Search

Single photon emission computed tomography (SPECT) has been employed to detect Parkinson&rsquo, s disease (PD). However, analysis of the SPECT PD images was mostly based on the region of interest (ROI) approach. Due to limited size of the ROI, especially in the multi-stage classification of PD, this study utilizes deep learning methods to establish a multiple stages classification model of PD. In the retrospective study, the 99mTc-TRODAT-1 was used for brain SPECT imaging. A total of 202 cases were collected, and five slices were selected for analysis from each subject. The total number of images was thus 1010. According to the Hoehn and Yahr Scale standards, all the cases were divided into healthy, early, middle, late four stages, and HYS I~V six stages. Deep learning is compared with five convolutional neural networks (CNNs). The input images included grayscale and pseudo color of two types. The training and validation sets were 70% and 30%. The accuracy, recall, precision, F-score, and Kappa values were used to evaluate the models&rsquo, performance. The best accuracy of the models based on grayscale and color images in four and six stages were 0.83 (AlexNet), 0.85 (VGG), 0.78 (DenseNet) and 0.78 (DenseNet).

The neuroimaging techniques such as dopaminergic imaging using Single Photon Emission Computed Tomography (SPECT) with 99mTc-TRODAT-1 have been employed to detect the stages of Parkinson&rsquo, s disease (PD). In this retrospective study, a total of 202 99mTc-TRODAT-1 SPECT imaging were collected. All of the PD patient cases were separated into mild (HYS Stage 1 to Stage 3) and severe (HYS Stage 4 and Stage 5) PD, according to the Hoehn and Yahr Scale (HYS) standard. A three-dimensional method was used to estimate six features of activity distribution and striatal activity volume in the images. These features were skewness, kurtosis, Cyhelsky&rsquo, s skewness coefficient, Pearson&rsquo, s median skewness, dopamine transporter activity volume, and dopamine transporter activity maximum. Finally, the data were modeled using logistic regression (LR) and support vector machine (SVM) for PD classification. The results showed that SVM classifier method produced a higher accuracy than LR. The sensitivity, specificity, PPV, NPV, accuracy, and AUC with SVM method were 0.82, 1.00, 0.84, 0.67, 0.83, and 0.85, respectively. Additionally, the Kappa value was shown to reach 0.68. This claimed that the SVM-based model could provide further reference for PD stage classification in medical diagnosis. In the future, more healthy cases will be expected to clarify the false positive rate in this classification model.

When using cardiac nuclear medicine images for diagnosis, the filtered back-projection (FBP) algorithm can reconstruct positron emission tomography (PET) images under low count rates. However, background strike artifacts in PET images are affected by diagnostic judgment. Hence, this study developed a robust method of removing background strike artifacts from FBP images without reducing image quality. A Jaszczak anthropomorphic torso phantom and a laboratory rabbit were used for performance tests of the proposed method. Parallel computing was applied to optimize the mask size of morphological structure operator (MSO) by minimizing the background standard deviation (Std). The optimal MSO mask size for the evaluated Jaszczak phantom was 3×3. The FBP images processed by MSO had significantly reduced strike artifacts measured by background Std ( P =1E-5). After MSO processing, the time activity curve (TAC) of FBP images was stable and resembled the original FBP images ( P =0.5). The proposed approach is highly stable and reduces noise by 13.08±2.32 in FBP images after MSO processing with 3×3 mask.

This study examines the utilization patterns of myocardial perfusion imaging (MPI) and cardiac catheterization (CC) under Taiwan’s national health insurance program. This study used the longitudinal health insurance database with 1,000,000 people were randomly selected from the national health insurance research database. This study obtained data from these patients with coronary artery disease (CAD) and comparison with the utilization of MPI or CC between 2005 and 2009. The incidence of CAD did not significantly change, while the prevalence of CAD, utilization of MPI, and the utilization of CC for the CAD patients increased annually. There were the most CAD patients in Northern Taiwan (43.5 %), followed by Southern, Central, and Eastern Taiwan. The utilizations of both of MPI (12.7 per 100 CAD patients) and CC (10.6 per 100 CAD patients) were most frequent in Northern Taiwan followed by Southern, Central, and Eastern Taiwan. However, the MPI/CC ratio was 1.20 in Northern Taiwan, followed by Southern, Central, and Eastern Taiwan (0.88, 0.64, and 0.52, respectively, P = 0.0008). The use of MPI was higher than CC only in Northern Taiwan. MPI may be underutilized to serve the role of gatekeeper for CC in the other regions.

Positron emission tomography (PET) had been utilized to image gene therapy, estimate tumor growth, detect neural function of the brain, and diagnose disease. However, sinogram noise always results inaccurate PET images. The factorial design of experiment (DOE), a statistical method, was applied to investigate, correct and estimate the fraction of scattering of 2D sinogram in PET. The DOE was included as factors of angle views and scatter media with two levels designed. The PET sinogram after scattering correction was then reconstructed by filtered back projection (FBP). Both Ge-68 uniform phantom and Jaszczak anthropomorphic torso phantom were applied to exam the performance of presented scattering correction algorithm. The signal-to-noise ratio (SNR), standard deviation (STD) of background, and full width at half maximum (FWHM), and uniformity test were applied to validate the performance of presented method. The proposed method provides a narrower FWHM, smaller STD of the background, higher SNR and better uniformity than those of original protocols. This method should be tested for accuracy and feasibility with three-dimensional phantoms or real animal studies and consideration effects of cross-talk between slices in future work.