Subtyping and grading of lower-grade gliomas using integrated feature selection and support vector machine

Abstract

Classifying lower-grade gliomas (LGGs) is a crucial step for accurate therapeutic intervention. The histopathological classification of various subtypes of LGG, including astrocytoma, oligodendroglioma and oligoastrocytoma, suffers from intraobserver and interobserver variability leading to inaccurate classification and greater risk to patient health. We designed an efficient machine learning-based classification framework to diagnose LGG subtypes and grades using transcriptome data. First, we developed an integrated feature selection method based on correlation and support vector machine (SVM) recursive feature elimination. Then, implementation of the SVM classifier achieved superior accuracy compared with other machine learning frameworks. Most importantly, we found that the accuracy of subtype classification is always high (>90%) in a specific grade rather than in mixed grade (~80%) cancer. Differential co-expression analysis revealed higher heterogeneity in mixed grade cancer, resulting in reduced prediction accuracy. Our findings suggest that it is necessary to identify cancer grades and subtypes to attain a higher classification accuracy. Our six-class classification model efficiently predicts the grades and subtypes with an average accuracy of 91% (±0.02). Furthermore, we identify several predictive biomarkers using co-expression, gene set enrichment and survival analysis, indicating our framework is biologically interpretable and can potentially support the clinician.

A data-centric machine learning approach to improve prediction of glioma grades using low-imbalance TCGA data

“…Joshi et al 22 proposed a two-stage ensemble for glioma detection and grading based on clinical and histological data. Munquad et al 23 employed a correlation-based feature selection scheme and an SVM to predict LGG and subtypes, achieving an average accuracy of 91%. Ren et al 24 predicted IDH1 (isocitrate dehydrogenase 1)…”

Uncovering the subtype-specific disease module and the development of drug response prediction models for glioma

“…Each subtype has distinct molecular features, and they can be classified using genomics and epigenomics profiles. Despite the presence of intratumor molecular heterogeneity, recent research has shown that deep learning (DL) and machine learning (ML)-based methods may accurately identify glioma subtypes [ 6 , 7 ]. Due to distinct molecular characteristics, the subtypes of glioma have different clinical outcomes and responses to treatment, highlighting the importance of personalized medicine for brain cancer treatment [ 8 ].…”

Immune landscape-based machine-learning–assisted subclassification, prognosis, and immunotherapy prediction for glioblastoma

“…Meanwhile, machine learning (ML)-based methods could detect key features from complex datasets and have been popular applications in clinical cancer research in recent years, such as early diagnosis, subtype identification, prognosis prediction, and so on ( 20 ). It has been used to classify various cancer subtypes, for example, breast cancer ( 21 ), adult T-cell leukemia/lymphoma ( 22 ), kidney cancer ( 23 ), and glioma ( 24 ). The application of ML for cancer subtype identification will enable accurate diagnosis and regard to the clinical management of patients.…”