Related Experiment Video
Updated: Aug 31, 2025
05:33
Three-Dimensional Shape Modeling and Analysis of Brain Structures
Published on: November 14, 2019
7.2K
Deep Neural Network-Based Novel Mathematical Model for 3D Brain Tumor Segmentation
Ajay S Ladkat1, Sunil L Bangare2, Vishal Jagota3
1Department of Instrumentation Engineering, Vishwakarma Institute of Technology, Pune, India.
Computational Intelligence and Neuroscience
|August 22, 2022
Summary
This study introduces an automated method using a mathematical model and deep neural networks (DNNs) for precise brain tumor segmentation from MRI scans. The novel approach achieves 98.90% pixel-level accuracy, aiding in better patient care.
Area of Science:
- Neuroscience
- Medical Imaging
- Artificial Intelligence
Background:
- Accurate brain tumor segmentation is crucial for diagnosis, treatment planning, and monitoring.
- Current segmentation methods can be time-consuming and prone to variability.
- Multimodal magnetic resonance imaging (MRI) offers rich data for tumor characterization.
Purpose of the Study:
- To develop a fully automated method for brain tumor segmentation using multimodal MRI.
- To enhance tumor and subregion delineation for improved measurement consistency.
- To contribute a novel approach to neuroscience research in brain tumor analysis.
Main Methods:
- A novel mathematical model was developed for enhancing individual MRI slices.
- A 3D attention U-Net deep neural network was employed for segmentation.
- The system was trained and validated on the BraTS 2019 dataset.
Main Results:
- The automated system achieved a pixel-level accuracy of 98.90% for tumor segmentation.
- Performance was validated against existing state-of-the-art methods.
- Time complexity analysis was conducted on various processing units.
Conclusions:
- The proposed automated segmentation method demonstrates high accuracy and efficiency.
- This approach has the potential to significantly aid in brain tumor patient treatment.
- The study highlights the synergy between mathematical modeling, deep learning, and neuroscience for medical image analysis.
