ABSTRACT
BACKGROUND: To generate high-quality bone scan SPECT images from only 1/7 scan time SPECT images using deep learning-based enhancement method. MATERIALS AND METHODS: Normal-dose (925-1110 MBq) clinical technetium 99 m-methyl diphosphonate (99mTc-MDP) SPECT/CT images and corresponding SPECT/CT images with 1/7 scan time from 20 adult patients with bone disease and a phantom were collected to develop a lesion-attention weighted U2-Net (Qin et al. in Pattern Recognit 106:107404, 2020), which produces high-quality SPECT images from fast SPECT/CT images. The quality of synthesized SPECT images from different deep learning models was compared using PSNR and SSIM. Clinic evaluation on 5-point Likert scale (5 = excellent) was performed by two experienced nuclear physicians. Average score and Wilcoxon test were constructed to assess the image quality of 1/7 SPECT, DL-enhanced SPECT and the standard SPECT. SUVmax, SUVmean, SSIM and PSNR from each detectable sphere filled with imaging agent were measured and compared for different images. RESULTS: U2-Net-based model reached the best PSNR (40.8) and SSIM (0.788) performance compared with other advanced deep learning methods. The clinic evaluation showed the quality of the synthesized SPECT images is much higher than that of fast SPECT images (P < 0.05). Compared to the standard SPECT images, enhanced images exhibited the same general image quality (P > 0.999), similar detail of 99mTc-MDP (P = 0.125) and the same diagnostic confidence (P = 0.1875). 4, 5 and 6 spheres could be distinguished on 1/7 SPECT, DL-enhanced SPECT and the standard SPECT, respectively. The DL-enhanced phantom image outperformed 1/7 SPECT in SUVmax, SUVmean, SSIM and PSNR in quantitative assessment. CONCLUSIONS: Our proposed method can yield significant image quality improvement in the noise level, details of anatomical structure and SUV accuracy, which enabled applications of ultra fast SPECT bone imaging in real clinic settings.
ABSTRACT
Quantitative susceptibility mapping (QSM) is used to quantify iron deposition in non-human primates in our study. Although QSM has many applications in detecting iron deposits in the human brain, including the distribution of iron deposits in specific brain regions, the change of iron deposition with aging, and the comparison of iron deposits between diseased groups and healthy controls, few studies have applied QSM to non-human primates, while most animal brain experiments focus on biochemical and anatomical results instead of non-invasive experiments. Additionally, brain imaging in children's research is difficult, but can be substituted using young rhesus monkeys, which are very similar to humans, as research animals. Therefore, understanding the relationship between iron deposition and age in rhesus macaques' brains can offer insights into both the developmental trajectory of magnetic susceptibility in the animal model and the correlated evidence in children's research. Twenty-three healthy rhesus macaque monkeys (23 ± 7.85 years, range 2-29 years) were included in this research. Seven regions of interest (ROIs-globus pallidus, substantia nigra, dentate nucleus, caudate nucleus, putamen, thalamus, red nucleus) have been analyzed in terms of QSM and R2 * (apparent relaxation rate). Susceptibility in most ROIs correlated significantly with the growth of age, similarly to the results for R2 *, but showed different trends in the thalamus and red nucleus, which may be caused by the different sensitivities of myelination and iron deposition in R2 * and QSM analysis. By assessing the correlation between iron content and age in healthy rhesus macaques' brains using QSM, we provide a piece of pilot information on normality for advanced animal disease models. Meanwhile, this study also could serve as the normative basis for further clinical studies using QSM for iron content quantification. Due to the comparison of the susceptibility on the same experimental objects, this research can also provide practical support for future research on characteristics for QSM and R2 *.
