Deep learning-based attenuation correction method in 99m Tc-GSA SPECT/CT hepatic imaging: a phantom study.

This study aimed to evaluate a deep learning-based attenuation correction (AC) method to generate pseudo-computed tomography (CT) images from non-AC single-photon emission computed tomography images (SPECT _NC ) for AC in ^99m Tc-galactosyl human albumin diethylenetriamine pentaacetic acid (GSA) scintigraphy and to reduce patient dosage. A cycle-consistent generative network (CycleGAN) model was used to generate pseudo-CT images. The training datasets comprised approximately 850 liver phantom images obtained from SPECT _NC and real CT images. The training datasets were then input to CycleGAN, and pseudo-CT images were output. SPECT images with real-time CT attenuation correction (SPECT _CTAC ) and pseudo-CT attenuation correction (SPECT _GAN ) were acquired. The difference in liver volume between real CT and pseudo-CT images was evaluated. Total counts and uniformity were then used to evaluate the effects of AC. Additionally, the similarity coefficients of SPECT _CTAC and SPECT _GAN were assessed using a structural similarity (SSIM) index. The pseudo-CT images produced a lower liver volume than the real CT images. SPECT _CTAC exhibited a higher total count than SPECT _NC and SPECT _GAN , which were approximately 60% and 7% lower, respectively. The uniformities of SPECT _CTAC and SPECT _GAN were better than those of SPECT _NC . The mean SSIM value for SPECT _CTAC and SPECT _GAN was 0.97. We proposed a deep learning-based AC approach to generate pseudo-CT images from SPECT _NC images in ^99m Tc-GSA scintigraphy. SPECT _GAN with AC using pseudo-CT images was similar to SPECT _CTAC , demonstrating the possibility of SPECT/CT examination with reduced exposure to radiation.
(© 2023. The Author(s), under exclusive licence to Japanese Society of Radiological Technology and Japan Society of Medical Physics.)