CTLESS: A Scatter-Window Projection and Deep Learning-Based Transmission-Less Attenuation Compensation Method for Myocardial Perfusion SPECT
Open Access
- 25 November 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (3) , 1308-1320
Abstract
Attenuation compensation (AC), while being beneficial for visual-interpretation tasks in myocardial perfusion imaging (MPI) by single-photon emission computed tomography (SPECT), typically requires the availability of a separate X-ray CT component, leading to additional radiation dose, higher costs, and potentially inaccurate diagnosis in case of misalignment between SPECT and CT images. To address these issues, we developed a method for cardiac SPECT AC using deep learning and emission scatter-window photons without a separate transmission scan (CTLESS). In this method, an estimated attenuation map reconstructed from scatter-energy window projections is segmented into different regions using a multi-channel input multi-decoder network trained on CT scans. Pre-defined attenuation coefficients are assigned to these regions, yielding the attenuation map used for AC. We objectively evaluated this method in a retrospective study with anonymized clinical SPECT/CT stress MPI images on the clinical task of detecting perfusion defects with an anthropomorphic model observer. CTLESS yielded statistically non-inferior performance compared to a CT-based AC (CTAC) method and significantly outperformed a non-AC (NAC) method on this clinical task. Similar results were observed in stratified analyses with different sexes, defect extents, and defect severities. The method was observed to generalize across two SPECT scanners, each with a different camera. In addition, CTLESS yielded similar performance as CTAC and outperformed NAC method on the fidelity-based figures of merit, namely, root mean squared error (RMSE) and structural similarity index measure (SSIM). Moreover, as we reduced the training dataset size, CTLESS yielded relatively stable AUC values and generally outperformed another DL-based AC method that directly estimated the attenuation coefficient within each voxel. These results demonstrate the capability of the CTLESS method for transmission-less AC in SPECT and motivate further clinical evaluation.Keywords
All Related Versions
Funding Information
- National Institute of Biomedical Imaging and Bioengineering of National Institute of Health (R01-EB031051, R01-EB031962)
- Bradley-Alavi Student Fellowship from the Education and Research Foundation for Nuclear Medicine and Molecular Imaging (ERF) of the Society of Nuclear Medicine and Molecular Imaging
- National Science Foundation (NSF) Early Career Development (CAREER) Program (2239707)
- National Institute of Biomedical Imaging and Bioengineering of National Institute of Health (R21-EB024647)
This publication has 70 references indexed in Scilit:
- Need for objective task‐based evaluation of deep learning‐based denoising methods: A study in the context of myocardial perfusion SPECTMedical Physics, 2023
- “Virtual” attenuation correction: improving stress myocardial perfusion SPECT imaging using deep learningEuropean Journal of Nuclear Medicine and Molecular Imaging, 2022
- Direct and indirect strategies of deep-learning-based attenuation correction for general purpose and dedicated cardiac SPECTEuropean Journal of Nuclear Medicine and Molecular Imaging, 2022
- Standard SPECT myocardial perfusion estimation from half-time acquisitions using deep convolutional residual neural networksJournal of Nuclear Cardiology, 2020
- Deep learning-based attenuation map generation for myocardial perfusion SPECTEuropean Journal of Nuclear Medicine and Molecular Imaging, 2020
- The important role of task-based model observers and related techniques in medical imagingJournal of Nuclear Cardiology, 2019
- Statistical Methods in Diagnostic Medicine2011
- Basic performance and stability of a CdTe solid-state detector panelAnnals of Nuclear Medicine, 2010
- Maximum likelihood estimation of receiver operating characteristic (ROC) curves from continuously-distributed dataStatistics in Medicine, 1998
- Detecting and assessing severity of coronary artery disease in humansCardioVascular and Interventional Radiology, 1990