Bonnet: Ultra-fast whole-body bone segmentation from CT scans
This work addresses the need for fast bone segmentation in time-critical medical applications like surgical planning, though it is incremental as it builds on existing U-Net architectures with novel optimizations.
The authors tackled the problem of slow whole-body bone segmentation from CT scans by proposing Bonnet, an ultra-fast sparse-volume pipeline that achieves high Dice scores across ribs, pelvis, and spine while reducing inference time by roughly 25x compared to baselines, running in only 2.69 seconds per scan.
This work proposes Bonnet, an ultra-fast sparse-volume pipeline for whole-body bone segmentation from CT scans. Accurate bone segmentation is important for surgical planning and anatomical analysis, but existing 3D voxel-based models such as nnU-Net and STU-Net require heavy computation and often take several minutes per scan, which limits time-critical use. The proposed Bonnet addresses this by integrating a series of novel framework components including HU-based bone thresholding, patch-wise inference with a sparse spconv-based U-Net, and multi-window fusion into a full-volume prediction. Trained on TotalSegmentator and evaluated without additional tuning on RibSeg, CT-Pelvic1K, and CT-Spine1K, Bonnet achieves high Dice across ribs, pelvis, and spine while running in only 2.69 seconds per scan on an RTX A6000. Compared to strong voxel baselines, Bonnet attains a similar accuracy but reduces inference time by roughly 25x on the same hardware and tiling setup. The toolkit and pre-trained models will be released at https://github.com/HINTLab/Bonnet.