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

2025-06-23

AI Summary

To address the degradation in retinal vessel segmentation accuracy caused by dual-level class imbalance—between vessel and non-vessel pixels, and further between thick and thin vessels—this paper proposes a two-tiered class-balancing framework based on deep convolutional neural networks. The first tier balances the vessel/non-vessel pixel distribution, while the second tier explicitly rebalances thick- and thin-vessel pixel distributions. Preprocessing integrates global contrast normalization, contrast-limited adaptive histogram equalization, and gamma correction to enhance image contrast and intensity consistency. Evaluated on standard benchmarks, the method achieves 98.23% AUC and 96.22% accuracy. External validation on the STARE dataset demonstrates strong generalizability. Notably, the approach significantly improves detection accuracy for fine vessels, delivering a more robust segmentation foundation for clinical decision support.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Security and Privacy: Large-scale security measurements

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.

Problem

• Segments retinal vessels in imbalanced fundus images
• Balances pixel distribution for vessel/non-vessel classes
• Enhances contrast and intensity uniformity in preprocessing

Innovation

• BLCB-CNN with bi-level class balancing
• Global Contrast Normalization preprocessing
• CNN architecture for vessel segmentation