Data Management and Network Architecture Effect on Performance Variability in Direct Attenuation Correction via Deep Learning for Cardiac SPECT: A Feasibility Study

Mahsa Torkaman; Jaewon Yang; Luyao Shi; Rui Wang; Edward J Miller; Albert J Sinusas; Chi Liu; Grant T Gullberg; Youngho Seo

doi:10.1109/trpms.2021.3138372

Data Management and Network Architecture Effect on Performance Variability in Direct Attenuation Correction via Deep Learning for Cardiac SPECT: A Feasibility Study

IEEE Trans Radiat Plasma Med Sci. 2022 Sep;6(7):755-765. doi: 10.1109/trpms.2021.3138372. Epub 2021 Dec 24.

Authors

Mahsa Torkaman¹, Jaewon Yang¹, Luyao Shi², Rui Wang³, Edward J Miller³, Albert J Sinusas⁴, Chi Liu⁴, Grant T Gullberg¹, Youngho Seo¹

Affiliations

¹ Radiology and Biomedical Imaging Department, University of California, San Francisco, CA, USA.
² Biomedical Engineering Department, Yale University, New Haven, CT, USA.
³ Radiology and Biomedical Imaging Department, Yale University, New Haven, CT, USA.
⁴ Biomedical Engineering Department, Yale University, New Haven, CT, USA; Radiology and Biomedical Imaging Department, Yale University, New Haven, CT, USA.

Abstract

Attenuation correction (AC) is important for accurate interpretation of SPECT myocardial perfusion imaging (MPI). However, it is challenging to perform AC in dedicated cardiac systems not equipped with a transmission imaging capability. Previously, we demonstrated the feasibility of generating attenuation-corrected SPECT images using a deep learning technique (SPECT_DL) directly from non-corrected images (SPECT_NC). However, we observed performance variability across patients which is an important factor for clinical translation of the technique. In this study, we investigate the feasibility of overcoming the performance variability across patients for the direct AC in SPECT MPI by proposing to develop an advanced network and a data management strategy. To investigate, we compared the accuracy of the SPECT_DL for the conventional U-Net and Wasserstein cycle GAN (WCycleGAN) networks. To manage the training data, clustering was applied to a representation of data in the lower-dimensional space, and the training data were chosen based on the similarity of data in this space. Quantitative analysis demonstrated that DL model with an advanced network improves the global performance for the AC task with the limited data. However, the regional results were not improved. The proposed data management strategy demonstrated that the clustered training has potential benefit for effective training.

Keywords: Attenuation correction; Deep learning; Hierarchical clustering; Myocardial perfusion imaging (MPI); Performance variability; SPECT; Wasserstein cycle GAN; t-SNE.

Abstract

Grants and funding