Background and Objectives: 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) (PETFDG) image can visualize neuronal injury of the brain in Alzheimer’s disease. Early-phase amyloid PET image is reported to be similar to PETFDG image. This study aimed to generate PETFDG images from 18F-florbetaben PET (PETFBB) images using a generative adversarial network (GAN) and compare the generated PETFDG (PETGE-FDG) with real PETFDG (PETRE-FDG) images using the structural similarity index measure (SSIM) and the peak signal-to-noise ratio (PSNR). Materials and Methods: Using the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, 110 participants with both PETFDG and PETFBB images at baseline were included. The paired PETFDG and PETFBB images included six and four subset images, respectively. Each subset image had a 5 min acquisition time. These subsets were randomly sampled and divided into 249 paired PETFDG and PETFBB subset images for the training datasets and 95 paired subset images for the validation datasets during the deep-learning process. The deep learning model used in this study is composed of a GAN with a U-Net. The differences in the SSIM and PSNR values between the PETGE-FDG and PETRE-FDG images in the cycleGAN and pix2pix models were evaluated using the independent Student’s t-test. Statistical significance was set at p ≤ 0.05. Results: The participant demographics (age, sex, or diagnosis) showed no statistically significant differences between the training (82 participants) and validation (28 participants) groups. The mean SSIM between the PETGE-FDG and PETRE-FDG images was 0.768 ± 0.135 for the cycleGAN model and 0.745 ± 0.143 for the pix2pix model. The mean PSNR was 32.4 ± 9.5 and 30.7 ± 8.0. The PETGE-FDG images of the cycleGAN model showed statistically higher mean SSIM than those of the pix2pix model (p < 0.001). The mean PSNR was also higher in the PETGE-FDG images of the cycleGAN model than those of pix2pix model (p < 0.001). Conclusions: We generated PETFDG images from PETFBB images using deep learning. The cycleGAN model generated PETGE-FDG images with a higher SSIM and PSNR values than the pix2pix model. Image-to-image translation using deep learning may be useful for generating PETFDG images. These may provide additional information for the management of Alzheimer’s disease without extra image acquisition and the consequent increase in radiation exposure, inconvenience, or expenses.
Background and Objectives: The optimal assessment of cognitive function, including the impact of education, is crucial in managing Alzheimer’s disease (AD). This study aimed to evaluate the role of cognitive reserve (CR), represented by the metabolic status of regions of the cerebral cortex, to evaluate cognitive decline considering the educational attainment of patients with AD. Materials and Methods: We used data from the Alzheimer’s Disease Neuroimaging Initiative database, and selected 124 patients who underwent both baseline F-18 fluorodeoxyglucose (FDG) and F-18 florbetaben (FBB) positron emission tomography (PET) scans. Demographics, cognitive function variables (Clinical Dementia Rating—Sum of Boxes [CDR]; AD Assessment Scale 11/13 [ADAS11/13] Mini-Mental State Examination [MMSE]), and the average standardized uptake value ratio (SUVR) of cerebral cortex regions to those of the cerebellum were obtained from the data. The participants’ education level was divided into low and high education subgroups using four cut-offs of 12, 14, 16, and 18 years of educational attainment (G12, G14, G16, and G18, respectively). Demographic and cognitive function variables were compared between the two subgroups in each of the four groups, and their correlations with the SUVRs were evaluated. Results: There was no significant difference between the high and low education subgroups in each of the four groups, except for ADAS11/13 and MMSE in G14 and age in G16. The SUVRs of FDG PET (FDGSUVR) were significantly correlated with CDR, ADAS11/13, and MMSE scores. FDGSUVR showed different trajectories of neurodegeneration between the low and high education groups. Conclusions: FDGSUVR correlated moderately but significantly with neuropsychological test results, without being influenced by education level. Therefore, FDG PET may reflect CR independent of education level, and therefore could be a reliable tool to evaluate cognitive decline in AD.
28 April 2025 in Computers in Biology and Medicine
Purpose Left ventricular (LV) ejection fraction (EF) is an important parameter for assessing cardiac systolic function and predicting prognosis in patients with cardiovascular disease. The aim of this study was to evaluate the feasibility of assessing LVEF by Tl-201 hybrid myocardial single-photon emission computed tomography (SPECT)/CT using two attenuation correction methods in patients with angina pectoris. Methods A total of 339 patients with angina pectoris (62.8 ± 12.9 years, male:female = 206:133) were analyzed. All patients underwent Tl-201 myocardial SPECT/CT and transthoracic two-dimensional (2D) echocardiograph. We compared LVEF assessed by SPECT/CT using two attenuation correction methods: CT-based attenuation correction (CTAC) and non-attenuation correction (non-AC) methods and 2D echocardiography. Results LVEF assessed by either of the two attenuation correction techniques and 2D echocardiography showed moderate correlation in all patients with angina pectoris (r = 0.487 for CTAC and r = 0.473 for non-AC, p < 0.001). Results were similar in the subgroup of patients with perfusion abnormalities on myocardial SPECT/CT images. Overall diagnostic performances were similar for the CTAC and non-AC methods for evaluating normal and decreased LVEF by myocardial SPECT/CT. Conclusion LVEF measured by the CTAC method of Tl-201-gated myocardial SPECT/CT was comparable with the conventional non-AC method in patients with angina pectoris and in the subgroup of patients with perfusion abnormality. Tl-201-gated myocardial hybrid SPECT/CT can be a reliable tool in the assessment of LVEF in clinic.