Quantitative and clinical analysis of SPECT image registration for epilepsy studies

This study reports quantitative measurements of the accuracy of two popular voxel-based registration algorithms-Woods automated image registration alg orithm and mutual information correlation-and compares these with conventio nal surface matching (SM) registration. Methods: The registration algorithm s were compared (15 different matches each) for (a) three-dimensional brain phantom images, (b) an ictal SPECT image from a patient with partial epile psy matched to itself after modification to simulate changes in the cerebra l blood flow pattern and (c) ictal/interictal SPECT images from 15 patients with partial epilepsy. Blinded visual ranking and localization of the subt raction images derived from the patient images were also performed. Results : Both voxel-based registration methods were more accurate than SM registra tion (P < 0.0005). Automated image registration algorithm was more accurate than mutual information correlation for the computer-simulated ictal/inter ictal images and the patient ictal/interictal studies (P < 0.05). The subtr action SPECTs from SM were poorer in visual ranking more often than the vox el-based methods (P < 0.05). Conclusion: Voxel intensity-based registration algorithms provide significant improvement in ictal/interictal SPECT regis tration accuracy and result in a clinically detectable improvement in the s ubtraction SPECT images.