Baseline cerebral structural morphology predict freezing of gait in early drug-naïve Parkinson’s disease

• DOI: 10.1038/s41531-022-00442-4
• Publication Date: 2022-12-29T08:03:30Z
• Authors: Yuting Li; Xiaofei Huang; Xiuhang Ruan; Dingna Duan; Yihe Zhang; Shaode Yu; Amei Chen; Zhaoxiu Wang; Yujian Zou; Mingrui Xia; Xinhua Wei
• Keywords: Brain morphometry

Freezing of gait (FOG) greatly impacts the daily life patients with Parkinson's disease (PD). However, predictors FOG in early PD are limited. Moreover, recent neuroimaging evidence cerebral morphological alterations is heterogeneous. We aimed to develop a model that could predict occurrence using machine learning, collaborating clinical, laboratory, and structural imaging information drug-naïve investigate morphology PD. Data from 73 healthy controls (HCs) 158 at baseline were obtained Progression Markers Initiative cohort. The CIVET pipeline was used generate features T1-weighted (T1WI). Five learning algorithms calculated assess predictive performance future during 5-year follow-up period. found models trained showed fair good (accuracy range, 0.67-0.73). Performance improved when clinical laboratory data added 0.71-0.78). For algorithms, elastic net-support vector 0.69-0.78) performed best. main based on mainly distributed left cerebrum. bilateral olfactory cortex (OLF) significantly higher surface area than HCs. Overall, we T1WI morphometric markers helped individual level. OLF exhibits predominantly cortical expansion