Unpaired Translation of Chest X-ray Images for Lung Opacity Diagnosis via Adaptive Activation Masks and Cross-Domain Alignment

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ning, Junzhi, Marshall, Dominic, Gao, Yijian, Nan, Xiaodan Xing Yang, Fang, Yingying, Zhang, Sheng, Komorowski, Matthieu, Yang, Guang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912297791782912
author Ning, Junzhi
Marshall, Dominic
Gao, Yijian
Nan, Xiaodan Xing Yang
Fang, Yingying
Zhang, Sheng
Komorowski, Matthieu
Yang, Guang
author_facet Ning, Junzhi
Marshall, Dominic
Gao, Yijian
Nan, Xiaodan Xing Yang
Fang, Yingying
Zhang, Sheng
Komorowski, Matthieu
Yang, Guang
contents Chest X-ray radiographs (CXRs) play a pivotal role in diagnosing and monitoring cardiopulmonary diseases. However, lung opacities in CXRs frequently obscure anatomical structures, impeding clear identification of lung borders and complicating the localization of pathology. This challenge significantly hampers segmentation accuracy and precise lesion identification, which are crucial for diagnosis. To tackle these issues, our study proposes an unpaired CXR translation framework that converts CXRs with lung opacities into counterparts without lung opacities while preserving semantic features. Central to our approach is the use of adaptive activation masks to selectively modify opacity regions in lung CXRs. Cross-domain alignment ensures translated CXRs without opacity issues align with feature maps and prediction labels from a pre-trained CXR lesion classifier, facilitating the interpretability of the translation process. We validate our method using RSNA, MIMIC-CXR-JPG and JSRT datasets, demonstrating superior translation quality through lower Frechet Inception Distance (FID) and Kernel Inception Distance (KID) scores compared to existing methods (FID: 67.18 vs. 210.4, KID: 0.01604 vs. 0.225). Evaluation on RSNA opacity, MIMIC acute respiratory distress syndrome (ARDS) patient CXRs and JSRT CXRs show our method enhances segmentation accuracy of lung borders and improves lesion classification, further underscoring its potential in clinical settings (RSNA: mIoU: 76.58% vs. 62.58%, Sensitivity: 85.58% vs. 77.03%; MIMIC ARDS: mIoU: 86.20% vs. 72.07%, Sensitivity: 92.68% vs. 86.85%; JSRT: mIoU: 91.08% vs. 85.6%, Sensitivity: 97.62% vs. 95.04%). Our approach advances CXR imaging analysis, especially in investigating segmentation impacts through image translation techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19860
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unpaired Translation of Chest X-ray Images for Lung Opacity Diagnosis via Adaptive Activation Masks and Cross-Domain Alignment
Ning, Junzhi
Marshall, Dominic
Gao, Yijian
Nan, Xiaodan Xing Yang
Fang, Yingying
Zhang, Sheng
Komorowski, Matthieu
Yang, Guang
Image and Video Processing
Computer Vision and Pattern Recognition
Chest X-ray radiographs (CXRs) play a pivotal role in diagnosing and monitoring cardiopulmonary diseases. However, lung opacities in CXRs frequently obscure anatomical structures, impeding clear identification of lung borders and complicating the localization of pathology. This challenge significantly hampers segmentation accuracy and precise lesion identification, which are crucial for diagnosis. To tackle these issues, our study proposes an unpaired CXR translation framework that converts CXRs with lung opacities into counterparts without lung opacities while preserving semantic features. Central to our approach is the use of adaptive activation masks to selectively modify opacity regions in lung CXRs. Cross-domain alignment ensures translated CXRs without opacity issues align with feature maps and prediction labels from a pre-trained CXR lesion classifier, facilitating the interpretability of the translation process. We validate our method using RSNA, MIMIC-CXR-JPG and JSRT datasets, demonstrating superior translation quality through lower Frechet Inception Distance (FID) and Kernel Inception Distance (KID) scores compared to existing methods (FID: 67.18 vs. 210.4, KID: 0.01604 vs. 0.225). Evaluation on RSNA opacity, MIMIC acute respiratory distress syndrome (ARDS) patient CXRs and JSRT CXRs show our method enhances segmentation accuracy of lung borders and improves lesion classification, further underscoring its potential in clinical settings (RSNA: mIoU: 76.58% vs. 62.58%, Sensitivity: 85.58% vs. 77.03%; MIMIC ARDS: mIoU: 86.20% vs. 72.07%, Sensitivity: 92.68% vs. 86.85%; JSRT: mIoU: 91.08% vs. 85.6%, Sensitivity: 97.62% vs. 95.04%). Our approach advances CXR imaging analysis, especially in investigating segmentation impacts through image translation techniques.
title Unpaired Translation of Chest X-ray Images for Lung Opacity Diagnosis via Adaptive Activation Masks and Cross-Domain Alignment
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.19860