Saved in:
Bibliographic Details
Main Authors: Lei, Xiaoning, Sun, Jianwei, Cai, Wenhao, Xu, Xichen, Wang, Yanshu, Gao, Hu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.14187
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917360499163136
author Lei, Xiaoning
Sun, Jianwei
Cai, Wenhao
Xu, Xichen
Wang, Yanshu
Gao, Hu
author_facet Lei, Xiaoning
Sun, Jianwei
Cai, Wenhao
Xu, Xichen
Wang, Yanshu
Gao, Hu
contents Task-based measures of image quality (IQ) are critical for evaluating medical imaging systems, which must account for randomness including anatomical variability. Stochastic object models (SOMs) provide a statistical description of such variability, but conventional mathematical SOMs fail to capture realistic anatomy, while data-driven approaches typically require clean data rarely available in clinical tasks. To address this challenge, we propose AMID, an unsupervised Ambient Measurement-Integrated Diffusion with noise decoupling, which establishes clean SOMs directly from noisy measurements. AMID introduces a measurement-integrated strategy aligning measurement noise with the diffusion trajectory, and explicitly models coupling between measurement and diffusion noise across steps, an ambient loss is thus designed base on it to learn clean SOMs. Experiments on real CT and mammography datasets show that AMID outperforms existing methods in generation fidelity and yields more reliable task-based IQ evaluation, demonstrating its potential for unsupervised medical imaging analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2512_14187
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Establishing Stochastic Object Models from Noisy Data via Ambient Measurement-Integrated Diffusion
Lei, Xiaoning
Sun, Jianwei
Cai, Wenhao
Xu, Xichen
Wang, Yanshu
Gao, Hu
Graphics
Computer Vision and Pattern Recognition
Task-based measures of image quality (IQ) are critical for evaluating medical imaging systems, which must account for randomness including anatomical variability. Stochastic object models (SOMs) provide a statistical description of such variability, but conventional mathematical SOMs fail to capture realistic anatomy, while data-driven approaches typically require clean data rarely available in clinical tasks. To address this challenge, we propose AMID, an unsupervised Ambient Measurement-Integrated Diffusion with noise decoupling, which establishes clean SOMs directly from noisy measurements. AMID introduces a measurement-integrated strategy aligning measurement noise with the diffusion trajectory, and explicitly models coupling between measurement and diffusion noise across steps, an ambient loss is thus designed base on it to learn clean SOMs. Experiments on real CT and mammography datasets show that AMID outperforms existing methods in generation fidelity and yields more reliable task-based IQ evaluation, demonstrating its potential for unsupervised medical imaging analysis.
title Establishing Stochastic Object Models from Noisy Data via Ambient Measurement-Integrated Diffusion
topic Graphics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.14187