MetaAD
Metabolism-aware anomaly detection for Parkinson's disease in 3D 18F-FDG PET.
MetaAD highlights abnormal metabolic cues of Parkinson’s disease in 18F-FDG PET. It learns healthy metabolic characteristics through cyclic modality translation and uses the difference between the input and reconstructed FDG images as an interpretable anomaly signal.
The work was published at MICCAI 2024 and received the Young Scientist Award.