Efficient KernelSHAP Explanations for Patch-based 3D Medical Image Segmentation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Brioso, Ricardo Coimbra, Sichili, Giulio, Dei, Damiano, Lambri, Nicola, Mancosu, Pietro, Scorsetti, Marta, Loiacono, Daniele
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914469685231616
author Brioso, Ricardo Coimbra
Sichili, Giulio
Dei, Damiano
Lambri, Nicola
Mancosu, Pietro
Scorsetti, Marta
Loiacono, Daniele
author_facet Brioso, Ricardo Coimbra
Sichili, Giulio
Dei, Damiano
Lambri, Nicola
Mancosu, Pietro
Scorsetti, Marta
Loiacono, Daniele
contents Perturbation-based explainability methods such as KernelSHAP provide model-agnostic attributions but are typically impractical for patch-based 3D medical image segmentation due to the large number of coalition evaluations and the high cost of sliding-window inference. We present an efficient KernelSHAP framework for volumetric CT segmentation that restricts computation to a user-defined region of interest and its receptive-field support, and accelerates inference via patch logit caching, reusing baseline predictions for unaffected patches while preserving nnU-Net's fusion scheme. To enable clinically meaningful attributions, we compare three automatically generated feature abstractions within the receptive-field crop: whole-organ units, regular FCC supervoxels, and hybrid organ-aware supervoxels, and we study multiple aggregation/value functions targeting stabilizing evidence (TP/Dice/Soft Dice) or false-positive behavior. Experiments on whole-body CT segmentations show that caching substantially reduces redundant computation (with computational savings ranging from 15% to 30%) and that faithfulness and interpretability exhibit clear trade-offs: regular supervoxels often maximize perturbation-based metrics but lack anatomical alignment, whereas organ-aware units yield more clinically interpretable explanations and are particularly effective for highlighting false-positive drivers under normalized metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11775
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Efficient KernelSHAP Explanations for Patch-based 3D Medical Image Segmentation
Brioso, Ricardo Coimbra
Sichili, Giulio
Dei, Damiano
Lambri, Nicola
Mancosu, Pietro
Scorsetti, Marta
Loiacono, Daniele
Computer Vision and Pattern Recognition
Artificial Intelligence
Perturbation-based explainability methods such as KernelSHAP provide model-agnostic attributions but are typically impractical for patch-based 3D medical image segmentation due to the large number of coalition evaluations and the high cost of sliding-window inference. We present an efficient KernelSHAP framework for volumetric CT segmentation that restricts computation to a user-defined region of interest and its receptive-field support, and accelerates inference via patch logit caching, reusing baseline predictions for unaffected patches while preserving nnU-Net's fusion scheme. To enable clinically meaningful attributions, we compare three automatically generated feature abstractions within the receptive-field crop: whole-organ units, regular FCC supervoxels, and hybrid organ-aware supervoxels, and we study multiple aggregation/value functions targeting stabilizing evidence (TP/Dice/Soft Dice) or false-positive behavior. Experiments on whole-body CT segmentations show that caching substantially reduces redundant computation (with computational savings ranging from 15% to 30%) and that faithfulness and interpretability exhibit clear trade-offs: regular supervoxels often maximize perturbation-based metrics but lack anatomical alignment, whereas organ-aware units yield more clinically interpretable explanations and are particularly effective for highlighting false-positive drivers under normalized metrics.
title Efficient KernelSHAP Explanations for Patch-based 3D Medical Image Segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2604.11775