Gated SPECT-Derived Myocardial Strain Estimated From Deep-Learning Image Translation Validated From N-13 Ammonia PET.

RATIONALE AND OBJECTIVES: This study investigated the use of deep learning-generated virtual positron emission tomography (PET)-like gated single-photon emission tomography (SPECTVP) for assessing myocardial strain, overcoming limitations of conventional SPECT. MATERIALS AND METHODS: SPECT-to-PET translation models for short-axis, horizontal, and vertical long-axis planes were trained using image pairs from the same patients in stress (720 image pairs from 18 patients) and resting states (920 image pairs from 23 patients). Patients without ejection-fraction changes during SPECT and PET were selected for training. We independently analyzed circumferential strains from short-axis-gated SPECT, PET, and model-generated SPECTVP images using a feature-tracking algorithm. Longitudinal strains were similarly measured from horizontal and vertical long-axis images. Intraclass correlation coefficients (ICCs) were calculated with two-way random single-measure SPECT and SPECTVP (PET). ICCs (95% confidence intervals) were defined as excellent (≥0.75), good (0.60-0.74), moderate (0.40-0.59), or poor (≤0.39). RESULTS: Moderate ICCs were observed for SPECT-derived stressed circumferential strains (0.56 [0.41-0.69]). Excellent ICCs were observed for SPECTVP-derived stressed circumferential strains (0.78 [0.68-0.85]). Excellent ICCs of stressed longitudinal strains from horizontal and vertical long axes, derived from SPECT and SPECTVP, were observed (0.83 [0.73-0.90], 0.91 [0.85-0.94]). CONCLUSION: Deep-learning SPECT-to-PET transformation improves circumferential strain measurement accuracy using standard-gated SPECT. Furthermore, the possibility of applying longitudinal strain measurements via both PET and SPECTVP was demonstrated. This study provides preliminary evidence that SPECTVP obtained from standard-gated SPECT with postprocessing potentially adds clinical value through PET-equivalent myocardial strain analysis without increasing the patient burden.

Deep learning approach using SPECT-to-PET translation for attenuation correction in CT-less myocardial perfusion SPECT imaging.

OBJECTIVE: Deep learning approaches have attracted attention for improving the scoring accuracy in computed tomography-less single photon emission computed tomography (SPECT). In this study, we proposed a novel deep learning approach referring to positron emission tomography (PET). The aims of this study were to analyze the agreement of representative voxel values and perfusion scores of SPECT-to-PET translation model-generated SPECT (SPECTSPT) against PET in 17 segments according to the American Heart Association (AHA). METHODS: This retrospective study evaluated the patient-to-patient stress, resting SPECT, and PET datasets of 71 patients. The SPECTSPT generation model was trained (stress: 979 image pairs, rest: 987 image pairs) and validated (stress: 421 image pairs, rest: 425 image pairs) using 31 cases of SPECT and PET image pairs using an image-to-image translation network. Forty of 71 cases of left ventricular base-to-apex short-axis images were translated to SPECTSPT in the stress and resting state (stress: 1830 images, rest: 1856 images). Representative voxel values of SPECT and SPECTSPT in the 17 AHA segments against PET were compared. The stress, resting, and difference scores of 40 cases of SPECT and SPECTSPT were also compared in each of the 17 segments. RESULTS: For AHA 17-segment-wise analysis, stressed SPECT but not SPECTSPT voxel values showed significant error from PET at basal anterior regions (segments #1, #6), and at mid inferoseptal regions (segments #8, #9, and #10). SPECT, but not SPECTSPT, voxel values at resting state showed significant error at basal anterior regions (segments #1, #2, and #6), and at mid inferior regions (segments #8, #9, and #11). Significant SPECT overscoring was observed against PET in basal-to-apical inferior regions (segments #4, #10, and #15) during stress. No significant overscoring was observed in SPECTSPT at stress, and only moderate over and underscoring in the basal inferior region (segment #4) was found in the resting and difference states. CONCLUSIONS: Our PET-supervised deep learning model is a new approach to correct well-known inferior wall attenuation in SPECT myocardial perfusion imaging. As standalone SPECT systems are used worldwide, the SPECTSPT generation model may be applied as a low-cost and practical clinical tool that provides powerful auxiliary information for the diagnosis of myocardial blood flow.

