Table 2 Cross-validated predictive performance for MDS-UPDRS Part III classification

From: Integrating digital gait data with metabolomics and clinical data to predict outcomes in Parkinson’s disease

Models

Linear SVM

RBF SVM

RF

Stochastic GBM

DEEP

XGB

Clinical confounders (median)

10-fold CV

AUC

median

(mad)

Extracted gait parameters

0.610

(0.21)

0.654

(0.09)

0.705

(0.12)

0.615

(0.16)

0.667

(0.12)

a0.76

(0.12)

0.554

(0.12)

Raw signal time series features

0.664

(0.09)

0.666

(0.08)

a0.754

(0.22)

0.655

(0.13)

0.740

(0.10)

0.745

(0.06)

  1. Cross-validated predictive performance for MDS-UPDRS Part III classification of whether MDS-UPDRS Part III sum scores are above or below the media score using different machine learning methods and time series features computed from either the extracted gait parameters or the raw gait signal data as input. “Clinical confounders” refers to a model that was solely trained with age and sex as predictors and serves as a comparator.
  2. AUC area under the Receiver Operating Characteristic Curve, median and median absolute deviation across 10 cross-validation (CV) cycles, SVM support vector machine, RBF radial basis function, RF random forest, DEEP deep boosting, XGB extreme gradient boosting, GBM gradient boosting machines.
  3. aThe highest median AUC for each row.
Back to article page