Deep learning-guided attenuation correction in the image domain for myocardial perfusion SPECT imaging

Abstract

We investigate the accuracy of direct attenuation correction (AC) in the image domain for myocardial perfusion SPECT (single-photon emission computed tomography) imaging (MPI-SPECT) using residual (ResNet) and UNet deep convolutional neural networks. MPI-SPECT 99mTc-sestamibi images of 99 patients were retrospectively included. UNet and ResNet networks were trained using non-attenuation-corrected SPECT images as input, whereas CT-based attenuation-corrected (CT-AC) SPECT images served as reference. Chang’s calculated AC approach considering a uniform attenuation coefficient within the body contour was also implemented. Clinical and quantitative evaluations of the proposed methods were performed considering SPECT CT-AC images of 19 subjects (external validation set) as reference. Image-derived metrics, including the voxel-wise mean error (ME), mean absolute error, relative error, structural similarity index (SSI), and peak signal-to-noise ratio, as well as clinical relevant indices, such as total perfusion deficit (TPD), were utilized. Overall, AC SPECT images generated using the deep learning networks exhibited good agreement with SPECT CT-AC images, substantially outperforming Chang’s method. The ResNet and UNet models resulted in an ME of −6.99 ± 16.72 and −4.41 ± 11.8 and an SSI of 0.99 ± 0.04 and 0.98 ± 0.05, respectively. Chang’s approach led to ME and SSI of 25.52 ± 33.98 and 0.93 ± 0.09, respectively. Similarly, the clinical evaluation revealed a mean TPD of 12.78 ± 9.22% and 12.57 ± 8.93% for ResNet and UNet models, respectively, compared to 12.84 ± 8.63% obtained from SPECT CT-AC images. Conversely, Chang’s approach led to a mean TPD of 16.68 ± 11.24%. The deep learning AC methods have the potential to achieve reliable AC in MPI-SPECT imaging.

The impact of MR-guided attenuation correction (compared to CT-based AC) on the diagnosis of anosmia based on 99m-Tc Ethyl-Cysteinate-Dimer SPECT images

“…They also implemented Chang's attenuation correction method (which is equivalent to the 2-class attenuation map) to provide a baseline for the assessment of the deep learning approach. The clinical evaluation, as well as quantitative results, demonstrated excellent performance of the deep learning approach (negligible bias) while systematic activity concentration bias was observed when Chang's method was used (43) .…”

Deep learning-enhanced nuclear medicine SPECT imaging applied to cardiac studies

“…Mostafapour et al [ 50 ] investigated the generated attenuation-corrected images utilizing ResNet and U-Net. This research enrolled 99 patient cases.…”

Deep learning-based synthetic CT generation from MR images: comparison of generative adversarial and residual neural networks

“…Although many approaches/algorithms have been proposed in the previous works concerning the generation of synthetic (pseudo) CT images from MRI data (6-8, 13, 14 ), deep learning-based approaches are of special interest owing to their promising and superior performance (10,26) . Among the various deep learning models, GAN and residual deep learning models are frequently used for different purposes in clinical and research settings (25)(26)(27)(28)(29) . The major aim of this study was to compare the two popular deep learning models for the challenging task of MR-guided synthetic CT generation related to their application in MR-only radiation planning (11,19) and MR-guided PET attenuation correction (11) .…”