The Diagnostic Value of 18 F-FDG PET in Parkinson Disease Based on Voxel Analysis

To evaluate the accuracy of statistical parametric mapping (SPM) and Scenium in the differential diagnosis of Parkinson disease (PD) and atypical Parkinsonian syndromes based on 18 F-fluoro-deoxy-glucose ( 18 F-FDG) imaging, and to explore the application of these 2 software programs in analyzing patients with Parkinson disease of varying severity, as well as to construct and evaluate the metabolic profiles of PD patients using Scenium.

A total of 64 patients with Parkinsonian syndrome who met the diagnostic criteria were included in this study. PET images were used for disease diagnosis with SPM and Scenium based on diagnostic charts, and the diagnostic accuracy of both software programs was assessed through consistency analysis. Meanwhile, an in-depth analysis was performed to compare the sensitivity, specificity, positive predictive value, and negative predictive value of the 2 software programs. In addition, Scenium was used to construct a diagnostic model for PD.

SPM demonstrated greater accuracy in distinguishing between PD and APS, with a significantly higher Kappa value (K_spm=0.704) compared with Scenium (K_scenium=0.440). The sensitivity and specificity of SPM were 82.5% and 91.7%, respectively. Further, a PD diagnostic model was constructed by incorporating PET parameters from the contralateral central region and basal ganglia, achieving a diagnostic accuracy of 82.9%.

SPM can more accurately differentiate the diagnosis of Parkinson disease from atypical Parkinson syndrome compared with Scenium.