AI and the World's Hard Problems · Medicine & drug discovery · Nature (computational biology) · 2026-10-21
Automatic segmentation of histological zones in rat distraction osteogenesis using a self-configuring deep learning pipeline
Histologic assessment of regenerate composition is essential for evaluating tissue maturation in distraction osteogenesis, but manual delineation of tissue compartments is labor-intensive and subject to observer variability. The study evaluated a deep learning workflow that automatically segments three histologic zones in a rat distraction osteogenesis model: the fibrous interzone, the microcolumn formation zone, and the remodeling zone. Segmentation was most reliable for the fibrous interzone (Dice 0.76 in H&E; 0.75 in pentachrome) and hardest for the remodeling zone (Dice 0.29; 0.44); combined quantitative and qualitative evaluation favored the pentachrome multi-class configuration, and implementation in 3D Slicer makes the pipeline practical for translational research.