Lung-to-heart ratio analysis using virtual planar images obtained from myocardial perfusion SPECT data: A phantom and clinical studies.

BACKGROUNDS: The lung-to-heart ratio (L/H ratio) in myocardial perfusion scintigraphy (MPS) is a useful marker that complements the sensitivity of ischemia detection. However, it requires planar imaging acquired following a separate protocol in addition to single-photon emission computed tomography (SPECT). We developed a novel method for constructing virtual planar image (VPI) from SPECT data. METHODS: Myocardial phantoms using Tl-201 were built with different amounts of radioactivity in the lungs. SPECT data and conventional planar images of these phantoms were collected with an Anger-type gamma camera. VPIs were constructed by adding all coronal images reconstructed from SPECT data. The clinical utility of VPIs obtained from 52 patients who underwent MPS with Tc-99m sestamibi was evaluated. RESULTS: The radioactivity linearity of VPIs was satisfactory, with a correlation coefficient of r ≥ .99 between the measured amounts of radioactivity and image counts. The L/H ratios obtained from VPI analysis were strongly correlated with those of conventional planar images with a correlation coefficient of r ≥ .99 in the phantom study and r = .929 in clinical application. CONCLUSION: The accuracy of VPI-based L/H ratio analysis was comparable to that of conventional planar image-based analysis. VPIs could be used as an alternative method of obtaining planar images in clinical settings.

Generative adversarial network-based post-processed image super-resolution technology for accelerating brain MRI: comparison with compressed sensing.

BACKGROUND: It is unclear whether deep-learning-based super-resolution technology (SR) or compressed sensing technology (CS) can accelerate magnetic resonance imaging (MRI) . PURPOSE: To compare SR accelerated images with CS images regarding the image similarity to reference 2D- and 3D gradient-echo sequence (GRE) brain MRI. MATERIAL AND METHODS: We prospectively acquired 1.3× and 2.0× faster 2D and 3D GRE images of 20 volunteers from the reference time by reducing the matrix size or increasing the CS factor. For SR, we trained the generative adversarial network (GAN), upscaling the low-resolution images to the reference images with twofold cross-validation. We compared the structural similarity (SSIM) index of accelerated images to the reference image. The rate of incorrect answers of a radiologist discriminating faster and reference image was used as a subjective image similarity (ISM) index. RESULTS: The SR demonstrated significantly higher SSIM than the CS (SSIM=0.9993-0.999 vs. 0.9947-0.9986; P < 0.001). In 2D GRE, it was challenging to discriminate the SR image from the reference image, compared to the CS (ISM index 40% vs. 17.5% in 1.3×; P = 0.039 and 17.5% vs. 2.5% in 2.0×; P = 0.034). In 3D GRE, the CS revealed a significantly higher ISM index than the SR (22.5% vs. 2.5%; P = 0.011) in 2.0 × faster images. However, the ISM index was identical for the 2.0× CS and 1.3× SR (22.5% vs. 27.5%; P = 0.62) with comparable time costs. CONCLUSION: The GAN-based SR outperformed CS in image similarity with 2D GRE for MRI acceleration. In addition, CS was more advantageous in 3D GRE than SR.

[Development of Body Phantom for Evaluation of Appropriate Administered Radioactivities and Image Quality on 99mTc-DMSA Scintigraphy in Pediatric Nuclear Medicine].

PURPOSE: We conducted a field survey about pediatric nuclear medicine. As a result, it was suggested that 99mTc-DMSA scintigraphy was performed at many institutions, whereas various examinations such as image acquisition and processing are not carried out using the renal phantom. Therefore, we developed the body phantom for the evaluation of appropriate administered radioactivities and image quality with renal scintigraphy in pediatric nuclear medicine. METHODS: We created three differently sized body phantoms (1-, 5-, and 20-year-old models). These pediatric body phantoms were filled with a 99mTc solution based on the consensus guideline of pediatric radiopharmaceutical administered radioactivity in Japan. The planar image was evaluated using acquisition count, uniformity and defect contrast. SPECT images were evaluated with a recovery coefficient (RC). RESULTS: The acquisition counts for pediatric body phantoms were relatively corresponded to the clinical study. The appropriate acquisition counts and the pixel size for the planar image were approximately 140 counts per pixel and 1.23-1.35 mm at 5 min acquisition times in 1- and 5-year-old pediatric body phantom studies, respectively. Although the uniformity and the cold contrast did not depend on pixel size and body size, the cold contrast was affected by body size. The RC for SPECT images depended on the performance of SPECT systems, the resolution recovery algorithm and body phantom size. CONCLUSION: The developed pediatric body phantom could allow us to establish optimal image acquisition and more evidence on renal scintigraphy in pediatric nuclear medicine.

[Usefulness of attenuation correction with transmission source in myocardial SPECT].

Attenuation correction in SPECT has been used for uniformly absorptive objects like the head. On the other hand, it has seldom been applied to nonuniform absorptive objects like the heart and surrounding lungs because of the difficulty and inaccuracy of data processing. However, since attenuation correction using a transmission source recently became practical, we were able to apply this method to a nonuniform absorptive object. Therefore, we evaluated the usefulness of this attenuation correction system with a transmission source in myocardial SPECT. The dose linearity, defect/normal ratio using a myocardial phantom, and myocardial count distribution in clinical cases was examined with and without the attenuation correction system. We found that all data processed with attenuation correction were better than those without attenuation correction. For example, in myocardial count distribution, while there was a difference between men and women without attenuation correction, which was considered to be caused by differences in body shape, after processing with attenuation correction, myocardial count distribution was almost the same in all cases. In conclusion, these results suggested that attenuation correction with a transmission source was useful in myocardial SPECT.