Fig. 2: Machine learning models. | npj Parkinson's Disease

Fig. 2: Machine learning models.

From: Prediction of motor and non-motor Parkinson’s disease symptoms using serum lipidomics and machine learning: a 2-year study

Fig. 2

Comparison of random forest (blue) and elastic net linear regression (red) models to predict the (a) Geriatric Depression, (b) Hoehn and Yahr, (c) The Schwab and England Activities of Daily Living scale, (d) Unified Parkinson’s Disease Rating Scale (UPDRS III) and (e) University of Pennsylvania Smell Identification Test (UPSIT) scale scores after 2 years in Parkinson’s disease participants. The predictors used are combinations of baseline lipids and cytokines with and without baseline clinical data. ‘All’ lipids is the entire lipid dataset whereas ‘lipids’ is the subset of lipids that separate outcomes from the OPLS. The analysis was performed on the training set (n = 54). RMSE root mean square error.

Back to article page