P.167 Application of the Anatomical Fiducials Framework to a Clinical Dataset of Patients with Parkinson's Disease. (November 2021)

• Record Type: Journal Article
• Authors: Abbass, M; Gilmore, G; Taha, A; Chevalier, R; Jach, M; Peters, TM; Khan, AR; Lau, JC
• Is Part Of: Canadian journal of neurological sciences. Volume 48(2021)Supplement S3
• Page Start: S67
• Page End: S68
• Publication Date: 2021-11
• Subjects: Neurology -- Periodicals; Nervous system -- Surgery -- Periodicals; Electronic journals; 616.8
• DOI: 10.1017/cjn.2021.443
• Languages: English
• ISSNs: 0317-1671

Abstract

Abstract : Background: Establishing spatial correspondence between subject and template images is necessary in neuroimaging research and clinical applications. A point-based set of anatomical fiducials (AFIDs) was recently developed and validated to provide quantitative measures of image registration. We applied the AFIDs protocol to magnetic resonance images (MRIs) obtained from patients with Parkinson's Disease (PD). Methods: Two expert and three novice raters placed AFIDs on MRIs of 39 PD patients. Localization and registration errors were calculated. To investigate for unique morphometric features, pairwise distances between AFIDs were calculated and compared to 30 controls who previously had AFIDs placed. Wilcoxon rank-sum tests with Bonferroni corrections were used. Results: 6240 AFIDs were placed with a mean localization error (±SD) of 1.57mm±1.16mm and mean registration error of 3.34mm±1.94mm. Out of the 496 pairwise distances, 40 were statistically significant (p<0.05/496). PD patients had a decreased pairwise distance between the left temporal horn, brainstem and pineal gland. Conclusions: AFIDs can be successfully applied with millimetric accuracy in a clinical setting and utilized to provide localized and quantitative measures of registration error. AFIDs provide clinicians and researchers with a common, open framework for quality control and validation of spatial correspondence, facilitating accurate aggregation of imaging datasets and comparisons between various neurological conditions.