DiffSegLung: Diffusion Radiomic Distillation for Unsupervised Lung Pathology Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Khiati, Rezkellah Noureddine, Brillet, Pierre-Yves, Fetita, Catalin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909035413897216
author Khiati, Rezkellah Noureddine
Brillet, Pierre-Yves
Fetita, Catalin
author_facet Khiati, Rezkellah Noureddine
Brillet, Pierre-Yves
Fetita, Catalin
contents Unsupervised segmentation of pulmonary pathologies in CT remains an open challenge due to the absence of annotated multi pathology cohorts and the failure of existing diffusion-based methods to exploit the quantitative Hounsfield Unit (HU) signal that physically distinguishes tissue classes. To address this, we propose DiffSegLung,a framework that introduces Diffusion Radiomic Distillation, in which handcrafted radiomic descriptors serve as a physics grounded teacher to shape the bottleneck of a 3D diffusion U-Net via a contrastive objective, transferring pathology discriminative structure into the learned representation without any annotations. At inference, the teacher is discarded and multitimestep bottleneck features are clustered by a Gaussian Mixture Model with HU-guided label assignment, followed by Sobel Diffusion Fusion for boundary refinement. Evaluated on 190 expert annotated axial slices drawn from four heterogeneous CT cohorts, Diff-SegLung improves segmentation across all four pathology classes over unsupervised baselines and improves generation fidelity over prior CT diffusion models.
format Preprint
id arxiv_https___arxiv_org_abs_2605_11758
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DiffSegLung: Diffusion Radiomic Distillation for Unsupervised Lung Pathology Segmentation
Khiati, Rezkellah Noureddine
Brillet, Pierre-Yves
Fetita, Catalin
Image and Video Processing
Computer Vision and Pattern Recognition
Unsupervised segmentation of pulmonary pathologies in CT remains an open challenge due to the absence of annotated multi pathology cohorts and the failure of existing diffusion-based methods to exploit the quantitative Hounsfield Unit (HU) signal that physically distinguishes tissue classes. To address this, we propose DiffSegLung,a framework that introduces Diffusion Radiomic Distillation, in which handcrafted radiomic descriptors serve as a physics grounded teacher to shape the bottleneck of a 3D diffusion U-Net via a contrastive objective, transferring pathology discriminative structure into the learned representation without any annotations. At inference, the teacher is discarded and multitimestep bottleneck features are clustered by a Gaussian Mixture Model with HU-guided label assignment, followed by Sobel Diffusion Fusion for boundary refinement. Evaluated on 190 expert annotated axial slices drawn from four heterogeneous CT cohorts, Diff-SegLung improves segmentation across all four pathology classes over unsupervised baselines and improves generation fidelity over prior CT diffusion models.
title DiffSegLung: Diffusion Radiomic Distillation for Unsupervised Lung Pathology Segmentation
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.11758