Optimized machine learning methods for prediction of cognitive outcome in Parkinson's disease

Given the increasing recognition of the significance of non-motor symptoms in Parkinson's disease, we investigate the optimal use of machine learning methods for the prediction of the Montreal Cognitive Assessment (MoCA) score at year 4 from longitudinal data obtained at years 0 and 1. We selec...

Full description

Saved in:
Bibliographic Details
Published inComputers in biology and medicine Vol. 111; p. 103347
Main Authors Salmanpour, Mohammad R., Shamsaei, Mojtaba, Saberi, Abdollah, Setayeshi, Saeed, Klyuzhin, Ivan S., Sossi, Vesna, Rahmim, Arman
Format Journal Article
LanguageEnglish
Published United States Elsevier Ltd 01.08.2019
Elsevier Limited
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Given the increasing recognition of the significance of non-motor symptoms in Parkinson's disease, we investigate the optimal use of machine learning methods for the prediction of the Montreal Cognitive Assessment (MoCA) score at year 4 from longitudinal data obtained at years 0 and 1. We selected n = 184 PD subjects from the Parkinson's Progressive Marker Initiative (PPMI) database (93 features). A range of robust predictor algorithms (accompanied with automated machine learning hyperparameter tuning) and feature subset selector algorithms (FSSAs) were selected. We utilized 65%, 5% and 30% of patients in each arrangement for training, training validation and final testing respectively (10 randomized arrangements). For further testing, we enrolled 308 additional patients. First, we employed 10 predictor algorithms, provided with all 93 features; an error of 1.83 ± 0.13 was obtained by LASSOLAR (Least Absolute Shrinkage and Selection Operator - Least Angle Regression). Subsequently, we used feature subset selection followed by predictor algorithms. GA (Genetic Algorithm) selected 18 features; subsequently LOLIMOT (Local Linear Model Trees) reached an error of 1.70 ± 0.10. DE (Differential evolution) also selected 18 features and coupled with Thiel-Sen regression arrived at a similar performance. NSGAII (Non-dominated sorting genetic algorithm) yielded the best performance: it selected six vital features, which combined with LOLIMOT reached an error of 1.68 ± 0.12. Finally, using this last approach on independent test data, we reached an error of 1.65. By employing appropriate optimization tools (including automated hyperparameter tuning), it is possible to improve prediction of cognitive outcome. Overall, we conclude that optimal utilization of FSSAs and predictor algorithms can produce very good prediction of cognitive outcome in PD patients. •We explored a range of predictor machines and feature subset selector machines.•LASSOLAR was able to work with a significant number of inputs.•NSGAII selected 6 vital features for prediction of cognitive outcome (MoCA).•Using 6 vital features, LOLIMOT reached an error of 1.68 ± 0.12•Using independent testing, we reached an error of 1.65, further confirming our findings.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0010-4825
1879-0534
1879-0534
DOI:10.1016/j.compbiomed.2019.103347