A Methodology to Validate MRI/SPECT Registration Methods Using Realistic Simulated SPECT Data
MICCAI '01 Proceedings of the 4th International Conference on Medical Image Computing and Computer-Assisted Intervention
Hi-index | 0.03 |
This paper describes an unsupervised method to help detection of significant functional inter-hemispheric asymmetries in brain SPECT. A validation of this method was performed with realistic simulated SPECT data sets with known asymmetries (in size and amplitude) as gold standard. Detection performances were assessed through Receiver Operating Characteristic (ROC) analysis based on measures of the overlap between known and detected asymmetries.