Abstract #3018

Substantia Nigra Susceptibility Features Derived by Radiomics Predict Motor Outcome for STN-DBS in Parkinson’s Disease

Naying He¹, Yu Liu¹, Bin Xiao², Junchen Li³, Chencheng Zhang⁴, Yijie Lai⁴, Feng Shi⁵, Dinggang Shen⁵, Yan Li¹, Hongjiang Wei⁶, Ewart Mark Haacke^1,7, Weibo Chen⁸, Qian Wang², Dianyou Li⁴, and Fuhua Yan¹

¹Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, ²Institute for Medical Imaging Technology, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, Shanghai, China, ³Changshu Hospital Affiliated to Nanjing University of Chinese Medicine, No. 6 Huanghe Road, Changshu, China, Changshu, China, ⁴Department of Neurosurgery, Center for Functional Neurosurgery, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, Shanghai, China, ⁵Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China, Shanghai, China, ⁶Institute for Medical Imaging Technology, Department of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, Shanghai, China, ⁷Department of Radiology, Wayne State University, Detroit, Michigan, USA, Detroit, MI, United States, ⁸Philips Healthcare，Shanghai，China, Shanghai, China

Currently, there are neither individual objective nor quantitative indicators for predicting DBS motor outcome. We hypothesized that the distribution of SN iron changes in PD patients may reflect a specific disease trait and could potentially account for some variability in the motor outcomes after sub-thalamic nucleus (STN) deep brain stimulation (DBS). We developed a radiomics model with machine learning (RA-ML) based on preoperative individual QSM of the SN to predict motor outcome for STN-DBS in PD and it performed best with an AUC of 0.897. In addition, the threshold probability of the RA-ML model can differentiate surgical responders and non-responders.

This abstract and the presentation materials are available to members only; a login is required.

Join Here