This study, conducted by Roberts et al., investigates how quantitative susceptibility mapping (QSM) MRI data could be used to predict deep brain stimulation (DBS) outcomes in patients who have Parkinson’s disease. The researchers created a regression model using QSM features from the subthalamic nucleus and the substantia nigra to predict postoperative motor improvement scores (UPDRS-III) in Parkinson’s patients who are going through DBS. The regression model used data augmentation techniques on sixty seven patients from two different medical centers to assess the accuracy compared to a more traditional type of test, known as the levodopa challenge test (LCT). The QSM model predicted the DBS outcomes at both centers, meaning that there was a strong correlation between predicted motor improvements and observed motor improvements. On the other hand, the traditional LCT was not able to predict DBS success. The iron distribution homogeneity yielded the most accurate prediction features in the substantia nigra, while the iron distribution heterogeneity yielded the most accurate prediction features in the subthalamic nucleus, showing that QSM features were a strong predictor of DBS responsiveness. The iron distribution in the deep brain nuclei could reflect the structural changes that are associated with motor symptom improvement, which highlights the usefulness and potential of QSM.