Organ Segmentation with Machine Learning Models
TL;DR
Ten architectures spanning convolutional, attention-based, transformer, and state–space (Mamba) families on the AMOS CT dataset under one identical nnU-Net-style pipeline are benchmarked, with per-organ Dice, 95-percentile Hausdorff Distance (HD95), and Normalised Surface Dice, with pairwise significance tested on an independent external dataset (TotalSegmentator).
Abstract
Accurate segmentation of abdominal organs in Computed Tomography (CT) underpins radiotherapy planning, surgical planning, and disease monitoring. Existing benchmarks rank architectures by a single aggregate Dice score, without per-organ statistical testing or boundary-sensitive metrics, even though models are chosen organ by organ for clinical use. We benchmark ten architectures spanning convolutional, attention-based, transformer, and state–space (Mamba) families on the AMOS CT dataset under one identical nnU-Net-style pipeline; we report per-organ Dice, 95-percentile Hausdorff Distance (HD95), and Normalised Surface Dice, with pairwise significance tested on an independent external dataset (TotalSegmentator). A competitive cluster of convolutional and Mamba models leads; rankings are stable on large organs but reshuffle by 10–13% on the small, geometrically complex ones, and boundary fidelity separates the models into tiers that the Dice ranking hides. This ordering largely holds on the external set (Spearman ρ=0.84). Selecting a model on aggregate Dice alone is therefore unsafe for organ-specific clinical tasks: per-organ overlap and boundary metrics should be the primary acceptance criteria for selecting a model before clinical deployment.