Automated Joint Space Detection Improves Bone Segmentation Accuracy

Quantitative description of complex anatomical structures remains challenging due to the expertise necessary for manual segmentation, labor, and interobserver variability. To overcome this, automated detection of specific landmarks can be provided by digital image analysis techniques including deep learning (DL) models. To this end, we undertook supervised automated analysis of micro-computed tomography (micro-CT) datasets of murine hindpaws and forepaws (30-33 bones). Advancing beyond previously published semi-automated (SA) marker-based watershed algorithms, we added structure enhancement, tensor voting, and output dilation to identify joint spaces. Segmentation was enhanced by the use of a DL joint space prediction model (3D U-Net architecture, ResNet-18 backbone) using wild-type (WT) hindpaw labels as ground truth. Prediction was then extended to hindpaws (52.4% in test group) and forepaws from WT and tumor necrosis factor transgenic (TNF-Tg) mice with inflammatory-erosive arthritis of both sexes across age. Segmentation accuracy improved dramatically using the DL methodology (WT male: SA 79.39±5.73% vs DL 98.16±1.47%, p<0.0001; WT female: SA 79.16±4.84% vs DL 99.19±1.63%, p<0.0001). Accuracy declined with increased disease severity and age in TNF-Tg mice (TNF-Tg male 93.54±4.73%, female 91.81±2.80%, p≤0.01). Subsequent testing in forepaws also displayed progressive reduction in accuracy with increasing arthritic severity (i.e., WT 87.29±2.07%, TNF-Tg male 72.65±11.70%, p<0.0001). Overall, this supervised automated model outperforms recent SA approaches in healthy joints to enhance investigation of complex bone anatomy. Although flexible application to novel and disease-modified datasets demonstrates deprecated performance, utilization may nonetheless catalyze structure-specific segmentation model development.