Related Experiment Video
Updated: Apr 15, 2026
13:12
Translational Brain Mapping at the University of Rochester Medical Center: Preserving the Mind Through Personalized Brain Mapping
Published on: August 12, 2019
46.8K
Improving fMRI reliability in presurgical mapping for brain tumours
M Tynan R Stevens1, David B Clarke2, Gerhard Stroink3
1Department of Physics, Dalhousie University, Halifax, Nova Scotia, Canada Biomedical Translational Imaging Centre, IWK Health Sciences Centre, Halifax, Nova Scotia, Canada.
Journal of Neurology, Neurosurgery, and Psychiatry
|March 28, 2015
Summary
Receiver operating characteristic-reliability (ROC-r) analysis improves presurgical functional MRI (fMRI) by identifying reliable data and optimizing processing for better surgical mapping. This ensures higher quality fMRI scans for neurosurgical planning.
Area of Science:
- Neuroimaging
- Clinical Neuroscience
- Medical Physics
Background:
- Functional MRI (fMRI) is crucial for presurgical mapping in neurosurgery.
- Ensuring the reliability and quality of fMRI data is essential for accurate presurgical planning.
- Current methods for assessing fMRI data quality and optimizing processing are limited.
Purpose of the Study:
- To demonstrate the utility of receiver operating characteristic-reliability (ROC-r) analysis for presurgical fMRI.
- To identify reliable versus unreliable fMRI datasets.
- To automatically select processing options and individual thresholds for activation maps.
Main Methods:
- Presurgical fMRI was performed on 16 brain tumor patients.
- Within-session test-retest fMRI was used to assess ROC-reliability.
- Optimized preprocessing pipelines were applied to improve reliability.
- fMRI results were spatially compared to intraoperative cortical stimulation mapping.
Main Results:
- Average ROC-r reliability in patients (0.58±0.03) was lower than in healthy controls (0.72±0.02).
- Optimized preprocessing pipelines significantly increased patient group reliability to 0.65±0.02.
- Higher fMRI data reliability correlated with better spatial correspondence to cortical stimulation (8.3±0.9 mm vs 29±3 mm).
Conclusions:
- ROC-r analysis effectively identifies reliable fMRI datasets and optimizes processing pipelines for presurgical mapping.
- Higher fMRI reliability leads to improved spatial accuracy compared to direct cortical stimulation.
- ROC-r enables automated fMRI processing, improving presurgical mapping and allowing for repeat scans of poor-quality data.
