Related Experiment Video
Updated: Jul 20, 2025
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Brain MRI in Progressive Supranuclear Palsy with Richardson's Syndrome and Variant Phenotypes
Mike P Wattjes1, Hans-Jürgen Huppertz2, Nima Mahmoudi1
1Department of Neuroradiology, Hannover Medical School, Hannover, Germany.
Background:
Brain magnetic resonance imaging (MRI) is used to support the diagnosis of progressive supranuclear palsy (PSP). However, the value of visual descriptive, manual planimetric, automatic volumetric MRI markers and fully automatic categorization is unclear, particularly regarding PSP predominance types other than Richardson's syndrome (RS).
Objectives:
To compare different visual reading strategies and automatic classification of T1-weighted MRI for detection of PSP in a typical clinical cohort including PSP-RS and (non-RS) variant PSP (vPSP) patients.
Methods:
Forty-one patients (21 RS, 20 vPSP) and 46 healthy controls were included. Three readers using three strategies performed MRI analysis: exclusively visual reading using descriptive signs (hummingbird, morning-glory, Mickey-Mouse), visual reading supported by manual planimetry measures, and visual reading supported by automatic volumetry. Fully automatic classification was performed using a pre-trained support vector machine (SVM) on the results of atlas-based volumetry.
Results:
All tested methods achieved higher specificity than sensitivity. Limited sensitivity was driven to large extent by false negative vPSP cases. Support by automatic volumetry resulted in the highest accuracy (75.1% ± 3.5%) among the visual strategies, but performed not better than the midbrain area (75.9%), the best single planimetric measure. Automatic classification by SVM clearly outperformed all other methods (accuracy, 87.4%), representing the only method to provide clinically useful sensitivity also in vPSP (70.0%).
Conclusions:
Fully automatic classification of volumetric MRI measures using machine learning methods outperforms visual MRI analysis without and with planimetry or volumetry support, particularly regarding diagnosis of vPSP, suggesting the use in settings with a broad phenotypic PSP spectrum. © 2023 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.
Insights
Fully automatic MRI classification using machine learning significantly improves progressive supranuclear palsy (PSP) detection, especially for variant PSP (vPSP) cases, outperforming visual and manual methods.
Area of Science:
- Neuroimaging
- Machine Learning
- Neurology
Background:
- Brain magnetic resonance imaging (MRI) aids progressive supranuclear palsy (PSP) diagnosis.
- The effectiveness of various MRI markers and classification methods, especially for non-Richardson's syndrome (non-RS) PSP variants, remains unclear.
Purpose of the Study:
- To compare visual, manual planimetric, and automatic volumetric MRI analysis strategies for PSP detection.
- To evaluate fully automatic classification of MRI data for identifying PSP, including Richardson's syndrome (RS) and variant PSP (vPSP) subtypes.
Main Methods:
- T1-weighted MRI scans from 41 PSP patients (21 RS, 20 vPSP) and 46 controls were analyzed.
- Three visual reading strategies were employed: descriptive signs, manual planimetry, and automatic volumetry.
- A support vector machine (SVM) classifier used atlas-based volumetry for fully automatic PSP classification.
Main Results:
- All methods showed higher specificity than sensitivity, with vPSP cases often yielding false negatives.
- Automatic volumetry support yielded the highest accuracy (75.1%) among visual strategies.
- Fully automatic SVM classification achieved superior accuracy (87.4%) and provided clinically useful sensitivity (70.0%) for vPSP.
Conclusions:
- Fully automatic classification of volumetric MRI data using machine learning surpasses visual analysis for PSP diagnosis.
- This automated approach is particularly effective for vPSP, suggesting its utility in diverse PSP presentations.
- Machine learning-based MRI analysis offers a promising tool for diagnosing PSP across its phenotypic spectrum.
More Related Videos
Related Concept Videos
01:37Neural Regulation
01:24Magnetic Resonance Imaging