Optimization of Intraventricular Radioactive Concentration for 13N ammonia PET with Time-of-Flight Scanner: Simplified Phantom Study with Noise Equivalent Count Rate Analysis.

Background: Myocardial blood flow quantification (MBF) is one of the distinctive features for cardiac positron emission tomography. The MBF calculation is mostly obtained by estimating the input function from the time activity curve in dynamic scan. However, there is a substantial risk of count-loss because the high radioactivity pass through the left ventricular (LV) cavity within a short period. We aimed to determine the optimal intraventricular activity using the noise equivalent count rate (NECR) analysis with simplified phantom model. Methods: Positron emission tomography computed tomography scanner with LYSO crystal and time of flight was used for phantom study. 150 MBq/mL of 13N was filled in 10 mL of syringe, placed in neck phantom to imitate end-systolic small LV. 3D list-mode acquisition was repeatedly performed along radioactive decay. Net true and random count rate were calculated and compared to the theoretical activity in the syringe. NECR curve analysis was used to determine the optimal radioactive concentration. Result: The attenuation curves showed good correlation to the theoretical activity between 20 to 370, and 370 to 740 MBq (r2=1.0 ± 0.0001, p<0.0001; r2=0.99 ± 0.0001, p<0.0001 for 20 to 370, and 370 to 740, respectively), while did not over 740 MBq (p=0.62). NECR analysis revealed that the peak rate was at 2.9 Mcps, there at the true counts were significantly suppressed. The optimal radioactive concentration was determined as 36 MBq/mL. Conclusion: Simulative analysis for high-dose of 13N using the phantom imitating small LV confirmed that the risk of count-loss was increased. The result can be useful information in assessing the feasibility of MBF quantification in clinical routine.

Imaging of Heart Type Fatty Acid Binding Protein Under Acute Reperfusion Ischemia Using Radio-labeled Antibody in Rat Heart Model.

Purpose: Heart-type fatty acid binding protein (H-FABP) is primary transporter of free fatty acid and plays an important role in myocardial metabolism, which is characterized by high specificity and rapid appearance under ischemic condition. The objective of this study was to clarify the usefulness of imaging study of targeting H-FABP appearance using radio-labeled antibody, and correlation with myocardial fatty acid metabolism and perfusion in acute reperfusion ischemia. Method: Wistar rats were allotted to sham-operated control group (sham; n=4), ischemia non-reperfused group (IG; n=5), and ischemia-reperfusion group (RG; n=5). Ligation of left coronary artery (LCA) was performed for IG and RG. 20 min of ischemia was followed by 60min of reperfusion for RG. 125I labeled anti H-FABP antibody (anti H-FABP), BMIPP and 99mTc-sestamibi (MIBI) was injected intravenously. Multi-tracer digital autoradiogram was performed using µ-imager®. The ratio of radioactivity in LCA related (culprit) area to the inferior (remote) area (target uptake ratio=TUR) was generated. Results: In sham group, no visually detectable accumulation was observed for the anti H-FABP image, and TURMIBI and TURBMIPP were equivalent to 1. In IG, TURMIBI and TURBMIPP were remarkably low (0.12±0.01, 0.24±0.07). In RG, TURMIBI was significantly lower (0.20±0.03, p<0.05 vs. other groups). However, TURBMIPP was significantly higher (2.78±1.28, p<0.05) compared to the sham and IG, whereas anti H-FABP showed markedly higher ratio in the reperfused area compared to the sham and IG (3.43±0.73 vs. 0.31±0.13 and 1.09±0.07 for IG and sham; p<0.05, and <0.01, respectively). Conclusion: Anti H-FABP accumulated specifically in reperfused area under acute ischemia, and it accorded to the area where fatty acid metabolism was activated. This study has shown the future potential for clinical application in vivo imaging of acute coronary syndrome.