Molecular subtype classification of low-grade gliomas using magnetic resonance imaging-based radiomics and machine learning.
NMR Biomed
; 35(11): e4792, 2022 11.
Article
de En
| MEDLINE
| ID: mdl-35767281
ABSTRACT
In 2016, the World Health Organization (WHO) updated the glioma classification by incorporating molecular biology parameters, including low-grade glioma (LGG). In the new scheme, LGGs have three molecular subtypes isocitrate dehydrogenase (IDH)-mutated 1p/19q-codeleted, IDH-mutated 1p/19q-noncodeleted, and IDH-wild type 1p/19q-noncodeleted entities. This work proposes a model prediction of LGG molecular subtypes using magnetic resonance imaging (MRI). MR images were segmented and converted into radiomics features, thereby providing predictive information about the brain tumor classification. With 726 raw features obtained from the feature extraction procedure, we developed a hybrid machine learning-based radiomics by incorporating a genetic algorithm and eXtreme Gradient Boosting (XGBoost) classifier, to ascertain 12 optimal features for tumor classification. To resolve imbalanced data, the synthetic minority oversampling technique (SMOTE) was applied in our study. The XGBoost algorithm outperformed the other algorithms on the training dataset by an accuracy value of 0.885. We continued evaluating the XGBoost model, then achieved an overall accuracy of 0.6905 for the three-subtype classification of LGGs on an external validation dataset. Our model is among just a few to have resolved the three-subtype LGG classification challenge with high accuracy compared with previous studies performing similar work.
Mots clés
Texte intégral:
1
Collection:
01-internacional
Base de données:
MEDLINE
Sujet principal:
Tumeurs du cerveau
/
Gliome
Type d'étude:
Observational_studies
/
Prognostic_studies
/
Risk_factors_studies
Limites:
Humans
Langue:
En
Journal:
NMR Biomed
Sujet du journal:
DIAGNOSTICO POR IMAGEM
/
MEDICINA NUCLEAR
Année:
2022
Type de document:
Article
Pays d'affiliation:
Taïwan