# Radiomics-Driven Hybrid Deep Learning for MRI-Based Prediction of Glioma Grade and 1p/19q Codeletion

## TL;DR

This study introduces a non-invasive AI method using MRI scans to predict glioma tumor grade and a key genetic marker, 1p/19q codeletion, with high accuracy.

## Contribution

A novel hybrid deep learning model combining CNN and LSTM for MRI-based glioma grading and 1p/19q codeletion prediction is proposed.

## Key findings

- The hybrid CNN-LSTM model achieved 88.1% accuracy and 0.93 AUC in predicting glioma grade and 1p/19q codeletion.
- Tumor heterogeneity and morphology features were most influential in model predictions.
- The model outperformed traditional ML and single DL models in accuracy and reliability.

## Abstract

This study provides a non-invasive method for predicting the grade of glioma tumors and an important genetic marker called Chromosome 1p and 19q codeletion (1p/19q) codeletion through common Magnetic Resonance Imaging (MRI) scans. Unlike the current method of using surgical biopsies and genetic analysis in a laboratory setting, the new method utilizes quantitative features (radiomics) from MRI scans and sophisticated artificial intelligence (AI) models. This study shows that a hybrid deep learning (DL) model, which combines convolutional neural networks and recurrent neural networks, has the ability to accurately detect complex patterns of glioma tumors in terms of shape, texture, and internal heterogeneity. The hybrid model outperformed machine learning (ML) models and single DL models, achieving high accuracy and high reliability in tumor subtype classification. Most importantly, the study also explains the most important imaging features that are used in the predictions, enabling clinicians to understand and trust the predictions. In summary, the findings of this research study show that MRI-based radiomics and hybrid DL models can be a reliable, interpretable, and non-invasive tool for personalized diagnosis and treatment of patients with low-grade glioma.

Background: Correct preoperative evaluation of glioma grade and molecular profile is a prerequisite for tailored treatment strategies. Specifically, the 1p/19q codeletion status represents a major prognostic and therapeutic marker in low-grade gliomas (LGGs). Nevertheless, its assessment is presently performed through invasive histopathological and genetic studies, thus underlining the need for non-invasive alternative approaches. Methods: We introduce a non-invasive radiomics framework that combines quantitative MRI features with sophisticated ML and DL approaches for glioma grading and 1p/19q codeletion status prediction. High-dimensional radiomic features characterizing tumor geometry, intensity, and texture were derived from preoperative MRI-based tumor delineations. Features were normalized and optimized using correlation-based feature selection. Several traditional ML classifiers were compared and contrasted with DL models, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and a CNN-Long Short-Term Memory (LSTM) hybrid model tailored to exploit both spatial feature hierarchies and feature correlations. Model validation was conducted using five-fold cross-validation and an independent test dataset, with accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) metrics. Results: Among all the models tested, the hybrid CNN-LSTM model performed the best, with an accuracy of 88.1% and an AUC of 0.93, outperforming conventional ML approaches and single-model DL architectures. Explainability analysis showed that the radiomic features of tumor heterogeneity and morphology had the most prominent impact on model performance. Conclusions: These findings indicate that the combination of radiomic features with hybrid DL models is capable of making non-invasive predictions of glioma grade and 1p/19q codeletion status. The new computational model has the potential to be used as a supplementary approach in precision neuro-oncology.