Related Experiment Video
Updated: Jun 17, 2026
11:09
Deep Brain Stimulation with Simultaneous fMRI in Rodents
Published on: February 15, 2014
14.6K
Technical Feasibility of Quantitative Susceptibility Mapping Radiomics for Predicting Deep Brain Stimulation Outcomes
Alexandra G Roberts1,2, Jinwei Zhang3, Ceren Tozlu2
1Electrical and Computer Engineering, Cornell University, Ithaca , New York , USA.
Neurosurgery
|September 18, 2025
Summary
Quantitative susceptibility mapping (QSM) radiomics accurately predicts deep brain stimulation (DBS) outcomes in Parkinson disease (PD) patients. This novel MRI approach offers a more reliable alternative to the levodopa challenge test (LCT) for predicting surgical success.
Area of Science:
- Neuroimaging and Computational Analysis
- Neurological Surgery and Movement Disorders
Background:
- Deep brain stimulation (DBS) is a surgical option for Parkinson disease (PD) patients with motor complications, but predicting surgical success remains challenging.
- Current DBS candidacy assessment relies on the levodopa challenge test (LCT), which has limitations in accurately predicting post-surgical outcomes.
- Quantitative susceptibility mapping (QSM), an MRI technique visualizing brain iron, shows promise for presurgical planning due to its depiction of the substantia nigra and subthalamic nuclei.
Purpose of the Study:
- To demonstrate the technical feasibility of using presurgical QSM radiomics to predict DBS outcomes in Parkinson disease patients.
- To evaluate a novel QSM radiomics regression model, enhanced with data augmentation, for predicting motor symptom improvement after DBS surgery.
Main Methods:
- A novel presurgical QSM radiomics approach utilizing a regression model was developed to predict DBS outcomes based on spatial features in deep gray nuclei.
- Data augmentation techniques, including label noise injection, were employed to improve the regression model's predictive accuracy with limited training data.
- The QSM radiomics model was validated on a cohort of 67 PD patients who underwent DBS surgery across two medical centers.
Main Results:
- The QSM radiomics model demonstrated significant predictive accuracy for DBS-induced improvement in the Unified Parkinson Disease Rating Scale at both study centers (Center 1: r = [value], Center 2: r = [value]).
- In contrast, the levodopa challenge test (LCT) showed poor predictive performance for DBS improvement at both centers (Center 1: r = [value], Center 2: r = [value]).
Conclusions:
- QSM radiomics presents a promising tool for accurately predicting DBS outcomes in Parkinson disease patients.
- This advanced MRI-based method offers a potentially more accurate and reliable alternative to the conventional levodopa challenge test for assessing surgical candidacy and predicting response.
