Comprehensive Benchmarking of CNN-Based Tumor Segmentation Methods Using Multimodal MRI Data

Kavita Kundal,K Venkateswara Rao,Arunabha Majumdar,Neeraj Kumar,Rahul Kumar

DOI: https://doi.org/10.1101/2024.01.22.24301602

2024-01-23

Abstract:Magnetic resonance imaging (MRI) is become an essential and a frontline technique in the detection of brain tumor. However, manual segmentation of tumor from MRI scans is a time-consuming and labour-intensive process. There is a prevalent trend in employing fully automated methods for accurate tumor segmentation using MRI scans. The precision in brain tumor segmentation is essential for the better diagnosis, treatment and prognosis. This study focuses on benchmarking and evaluating the performance of four widely used convolutional neural network (CNN) based brain tumor segmentation methods CaPTk, 2DVNet, EnsembleUNets, and ResNet50. We used 1251 multimodal MRI scans from the BraTS2021 dataset, which encompasses T1, T2, T1ce, and Flair imaging modalities. We compared the performance of these CNN-based methods against a reference set of previously segmented images obtained with the assistance of radiologists. The evaluation encompasses direct utilization of segmented outputs and also by employing radiomic features. Performance evaluation with direct segments method using Dice Similarity Coefficient score (DSC) and Hausdorff Distance (HD) suggested better performance of EnsembleUNets with DSC and HD of 0.93 and 18 respectively, outperforming the other methods. Comparative analysis using radiomics features also revealed that EnsembleUNets is the most precise segmentation method as compared to CaPTk, 2DVNet, and ResNet50. EnsembleUNets achieved Concordance Correlation Coefficient (CCC), Total Deviation Index (TDI), and Root Mean Square Error (RMSE) of 0.79, 1.14 and 0.53 respectively and outperformed its counterparts. These findings contribute valuable insight into the comparative efficacy of EnsembleUNets, facilitating informed decision for accurate brain tumor segmentation.

Radiology and Imaging

What problem does this paper attempt to address?

This paper aims to evaluate and compare four convolutional neural network (CNN) based brain tumor segmentation methods: CaPTk, 2DVNet, EnsembleUNets, and ResNet50. These methods use multi-modal MRI data for tumor segmentation to improve the accuracy of diagnosis, treatment, and prognosis. In this study, the authors used the BraTS2021 dataset, which includes 1251 MRI scans with four imaging modes: T1, T2, T1ce, and Flair, and compared them with pre-segmentation images assisted by radiologists. By comparing the direct segmentation output and radiomics features, the study found that EnsembleUNets performed the best in terms of Dice similarity coefficient (DSC) and Hausdorff distance (HD), with values of 0.93 and 18 respectively, outperforming other methods. In the analysis of radiomics features, EnsembleUNets also demonstrated the highest accuracy with a high concordance correlation coefficient (CCC), total deviation index (TDI), and root mean square error (RMSE). The keywords of the paper include brain tumor segmentation, multi-modal MRI, radiomics, convolutional neural networks, and radiomics features. The research results show that EnsembleUNets has high accuracy and robustness in brain tumor segmentation, which is important for early detection, treatment planning, and improvement of patient survival outcomes.

Comprehensive Benchmarking of CNN-Based Tumor Segmentation Methods Using Multimodal MRI Data

Deep Learning-Based Automatic Segmentation for Size and Volumetric Measurement of Breast Cancer on Magnetic Resonance Imaging.

Ensemble 3D CNN and U-Net-based brain tumour classification with MKKMC segmentation

Multimodal CNN Networks for Brain Tumor Segmentation in MRI: A BraTS 2022 Challenge Solution

Automated brain tumor diagnostics: Empowering neuro-oncology with deep learning-based MRI image analysis

XAI-MRI: An Ensemble Dual-Modality Approach for 3D Brain Tumor Segmentation Using Magnetic Resonance Imaging

Ensemble classification and segmentation for intracranial metastatic tumors on MRI images based on 2D U-nets

Image-based MRI detection of brain tumours using convolutional neural networks

Mask region-based convolutional neural network and VGG-16 inspired brain tumor segmentation

A Novel End-to-end Brain Tumor Segmentation Method Using Improved Fully Convolutional Networks.

Brain Tumor MRI Classification Using a Novel Deep Residual and Regional CNN

Comprehensive Review on MRI-Based Brain Tumor Segmentation: A Comparative Study from 2017 Onwards

MRI tumor segmentation with densely connected 3D CNN

Robust clinical applicable CNN and U-Net based algorithm for MRI classification and segmentation for brain tumor

Automatic Segmentation of MRI of Brain Tumor Using Deep Convolutional Network

Brain tumor segmentation using neuro-technology enabled intelligence-cascaded U-Net model

Brain Tumor Segmentation from MRI Images Using Handcrafted Convolutional Neural Network

Two-headed UNetEfficientNets for parallel execution of segmentation and classification of brain tumors: incorporating postprocessing techniques with connected component labelling

Ensemble Technique for Brain Tumor Patient Survival Prediction

Brain tumor segmentation based on deep learning and an attention mechanism using MRI multi-modalities brain images

Deep learning-based ensemble model for brain tumor segmentation using multi-parametric MR scans