Application of the anatomical fiducials framework to a clinical dataset of patients with Parkinson’s disease

Abstract

Establishing spatial correspondence between subject and template images is necessary in neuroimaging research and clinical applications such as brain mapping and stereotactic neurosurgery. Our anatomical fiducial (AFID) framework has recently been validated to serve as a quantitative measure of image registration based on salient anatomical features. In this study, we sought to apply the AFIDs protocol to the clinic, focusing on structural magnetic resonance images obtained from patients with Parkinson’s disease (PD). We confirmed AFIDs could be placed to millimetric accuracy in the PD dataset with results comparable to those in normal control subjects. We evaluated subject-to-template registration using this framework by aligning the clinical scans to standard template space using a robust open preprocessing workflow. We found that registration errors measured using AFIDs were higher than previously reported, suggesting the need for optimization of image processing pipelines for clinical grade datasets. Finally, we examined the utility of using point-to-point distances between AFIDs as a morphometric biomarker of PD, finding evidence of reduced distances between AFIDs that circumscribe regions known to be affected in PD including the substantia nigra. Overall, we provide evidence that AFIDs can be successfully applied 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 and the location of anatomical structures, facilitating aggregation of imaging datasets and comparisons between various neurological conditions.

The impact of localization and registration accuracy on estimates of deep brain stimulation electrode position in stereotactic space

“…A Wilcoxon signed-rank test revealed no significant differences between any LEs in all axes. Furthermore, we expanded our analysis to a subset of 24 patients who previously had AFIDs placed by five raters (Abbass et al, 2022; Lau et al, 2019; Taha et al, 2023). The median AFLE (IQR) ranged from 0.56 mm (0.40-0.69 mm) for AFID02 (PC) to 2.25 mm (1.49-2.75 mm) for AFID25 (right inferior anteromedial temporal horn).…”

“…We previously reported and openly released AFIDs annotations in a population of thirty-nine patients undergoing STN DBS for PD (Abbass et al, 2022; Taha et al, 2023). Briefly, 5 raters localized AFIDs on subject scans and on the MNI template.…”

“…The median AFLE (IQR) ranged from 0.56 mm (0.40-0.69 mm) for AFID02 (PC) to 2.25 mm (1.49-2.75 mm) for AFID25 (right inferior anteromedial temporal horn). See Abbass et al (2022) for more details.…”

“…These voxel overlap measures are straightforward to obtain from common neuroimaging workflows, but are also relatively coarse metrics that do not capture focal misregistration (Rohlfing, 2012). Anatomically placed points (also referred to as fiducials or landmarks) can also be used to quantify registration accuracy measured as the millimetric distance between transformed points (Abbass et al, 2022; Lau et al, 2019; Schönecker et al, 2009). This motivated our group to identify and validate anatomical fiducials (AFIDs) which can be accurately and reliably placed.…”

Automatic Segmentation of Parkinson Disease Therapeutic Targets Using Nonlinear Registration and Clinical MR Imaging: Comparison of Methodology, Presence of Disease, and Quality Control

“…Alternatively, Abbass et al recently reported utilization of anatomical fiducials in the brain for such a purpose. However, this method includes the placement of 32 fiducial points which is impractical for clinical application [35]. Therefore, as automated segmentation tools continue to evolve, efficient and reliable methods of QC are needed to make them more clinically relevant.…”