Optimizing Prediction of MGMT Promoter Methylation from MRI Scans using Adversarial Learning

TL;DR

This paper presents a novel adversarial learning approach to predict MGMT promoter methylation status from MRI scans, achieving significant accuracy improvements over previous models, which could enhance personalized treatment for glioblastoma patients.

Contribution

The study introduces a new ML model called the Intermediate State Generator for MRI normalization and demonstrates its effectiveness in improving MGMT methylation prediction accuracy.

Findings

Achieved over 6% higher accuracy than Kaggle's best model.

Developed four models including a novel MRI normalization technique.

Significant statistical improvement in prediction performance.

Abstract

Glioblastoma Multiforme (GBM) is a malignant brain cancer forming around 48% of al brain and Central Nervous System (CNS) cancers. It is estimated that annually over 13,000 deaths occur in the US due to GBM, making it crucial to have early diagnosis systems that can lead to predictable and effective treatment. The most common treatment after GBM diagnosis is chemotherapy, which works by sending rapidly dividing cells to apoptosis. However, this form of treatment is not effective when the MGMT promoter sequence is methylated, and instead leads to severe side effects decreasing patient survivability. Therefore, it is important to be able to identify the MGMT promoter methylation status through non-invasive magnetic resonance imaging (MRI) based machine learning (ML) models. This is accomplished using the Brain Tumor Segmentation (BraTS) 2021 dataset, which was recently used for an international Kaggle competition. We developed four primary models - two radiomic models and two CNN models - each solving the binary classification task with progressive improvements. We built a novel ML model termed as the Intermediate State Generator which was used to normalize the slice thicknesses of all MRI scans. With further improvements, our best model was able to achieve performance significantly ($p < 0.05$) better than the best performing Kaggle model with a 6% increase in average cross-validation accuracy. This improvement could potentially lead to a more informed choice of chemotherapy as a treatment option, prolonging lives of thousands of patients with GBM each year.