Investigating certain choices of CNN configurations for brain lesion segmentation

TL;DR

This study systematically evaluates CNN configurations, specifically U-Net and DeepMedic, for MRI brain tumor segmentation, identifying optimal parameters to improve accuracy and generalization in clinical applications.

Contribution

It provides a comprehensive analysis of CNN design choices, such as loss functions and training parameters, for brain tumor segmentation, highlighting the superior performance of U-Net with specific settings.

Findings

U-Net outperformed DeepMedic in DSC scores.

Adam optimizer yielded better results for U-Net, SGD for DeepMedic.

Weighted soft Dice and cross-entropy loss improved stability and performance.

Abstract

Brain tumor imaging has been part of the clinical routine for many years to perform non-invasive detection and grading of tumors. Tumor segmentation is a crucial step for managing primary brain tumors because it allows a volumetric analysis to have a longitudinal follow-up of tumor growth or shrinkage to monitor disease progression and therapy response. In addition, it facilitates further quantitative analysis such as radiomics. Deep learning models, in particular CNNs, have been a methodology of choice in many applications of medical image analysis including brain tumor segmentation. In this study, we investigated the main design aspects of CNN models for the specific task of MRI-based brain tumor segmentation. Two commonly used CNN architectures (i.e. DeepMedic and U-Net) were used to evaluate the impact of the essential parameters such as learning rate, batch size, loss function, and optimizer. The performance of CNN models using different configurations was assessed with the BraTS 2018 dataset to determine the most performant model. Then, the generalization ability of the model was assessed using our in-house dataset. For all experiments, U-Net achieved a higher DSC compared to the DeepMedic. However, the difference was only statistically significant for whole tumor segmentation using FLAIR sequence data and tumor core segmentation using T1w sequence data. Adam and SGD both with the initial learning rate set to 0.001 provided the highest segmentation DSC when training the CNN model using U-Net and DeepMedic architectures, respectively. No significant difference was observed when using different normalization approaches. In terms of loss functions, a weighted combination of soft Dice and cross-entropy loss with the weighting term set to 0.5 resulted in an improved segmentation performance and training stability for both DeepMedic and U-Net models.