Automatic Striatal ROI Delineation for Semi-quantitative Analysis of 99mTc-TRODAT-1 Brain SPECT Imaging

Purpose: Single-photon emission computed tomography (SPECT) imaging, a relatively inexpensive nuclear imaging technique, has been used as a diagnostic tool for Parkinson’s disease (PD). The tracer 99mTc-TRODAT-1 is commonly used in Taiwan, where the imaging cost is covered under the national health insurance system. However, poor TRODAT-SPECT imaging resolution is a problem. Therefore, we developed a semi-quantitative program that can perform automatic region of interest (ROI) delineation for semi-quantifying striatal dopamine transporter activities. Methods: A total of 228 subjects were enrolled in this study, including PD patients (n = 62) and healthy controls (n = 166). The subjects received a 99mTc-TRODAT-1 brain SPECT scan. Some of the subjects (n = 40) also received a high-resolution magnetic resonance imaging (MRI) scan for the evaluation of the accuracy of the semi-quantitative assessment. In addition, the developed automatic ROI delineation method was used to calculate the striatal specific uptake ratio (SUR) in dopamine transporter imaging. The reproducibility of the SUR obtained using the method was compared to those in previous studies. Results: Based on linear regression analysis (r2 = 0.92, correlation coefficient = 0.96), a correlation was found between the results obtained using the automatic ROI delineation method and those of the manual method in striatum with the corresponding MR images. The age-related decline of SUR availability measured in this study was 6.4% per decade. There was almost no difference between our results and those in related clinical studies. Conclusions: The proposed automatic ROI delineation program for SUR analysis was tested and used to estimate the ROIs in TRODAT-SPECT images to aid PD diagnosis. The program could be helpful for physicians who are inexperienced with TRODAT-SPECT images.