Faster Region-Based Convolutional Neural Network in the Classification of Different Parkinsonism Patterns of the Striatum on Maximum Intensity Projection Images of [18F]FP-CIT Positron Emission Tomography

The aim of this study was to compare the performance of a deep-learning convolutional neural network (Faster R-CNN) model to detect imaging findings suggestive of idiopathic Parkinson's disease (PD) based on [F-18]FP-CIT PET maximum intensity projection (MIP) images versus that of nuclear medicine (NM) physicians. The anteroposterior MIP images of the [F-18]FP-CIT PET scan of 527 patients were classified as having PD (139 images) or non-PD (388 images) patterns according to the final diagnosis. Non-PD patterns were classified as overall-normal (ONL, 365 images) and vascular parkinsonism with definite defects or prominently decreased dopamine transporter binding (dVP, 23 images) patterns. Faster R-CNN was trained on 120 PD, 320 ONL, and 16 dVP pattern images and tested on the 19 PD, 45 ONL, and seven dVP patterns images. The performance of the Faster R-CNN and three NM physicians was assessed using receiver operating characteristics curve analysis. The difference in performance was assessed using Cochran's Q test, and the inter-rater reliability was calculated. Faster R-CNN showed high accuracy in differentiating PD from non-PD patterns and also from dVP patterns, with results comparable to those of NM physicians. There were no significant differences in the area under the curve and performance. The inter-rater reliability among Faster R-CNN and NM physicians showed substantial to almost perfect agreement. The deep-learning model accurately differentiated PD from non-PD patterns on MIP images of [F-18]FP-CIT PET, and its performance was comparable to that of NM physicians.

Automated summary

The study compared the performance of a deep-learning convolutional neural network model to detect imaging findings suggestive of idiopathic Parkinson's disease based on [F-18]FP-CIT PET images with that of nuclear medicine physicians. The results showed that the deep-learning model had high accuracy in differentiating PD from non-PD patterns, comparable to NM physicians, with substantial inter-rater reliability.