Diagnostic Accuracy of Machine Learning-Based Radiomics in Grading Gliomas: Systematic Review and Meta-Analysis

Abstract

Purpose. This study aimed to estimate the diagnostic accuracy of machine learning- (ML-) based radiomics in differentiating high-grade gliomas (HGG) from low-grade gliomas (LGG) and to identify potential covariates that could affect the diagnostic accuracy of ML-based radiomic analysis in classifying gliomas. Method. A primary literature search of the PubMed database was conducted to find all related literatures in English between January 1, 2009, and May 1, 2020, with combining synonyms for “machine learning,” “glioma,” and “radiomics.” Five retrospective designed original articles including LGG and HGG subjects were chosen. Pooled sensitivity, specificity, their 95% confidence interval, area under curve (AUC), and hierarchical summary receiver-operating characteristic (HSROC) models were obtained. Result. The pooled sensitivity when diagnosing HGG was higher (96% (95% CI: 0.93, 0.98)) than the specificity when diagnosing LGG (90% (95% CI 0.85, 0.93)). Heterogeneity was observed in both sensitivity and specificity. Metaregression confirmed the heterogeneity in sample sizes ( p = 0.05 ), imaging sequence types ( p = 0.02 ), and data sources ( p = 0.01 ), but not for the inclusion of the testing set ( p = 0.19 ), feature extraction number ( p = 0.36 ), and selection of feature number ( p = 0.18 ). The results of subgroup analysis indicate that sample sizes of more than 100 and feature selection numbers less than the total sample size positively affected the diagnostic performance in differentiating HGG from LGG. Conclusion. This study demonstrates the excellent diagnostic performance of ML-based radiomics in differentiating HGG from LGG.

The Diagnostic Performance of Machine Learning-Based Radiomics of DCE-MRI in Predicting Axillary Lymph Node Metastasis in Breast Cancer: A Meta-Analysis

“…In this setting, datasets usually consist of hundreds of patients at most, which is better than with deep learning in this case. Similar findings have been previously reported for ML in other applications (9,10,30). However, deep learning only included two studies.…”

“…First, a relatively small number of studies met the selection criteria. The second limitation was the significant heterogeneity, which is an issue similar to that in other metaanalyses of diagnostic accuracy using ML based on radiomics (9,10,30).…”

“…In recent years, radiomics and machine learning (ML) models have become increasingly popular for analyzing diagnostic images (9,10). The ability of radiomics analysis to maximize the number of features in quantitative images has excellent potential for evaluating ALNM in breast cancer patients (11)(12)(13)(14)(15).…”

Machine Learning Tools for Image-Based Glioma Grading and the Quality of Their Reporting: Challenges and Opportunities

“…This requirement may be advantageous for veteran diagnostic imagers, whose knowledge of brain tumor appearance may enhance feature design and selection. Hand-engineered features also can undergo feature reduction to mitigate the risks of overfitting, and prior works demonstrate better performance for glioma grading models using a smaller number of quantitative features [ 41 ]. However, hand-engineered features are limited since they cannot be adjusted during model training, and it is uncertain if they are optimal features for classification.…”

“…Areas for improvement included reporting of titles and abstracts, justification of sample size, full model specification and performance, and participant demographics, and missing data. Sohn et al’s meta-analysis of radiomics studies differentiating high- and low-grade gliomas estimated a high risk of bias according to QUADAS-2, attributing this to the fact that all their analyzed studies were retrospective (and have the potential for bias because patient outcomes are already known), the lack of control over acquisition factors in the studies using public imaging data, and unclear study flow and timing due to poor reporting [ 41 ]. Readers should refer directly to Navarro et al, Bahar et al and Sohn et al for more detailed discussion of shortcomings in study reporting and risk of bias.…”