Related Experiment Video
Updated: Mar 24, 2026
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images
Abstract:
Among brain tumors, gliomas are the most common and aggressive, leading to a very short life expectancy in their highest grade. Thus, treatment planning is a key stage to improve the quality of life of oncological patients. Magnetic resonance imaging (MRI) is a widely used imaging technique to assess these tumors, but the large amount of data produced by MRI prevents manual segmentation in a reasonable time, limiting the use of precise quantitative measurements in the clinical practice. So, automatic and reliable segmentation methods are required; however, the large spatial and structural variability among brain tumors make automatic segmentation a challenging problem. In this paper, we propose an automatic segmentation method based on Convolutional Neural Networks (CNN), exploring small 3 ×3 kernels. The use of small kernels allows designing a deeper architecture, besides having a positive effect against overfitting, given the fewer number of weights in the network. We also investigated the use of intensity normalization as a pre-processing step, which though not common in CNN-based segmentation methods, proved together with data augmentation to be very effective for brain tumor segmentation in MRI images. Our proposal was validated in the Brain Tumor Segmentation Challenge 2013 database (BRATS 2013), obtaining simultaneously the first position for the complete, core, and enhancing regions in Dice Similarity Coefficient metric (0.88, 0.83, 0.77) for the Challenge data set. Also, it obtained the overall first position by the online evaluation platform. We also participated in the on-site BRATS 2015 Challenge using the same model, obtaining the second place, with Dice Similarity Coefficient metric of 0.78, 0.65, and 0.75 for the complete, core, and enhancing regions, respectively.
Insights
This study introduces an automated brain tumor segmentation method using Convolutional Neural Networks (CNNs) and small kernels. The approach significantly improves segmentation accuracy on MRI images, outperforming existing methods in challenges.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neuro-oncology
Background:
- Gliomas are aggressive brain tumors with poor prognosis.
- Manual segmentation of brain tumors from MRI is time-consuming and limits clinical application.
- Automated segmentation methods are crucial for precise quantitative analysis and treatment planning.
Purpose of the Study:
- To develop an automated and reliable method for brain tumor segmentation in MRI.
- To explore the efficacy of Convolutional Neural Networks (CNNs) with small kernels for this task.
- To investigate the impact of intensity normalization and data augmentation on segmentation performance.
Main Methods:
- Proposed an automated segmentation method based on Convolutional Neural Networks (CNNs).
- Utilized small 3x3 kernels to enable deeper architectures and reduce overfitting.
- Incorporated intensity normalization and data augmentation as pre-processing steps.
Main Results:
- Achieved first place in the Brain Tumor Segmentation Challenge 2013 (BRATS 2013) for complete, core, and enhancing tumor regions (Dice scores: 0.88, 0.83, 0.77).
- Secured second place in the BRATS 2015 Challenge with Dice scores of 0.78, 0.65, and 0.75.
- Demonstrated the effectiveness of small kernels, intensity normalization, and data augmentation.
Conclusions:
- The proposed CNN-based method offers a robust and accurate solution for brain tumor segmentation in MRI.
- The findings highlight the potential of deep learning for improving oncological patient care.
- The method's success in multiple challenges validates its clinical applicability.
