Abstract
BackgroundIt is reported that radiomic features extracted from quantitative susceptibility mapping (QSM) had promising clinical value for the diagnosis of Parkinson’s disease (PD). We aimed to explore the usefulness of radiomics features based on magnitude images to distinguish PD from non-PD controls.MethodsWe retrospectively recruited PD patients and controls who underwent brain 3.0T MR including susceptibility-weighted imaging (SWI). A total of 396 radiomics features were extracted from the SN of 95 PD patients and 95 non-PD controls based on SWI. Intra-/inter-observer correlation coefficients (ICCs) were applied to measure the observer agreement for the radiomic feature extraction. Then the patients were randomly grouped into training and validation sets in a ratio of 7:3. In the training set, the maximum correlation minimum redundancy algorithm (mRMR) and the least absolute shrinkage and selection operator (LASSO) were conducted to filter and choose the optimized subset of features, and a radiomics signature was constructed. Moreover, radiomics signatures were constructed by different machine learning models. Area under the ROC curves (AUCs) were applied to evaluate the predictive performance of the models. Then correlation analysis was performed to evaluate the correlation between the optimized features and clinical factors.ResultsThe intro-observer CC ranged from 0.82 to 1.0, and the inter-observer CC ranged from 0.77 to 0.99. The LASSO logistic regression model showed good prediction efficacy in the training set [AUC = 0.82, 95% confidence interval (CI, 0.74–0.88)] and the validation set [AUC = 0.81, 95% CI (0.68–0.91)]. One radiomic feature showed a moderate negative correlation with Hoehn-Yahr stage (r = −0.49, P = 0.012).ConclusionRadiomic predictive features based on SWI magnitude images could reflect the Hoehn-Yahr stage of PD to some extent.
Highlights
MATERIALS AND METHODSParkinson’s disease (PD) is the second most common neurodegenerative disease and affects 1% of the population above 60 years (Tysnes and Storstein, 2017)
Radiomic predictive features based on susceptibility-weighted imaging (SWI) magnitude images could reflect the Hoehn-Yahr stage of PD to some extent
The 16 features with non-zero coefficients are shown in Figure 2C and the weight of each feature that contributed to the established signature is displayed
Summary
MATERIALS AND METHODSParkinson’s disease (PD) is the second most common neurodegenerative disease and affects 1% of the population above 60 years (Tysnes and Storstein, 2017). Postmortem study has shown that loss of dopaminergic neurons occurs most severely in the lateral ventral tier, followed by the medial ventral tier of the SN pars compacta (Fearnley et al, 1991). In an in vivo study, susceptibility-weighted imaging (SWI) was found to be more sensitive for detection brain mineralization than conventional MRI sequences. The inconsistency of the swallow tail sign occurrence in healthy subjects (Schmidt et al, 2017), and the disappearance of the swallow tail sign can be found in some cognitive disorders (Rizzo et al, 2019), these findings make it more difficult to diagnose Parkinson’s disease using only MRI signal changes. It is reported that radiomic features extracted from quantitative susceptibility mapping (QSM) had promising clinical value for the diagnosis of Parkinson’s disease (PD). We aimed to explore the usefulness of radiomics features based on magnitude images to distinguish PD from non-PD controls
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have