Cargando…

Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment

Background: Current research on the prediction of movement complications associated with levodopa therapy in Parkinson’s disease (PD) is limited. levodopa-induced dyskinesia (LID) is a movement complication that seriously affects the life quality of PD patients. One-third of PD patients develop LID...

Descripción completa

Detalles Bibliográficos
Autores principales: Luo, Yang, Chen, Huiqin, Gui, Mingzhen
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10417024/
https://www.ncbi.nlm.nih.gov/pubmed/37568874
http://dx.doi.org/10.3390/diagnostics13152511
_version_ 1785087921343168512
author Luo, Yang
Chen, Huiqin
Gui, Mingzhen
author_facet Luo, Yang
Chen, Huiqin
Gui, Mingzhen
author_sort Luo, Yang
collection PubMed
description Background: Current research on the prediction of movement complications associated with levodopa therapy in Parkinson’s disease (PD) is limited. levodopa-induced dyskinesia (LID) is a movement complication that seriously affects the life quality of PD patients. One-third of PD patients develop LID within 1 to 6 years of levodopa treatment. This study aimed to construct models based on radiomics and machine learning to predict early LID in PD. Methods: We extracted radiomics features from the T1-weighted MRI obtained in the baseline of 49 PD control and 54 PD with LID in the first 6 years of levodopa therapy. Six brain regions related to the onset of PD were segmented as regions of interest (ROIs). The least absolute shrinkage and selection operator (LASSO) was used for feature selection. Using the machine learning methods of support vector machine (SVM), random forest (RF), and AdaBoost, we constructed radiomics models and hybrid models. The hybrid models combined the radiomics features and the Unified Parkinson’s Disease Rating Scale part III (UPDRS III) total score. The five-fold cross-validation was performed and repeated 20 times to validate the stability of the classifiers. We used sensitivity, specificity, accuracy, receiver operating characteristic (ROC) curves, and area under the ROC curve (AUC) for model validation. Results: We selected 33 out of 6138 radiomics features. In the testing set of the radiomics model, the AUC values of the SVM, RF, and AdaBoost classifiers were 0.905, 0.808, and 0.778, respectively, and the accuracies were 0.839, 0.742, and 0.710. The hybrid models had better prediction performance. In the testing set, the AUC values of SVM, RF, and AdaBoost classifiers were 0.958, 0.861, and 0.832, respectively, and the accuracies were 0.903, 0.806, and 0.774. Conclusions: Our results indicate that T1-weighted MRI is valuable in predicting early LID in PD. This work demonstrates that the combination of radiomics features and clinical features has good potential and value for identifying early LID in PD.
format Online
Article
Text
id pubmed-10417024
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-104170242023-08-12 Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment Luo, Yang Chen, Huiqin Gui, Mingzhen Diagnostics (Basel) Article Background: Current research on the prediction of movement complications associated with levodopa therapy in Parkinson’s disease (PD) is limited. levodopa-induced dyskinesia (LID) is a movement complication that seriously affects the life quality of PD patients. One-third of PD patients develop LID within 1 to 6 years of levodopa treatment. This study aimed to construct models based on radiomics and machine learning to predict early LID in PD. Methods: We extracted radiomics features from the T1-weighted MRI obtained in the baseline of 49 PD control and 54 PD with LID in the first 6 years of levodopa therapy. Six brain regions related to the onset of PD were segmented as regions of interest (ROIs). The least absolute shrinkage and selection operator (LASSO) was used for feature selection. Using the machine learning methods of support vector machine (SVM), random forest (RF), and AdaBoost, we constructed radiomics models and hybrid models. The hybrid models combined the radiomics features and the Unified Parkinson’s Disease Rating Scale part III (UPDRS III) total score. The five-fold cross-validation was performed and repeated 20 times to validate the stability of the classifiers. We used sensitivity, specificity, accuracy, receiver operating characteristic (ROC) curves, and area under the ROC curve (AUC) for model validation. Results: We selected 33 out of 6138 radiomics features. In the testing set of the radiomics model, the AUC values of the SVM, RF, and AdaBoost classifiers were 0.905, 0.808, and 0.778, respectively, and the accuracies were 0.839, 0.742, and 0.710. The hybrid models had better prediction performance. In the testing set, the AUC values of SVM, RF, and AdaBoost classifiers were 0.958, 0.861, and 0.832, respectively, and the accuracies were 0.903, 0.806, and 0.774. Conclusions: Our results indicate that T1-weighted MRI is valuable in predicting early LID in PD. This work demonstrates that the combination of radiomics features and clinical features has good potential and value for identifying early LID in PD. MDPI 2023-07-27 /pmc/articles/PMC10417024/ /pubmed/37568874 http://dx.doi.org/10.3390/diagnostics13152511 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Luo, Yang
Chen, Huiqin
Gui, Mingzhen
Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title_full Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title_fullStr Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title_full_unstemmed Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title_short Radiomics and Hybrid Models Based on Machine Learning to Predict Levodopa-Induced Dyskinesia of Parkinson’s Disease in the First 6 Years of Levodopa Treatment
title_sort radiomics and hybrid models based on machine learning to predict levodopa-induced dyskinesia of parkinson’s disease in the first 6 years of levodopa treatment
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10417024/
https://www.ncbi.nlm.nih.gov/pubmed/37568874
http://dx.doi.org/10.3390/diagnostics13152511
work_keys_str_mv AT luoyang radiomicsandhybridmodelsbasedonmachinelearningtopredictlevodopainduceddyskinesiaofparkinsonsdiseaseinthefirst6yearsoflevodopatreatment
AT chenhuiqin radiomicsandhybridmodelsbasedonmachinelearningtopredictlevodopainduceddyskinesiaofparkinsonsdiseaseinthefirst6yearsoflevodopatreatment
AT guimingzhen radiomicsandhybridmodelsbasedonmachinelearningtopredictlevodopainduceddyskinesiaofparkinsonsdiseaseinthefirst6yearsoflevodopatreatment