Prediction of MGMT Methylation Status of Glioblastoma using Radiomics and Latent Space Shape Features

TL;DR

This study develops a machine learning approach combining radiomics and shape features from MR images to predict MGMT methylation status in glioblastoma, demonstrating effectiveness on a public dataset.

Contribution

It introduces a novel combination of radiomic and autoencoder-derived shape features for MGMT methylation prediction in gliomas.

Findings

Achieved accurate MGMT methylation status prediction

Utilized deep learning for tumor segmentation

Validated on BraTS 2021 dataset

Abstract

In this paper we propose a method for predicting the status of MGMT promoter methylation in high-grade gliomas. From the available MR images, we segment the tumor using deep convolutional neural networks and extract both radiomic features and shape features learned by a variational autoencoder. We implemented a standard machine learning workflow to obtain predictions, consisting of feature selection followed by training of a random forest classification model. We trained and evaluated our method on the RSNA-ASNR-MICCAI BraTS 2021 challenge dataset and submitted our predictions to the challenge.