin and thick vessels. Level-I is used for vessel/non-vessel balancing and Level-II is used for thick/thin vessel balancing. Additionally, pre-processing of the input retinal fundus image is performed by Global Contrast Normalization (GCN), Contrast Limited Adaptive Histogram Equalization (CLAHE), and gamma corrections to increase intensity uniformity as well as to enhance the contrast between vessels and background pixels. The resulting balanced dataset is used for classification-based segmentation of the retinal vascular tree. We evaluate the proposed scheme on standard retinal fundus images and achieve superior performance measures, including an area under the ROC curve of 98.23%, Accuracy of 96.22%, Sensitivity of 81.57%, and Specificity of 97.65%. We also demonstrate the method's efficacy through external cross-validation on STARE images, confirming its generalization ability.