Category: Parkinson's disease: Neuroimaging
Objective: To determine whether AI diagnostic performance for Parkinson’s disease differs across PET imaging platforms
Background: PET-based artificial intelligence (AI) has emerged as a promising approach for the diagnosis of Parkinson’s disease (PD). Whether diagnostic performance differs by imaging platform, however, remains unclear.
Method: We conducted a systematic review and diagnostic test accuracy meta-analysis of studies evaluating PET-based AI models for distinguishing PD from NC. Pooled sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were estimated.
Results: Fifteen studies met the inclusion criteria. Overall, PET-based AI showed excellent accuracy for differentiating PD from NC, with pooled sensitivity of 0.93 (95% CI, 0.89–0.96), pooled specificity of 0.92 (95% CI, 0.86–0.95), and AUC of 0.97 (95% CI, 0.95–0.98). No significant differences in sensitivity or specificity were observed across PET-only, PET/CT, and PET/MRI, and no platform demonstrated clear diagnostic superiority.
Conclusion: PET-based AI provides excellent discrimination between PD and NC, and its performance appears broadly consistent across PET-only, PET/CT, and PET/MRI platforms. These findings suggest that the PET imaging modality is not a major determinant of AI diagnostic accuracy in this setting. Variation in performance may instead be more strongly influenced by factors such as model architecture, tracer selection, multicenter design, and integration of clinical features.
References: Shen Z, Wang J, Huang H, Lu J, Ge J, Xiong H, et al. Cross-modality PET image synthesis for Parkinson’s disease diagnosis: a leap from [18F]FDG to [11C]CFT. Eur J Nucl Med Mol Imaging. 2025;52:1566–1575. doi:10.1007/s00259-025-07096-3
Lu W, Song T, Li J, Zhang Y, Lu J. Individual-specific metabolic network based on 18F-FDG PET revealing multi-level aberrant metabolisms in Parkinson’s disease. Hum Brain Mapp. 2024;45:e70026. doi:10.1002/hbm.70026
To cite this abstract in AMA style:
MH. Elkasaby, MSA. Ahmed. Does PET Imaging Modality Influence Artificial Intelligence Performance in Diagnosing Parkinson’s Disease? Diagnostic Test Accuracy Meta-analysis [abstract]. Mov Disord. 2026; 41 (suppl 1). https://www.mdsabstracts.org/abstract/does-pet-imaging-modality-influence-artificial-intelligence-performance-in-diagnosing-parkinsons-disease-diagnostic-test-accuracy-meta-analysis/. Accessed October 1, 2026.« Back to 2026 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/does-pet-imaging-modality-influence-artificial-intelligence-performance-in-diagnosing-parkinsons-disease-diagnostic-test-accuracy-meta-analysis/