Robust Quantification of Percent Emphysema on CT Via Domain Attention: The Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study

Xuzhe Zhang; Elsa D. Angelini; Eric A. Hoffman; Karol E. Watson; Benjamin M. Smith; R. Graham Barr; Andrew F. Laine

doi:10.1109/ISBI56570.2024.10635299

Back

Conference proceeding

Robust Quantification of Percent Emphysema on CT Via Domain Attention: The Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study

Xuzhe Zhang, Elsa D. Angelini, Eric A. Hoffman, Karol E. Watson, Benjamin M. Smith, R. Graham Barr and Andrew F. Laine

2024 IEEE International Symposium on Biomedical Imaging (ISBI), pp.1-5

05/27/2024

DOI: 10.1109/ISBI56570.2024.10635299

PMCID: PMC11388062

PMID: 39267982

Files and links (1)

url

https://pmc.ncbi.nlm.nih.gov/articles/PMC11388062/pdf/nihms-2021036.pdfView

Open Access

Abstract

Robust quantification of pulmonary emphysema on computed tomography (CT) remains challenging for large-scale research studies that involve scans from different scanner types and for translation to clinical scans. Although the domain shifts in different CT scanners are subtle compared to shifts existing in other modalities (e.g., MRI) or cross-modality, emphysema is highly sensitive to it. Such subtle difference limits the application of general domain adaptation methods, such as image translation-based methods, as the contrast difference is too subtle to be distinguished. Existing studies have explored several directions to tackle this challenge, including density correction, noise filtering, regression, hidden Markov measure field (HMMF) model-based segmentation, and volume-adjusted lung density. Despite some promising results, previous studies either required a tedious workflow or eliminated opportunities for downstream emphysema subtyping, limiting efficient adaptation on a large-scale study. To alleviate this dilemma, we developed an end-to-end deep learning framework based on an existing HMMF segmentation framework. We first demonstrate that a regular UNet cannot replicate the existing HMMF results because of the lack of scanner priors. We then design a novel domain attention block, a simple yet efficient cross-modal block to fuse image visual features with quantitative scanner priors (a sequence), which significantly improves the results.

Computed Tomography

Magnetic Resonance Imaging

Noise

deep learning

Hidden Markov models

Image segmentation

multi-modal learning

pulmonary emphysema

segmentation

Visualization

Volume measurement

Details

Title: Subtitle: Robust Quantification of Percent Emphysema on CT Via Domain Attention: The Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study
Creators: Xuzhe Zhang - Columbia University
Elsa D. Angelini - Columbia University
Eric A. Hoffman - University of Iowa
Karol E. Watson - University of California, Los Angeles
Benjamin M. Smith - Columbia University Irving Medical Center
R. Graham Barr - Columbia University Irving Medical Center
Andrew F. Laine - Columbia University
Resource Type: Conference proceeding
Publication Details: 2024 IEEE International Symposium on Biomedical Imaging (ISBI), pp.1-5
DOI: 10.1109/ISBI56570.2024.10635299
PMID: 39267982
PMCID: PMC11388062
NLM abbreviation: Proc IEEE Int Symp Biomed Imaging
eISSN: 1945-8452
Publisher: IEEE
Grant note: National Center for Advancing Translational Sciences (10.13039/100006108) National Institutes of Health (10.13039/100000002)
Language: English
Date published: 05/27/2024
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Radiology; Internal Medicine
Record Identifier: 9984699247702771

Metrics

8 Record Views