PADM: A Physics-aware Diffusion Model for Attenuation Correction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pham, Trung Kien, Vu, Hoang Minh, Chu, Anh Duc, Nguyen, Dac Thai, Nguyen, Trung Thanh, Truong, Thao Nguyen, Son, Mai Hong, Nguyen, Thanh Trung, Nguyen, Phi Le
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918193584406528
author Pham, Trung Kien
Vu, Hoang Minh
Chu, Anh Duc
Nguyen, Dac Thai
Nguyen, Trung Thanh
Truong, Thao Nguyen
Son, Mai Hong
Nguyen, Thanh Trung
Nguyen, Phi Le
author_facet Pham, Trung Kien
Vu, Hoang Minh
Chu, Anh Duc
Nguyen, Dac Thai
Nguyen, Trung Thanh
Truong, Thao Nguyen
Son, Mai Hong
Nguyen, Thanh Trung
Nguyen, Phi Le
contents Attenuation artifacts remain a significant challenge in cardiac Myocardial Perfusion Imaging (MPI) using Single-Photon Emission Computed Tomography (SPECT), often compromising diagnostic accuracy and reducing clinical interpretability. While hybrid SPECT/CT systems mitigate these artifacts through CT-derived attenuation maps, their high cost, limited accessibility, and added radiation exposure hinder widespread clinical adoption. In this study, we propose a novel CT-free solution to attenuation correction in cardiac SPECT. Specifically, we introduce Physics-aware Attenuation Correction Diffusion Model (PADM), a diffusion-based generative method that incorporates explicit physics priors via a teacher--student distillation mechanism. This approach enables attenuation artifact correction using only Non-Attenuation-Corrected (NAC) input, while still benefiting from physics-informed supervision during training. To support this work, we also introduce CardiAC, a comprehensive dataset comprising 424 patient studies with paired NAC and Attenuation-Corrected (AC) reconstructions, alongside high-resolution CT-based attenuation maps. Extensive experiments demonstrate that PADM outperforms state-of-the-art generative models, delivering superior reconstruction fidelity across both quantitative metrics and visual assessment.
format Preprint
id arxiv_https___arxiv_org_abs_2511_06948
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PADM: A Physics-aware Diffusion Model for Attenuation Correction
Pham, Trung Kien
Vu, Hoang Minh
Chu, Anh Duc
Nguyen, Dac Thai
Nguyen, Trung Thanh
Truong, Thao Nguyen
Son, Mai Hong
Nguyen, Thanh Trung
Nguyen, Phi Le
Computer Vision and Pattern Recognition
Attenuation artifacts remain a significant challenge in cardiac Myocardial Perfusion Imaging (MPI) using Single-Photon Emission Computed Tomography (SPECT), often compromising diagnostic accuracy and reducing clinical interpretability. While hybrid SPECT/CT systems mitigate these artifacts through CT-derived attenuation maps, their high cost, limited accessibility, and added radiation exposure hinder widespread clinical adoption. In this study, we propose a novel CT-free solution to attenuation correction in cardiac SPECT. Specifically, we introduce Physics-aware Attenuation Correction Diffusion Model (PADM), a diffusion-based generative method that incorporates explicit physics priors via a teacher--student distillation mechanism. This approach enables attenuation artifact correction using only Non-Attenuation-Corrected (NAC) input, while still benefiting from physics-informed supervision during training. To support this work, we also introduce CardiAC, a comprehensive dataset comprising 424 patient studies with paired NAC and Attenuation-Corrected (AC) reconstructions, alongside high-resolution CT-based attenuation maps. Extensive experiments demonstrate that PADM outperforms state-of-the-art generative models, delivering superior reconstruction fidelity across both quantitative metrics and visual assessment.
title PADM: A Physics-aware Diffusion Model for Attenuation Correction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.06948