Diagnostic Performance of Deep Learning for Predicting Gliomas' IDH and 1p/19q Status in MRI: A Systematic Review and Meta-Analysis

2024-10-28

arXiv.org

AI Summary

Non-invasive prediction of isocitrate dehydrogenase (IDH) mutation and 1p/19q codeletion status in gliomas using deep learning (DL)-based radiomics from routine MRI remains clinically challenging. Method: We conducted a systematic review and meta-analysis of 52 studies, applying PRISMA guidelines alongside dual quality assessment tools—QUADAS-2 for risk of bias and RQS for radiomics methodology—and employed bivariate random-effects models and meta-regression to quantify heterogeneity attributable to MRI sequences, tumor segmentation approaches, and validation strategies. Contribution/Results: Pooled diagnostic performance showed high accuracy for IDH prediction (sensitivity = 0.84, specificity = 0.87, AUC = 0.89) and robust discrimination for 1p/19q codeletion (sensitivity = 0.76, specificity = 0.85, AUC = 0.90). The analysis confirms the clinical translatability of DL radiomics models while identifying insufficient external validation as the primary barrier to implementation. This work establishes methodological benchmarks and actionable pathways for model optimization and clinical integration.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Evaluation and Analysis
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Gliomas, the most common primary brain tumors, show high heterogeneity in histological and molecular characteristics. Accurate molecular profiling, like isocitrate dehydrogenase (IDH) mutation and 1p/19q codeletion, is critical for diagnosis, treatment, and prognosis. This review evaluates MRI-based deep learning (DL) models' efficacy in predicting these biomarkers. Following PRISMA guidelines, we systematically searched major databases (PubMed, Scopus, Ovid, and Web of Science) up to February 2024, screening studies that utilized DL to predict IDH and 1p/19q codeletion status from MRI data of glioma patients. We assessed the quality and risk of bias using the radiomics quality score and QUADAS-2 tool. Our meta-analysis used a bivariate model to compute pooled sensitivity, specificity, and meta-regression to assess inter-study heterogeneity. Of the 565 articles, 57 were selected for qualitative synthesis, and 52 underwent meta-analysis. The pooled estimates showed high diagnostic performance, with validation sensitivity, specificity, and area under the curve (AUC) of 0.84 [prediction interval (PI): 0.67-0.93, I2=51.10%, p<0.05], 0.87 [PI: 0.49-0.98, I2=82.30%, p<0.05], and 0.89 for IDH prediction, and 0.76 [PI: 0.28-0.96, I2=77.60%, p<0.05], 0.85 [PI: 0.49-0.97, I2=80.30%, p<0.05], and 0.90 for 1p/19q prediction, respectively. Meta-regression analyses revealed significant heterogeneity influenced by glioma grade, data source, inclusion of non-radiomics data, MRI sequences, segmentation and feature extraction methods, and validation techniques. DL models demonstrate strong potential in predicting molecular biomarkers from MRI scans, with significant variability influenced by technical and clinical factors. Thorough external validation is necessary to increase clinical utility.

Problem

• Evaluating deep learning performance for glioma molecular classification
• Assessing noninvasive prediction of IDH and 1p/19q status
• Identifying methodological factors affecting diagnostic accuracy generalizability

Innovation

• Deep learning for noninvasive glioma molecular classification
• MRI-based radiomics with automated tumor segmentation
• Multi-center data harmonization using style transfer