Glomeruli detection and classification in histopathological images using deep learning semantic segmentation

Key Points

• Detection and classification of glomeruli showed a significant improvement in accuracy with deep learning methods.
• The deep learning model achieved an accuracy of 92% in identifying glomeruli in histopathological images.
• Using semantic segmentation, the model was trained on a dataset of annotated images to enhance classification reliability.
• The findings support the potential for AI-based tools in histopathology, though external validation is necessary.