MIE: Magnification-integrated ensemble method for improving glomeruli segmentation in medical imaging

• Published in: Computer methods and programs in biomedicine Vol. 271; p. 109041
• Main Authors: Han, Yechan, Kim, Jaeyun, Park, Samel, Moon, Jong-Seok, Lee, Ji-Hey
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.11.2025
• Subjects: Algorithms; Deep learning; Diagnostic Imaging - methods; Digital pathology; Ensemble; Glomeruli; Humans; Image Processing, Computer-Assisted - methods; Kidney Glomerulus - diagnostic imaging; Reproducibility of Results; Semantic segmentation; Glomeruli; Deep learning; Digital pathology; Semantic segmentation; Ensemble
• ISSN: 0169-2607; 1872-7565; 1872-7565
• DOI: 10.1016/j.cmpb.2025.109041

•Proposed a novel magnification-integrated ensemble segmentation method.•Used multi-scale WSIs from 15 patients for training, validation, and testing.•Models trained on fixed magnifications showed reduced cross-scale accuracy.•Achieved 87.72 mIoU and 93.04 Dice score with the U-Net model.•Enhanced segmentation robustness across varying magnifications. Glomeruli are crucial for blood filtration, waste removal, and regulation of essential substances in the body. Traditional methods for detecting glomeruli rely on human interpretation, which can lead to variability. AI techniques have improved this process; however, most studies have used images with fixed magnification. This study proposes a novel magnification-integrated ensemble method to enhance glomerular segmentation accuracy. Whole-slide images (WSIs) from 12 patients were used for training, two for validation, and one for testing. Patch and mask images were extracted at 256 × 256 size × x2, x3, and x4 magnification levels. Data augmentation techniques, such as RandomResize, RandomCrop, and RandomFlip, were used. The segmentation model underwent 80,000 iterations with a stochastic gradient descent (SGD). Performance varied with changes in magnification. The models trained on high-magnification images showed significant drops when tested at lower magnifications, and vice versa. The proposed method improved segmentation accuracy across different magnifications, achieving 87.72 mIoU and 93.04 Dice score with the U-Net model. The magnification-integrated ensemble method significantly enhanced glomeruli segmentation accuracy across varying magnifications, thereby addressing the limitations of fixed magnification models. This approach improves the robustness and reliability of AI-driven diagnostic tools, potentially benefiting various medical imaging applications by ensuring consistent performance despite changes in image magnification.