Weakly supervised neural network: segmentation of complex structures in X-ray microCT
Results indicate that weak supervision provides a meaningful learning signal, enabling reliable localization of glomeruli even in the absence of dense labels, and incorporation of a small set of high-quality annotations substantially improves segmentation performance, approaching that of a fully supervised model.
Daniele Rusconi, M. Ascolese, S. Fest-Santini et al.
· Journal of Instrumentation · 0 citations