Summary
Identifying small and complex structures in medical X-ray images usually needs many carefully labeled examples, which takes a lot of time. The authors explored using a special kind of AI that learns from just a few labeled points combined with a small number of fully labeled images to find tiny kidney parts called glomeruli in detailed scans. Their method could reliably locate these parts without needing fully labeled data for every image. Adding a bit more fully labeled data improved the AI’s accuracy to be close to fully supervised methods. This work shows that using less detailed labeling can still help computers analyze complex medical images efficiently.
weakly supervised learningsegmentationX-ray microCTdeep learningconvolutional neural networknnU-Netrenal glomerulisparse annotationsbiomedical imagingtomography
Authors
Daniele Rusconi, Michela Ascolese, Stephanie Fest-Santini, Alberto Bravin, Maurizio Santini
Abstract
Segmentation of complex structures in X-ray tomographic data is a fundamental task in biomedical research, but it often requires large amounts of precisely annotated data, making fully supervised approaches costly and difficult to scale. In this study, weakly supervised deep learning is investigated as a strategy to reduce annotation effort while maintaining accurate segmentation. A two-dimensional convolutional neural network based on the nnU-Net framework was adapted to a weak supervision setting using sparse dot-based annotations, complemented by a limited number of fully segmented images. The approach was evaluated on high-resolution microCT slices of rat kidneys, targeting the segmentation of renal glomeruli, which are small, low-contrast anatomical structures. Results indicate that weak supervision provides a meaningful learning signal, enabling reliable localization of glomeruli even in the absence of dense labels. Incorporating a small set of high-quality annotations substantially improves segmentation performance, approaching that of a fully supervised model. These findings highlight the potential of weakly supervised learning as an annotation-efficient strategy for the analysis of complex structures in X-ray tomographic data, and suggest that alternative loss formulations tailored to sparse annotations may further enhance performance.