# Submanifold Sparse Convolutional Networks for Automated 3D Segmentation of Kidneys and Kidney Tumours in Computed Tomography Source: [Submanifold Sparse Convolutional Networks for Automated 3D Segmentation of Kidneys and Kidney Tumours in Computed Tomography]() Collection index: [llms.txt]() Data updated: 2026-05-09T05:00:12.752Z Authors: Sa\\'ul Alonso-Monsalve, Leigh H. Whitehead, Adam Aurisano, Lorena Escudero Sanchez Published: 2025-11-06T13:17:16.000Z ## Abstract Accurate delineation of kidney tumours in Computed Tomography (CT) is essential for downstream quantitative analysis and precision oncology, but manual segmentation is a specialised task, time-consuming and difficult to scale. Automated 3D segmentation remains challenging because CT scans are large volumetric images, making high-resolution dense convolutional networks computationally expensive and often dependent on downsampling or patch-based inference. We propose a two-stage 3D segmentation methodology based on voxel sparsification and submanifold sparse convolutional networks (SSCNs). Stage 1 uses a low-resolution sparse network to identify a region of interest (ROI); Stage 2 applies a high-resolution sparse network for refined segmentation within the cropped ROI. This enables native high-resolution 3D processing while reducing memory use and inference time. We evaluate the method on the KiTS23 renal cancer CT dataset using 5-fold cross-validation. Our method achieved Dice similarity coefficients of 95.8% for kidneys + masses, 85.7% for tumours + cysts, and 80.3% for tumours alone, competitive with top KiTS23 approaches. In direct comparisons on the same cross-validation folds, the proposed sparse method achieves tumour + cyst and tumour-only Dice scores comparable to, and slightly higher than, a patch-based nnU-Net baseline, while consistently requiring less VRAM and shorter inference time across the tested hardware. Across the tested GPUs, our sparse model is markedly faster than both nnU-Net and the zero-shot zoom-out/zoom-in foundation model SegVol, which localises kidneys well but underperforms on small heterogeneous lesions. Compared to an equivalent dense implementation of the same architecture, the proposed sparse approach achieves up to a 60% reduction in inference time and up to a 75% reduction in VRAM usage across both CPU and the GPU configurations tested. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Visual Segmentation on KiTS23](): Visual Segmentation; KiTS23; KTC Dice Score: 95.8. - [Semantic Segmentation on KiTS23 (Fold 1)](): Semantic segmentation; KiTS23 (Fold 1); Dice (Kidneys + masses): 95.26. - [Semantic Segmentation on KiTS23 (Fold 2)](): Semantic segmentation; KiTS23 (Fold 2); Dice Score (Kidneys + Masses): 95.14. - [Semantic Segmentation on KiTS23 (Fold 3)](): Semantic segmentation; KiTS23 (Fold 3); Dice (Kidneys + masses): 96.76. - [Semantic Segmentation on KiTS23 (Fold 4)](): Semantic segmentation; KiTS23 (Fold 4); Dice (Kidneys + masses): 95.8. - [Semantic Segmentation on KiTS23 (Fold 5)](): Semantic segmentation; KiTS23 (Fold 5); Dice Score (Kidneys + masses): 96.1. - [Semantic Segmentation on KiTS23 (5-fold cross-val average)](): Semantic segmentation; KiTS23 (5-fold cross-val average); Dice (Kidneys + masses): 95.81. ## Original sources and implementation [Original paper]() [PDF]()