Objective: To include probabilistic information on the location of the middle cerebellar peduncle (MCP) into an existing subcortical segmentation atlas and to evaluate the diagnostic potential of the novel atlas in the differential diagnosis of multiple system atrophy (MSA) and Parkinson’s disease (PD).

Background: Previous studies reported that reduced MCP width is associated with a high specificity for a clinical diagnosis of MSA. These studies were based on manual segmentation of the MCP. Recently, automatic, investigator-independent segmentation algorithms were developed and evaluated in neurodegenerative disorders.

Methods: 48 healthy individuals (HC) without evidence of any neurological disorder on careful clinical examination underwent conventional 3T MRI and were used for the development of the novel segmentation atlas. The MCP was identified and segmented observer-independently by the help of the infratentorial probabilistic white matter atlas and integrated into FreeSurfer – a fully automated whole brain MRI segmentation procedure (www.freesurfer.net). To evaluate the diagnostic potential of automated segmentation of subcortical regions by the help of the novel atlas 32 MSA patients (19 MSA-P, 13 MSA-C) and 19 PD patients were selected from our MRI database based on the following inclusion criteria: (1) a clinical diagnosis of probable MSA or PD at the last visit according to diagnostic criteria; (2) a clinical follow-up of ≥24 months; (3) a disease duration of ≤5 years. Exclusion criteria were dementia, extensive white matter lesions and vascular or space-occupying lesions within the cerebrum or motion artefacts on MRI.

Results: There was no significant difference of gender distribution (p=0.885), age (p=0.155) and disease duration (p=0.992) among subject groups. Motor impairment measured by the Hoehn & Yahr staging and the UPDRS III was greater in MSA patients as compared with PD patients (p = 0.001). By applying a C4.5 classifier, atrophy of the putamen and atrophy of the middle cerebellar peduncle were selected as most useful measures in differentiating MSA from PD. Cross-validation provides support for the generalizability of the model with a weighted average F-measure and a Cohen’s kappa of 0.901 and 0.788, respectively.

Conclusions: 1) We successfully developed a segmentation atlas containing probabilistic information on the location of the MCP and (2) we were able to demonstrate an excellent diagnostic accuracy of automatic MCP and putamen volumetry in variants of MSA.

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/diagnostic-potential-of-automated-middle-cerebellar-peduncle-volumetry-in-variants-of-multiple-system-atrophy/