Related Experiment Video
Updated: May 26, 2025
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Design of an Optimal Convolutional Neural Network Architecture for MRI Brain Tumor Classification by Exploiting
Sofia El Amoury1, Youssef Smili2, Youssef Fakhri1
1Laboratory RI, Faculty of Sciences of Kenitra, Ibn Tofail University, Kenitra 14000, Morocco.
Abstract:
The classification of brain tumors using MRI scans is critical for accurate diagnosis and effective treatment planning, though it poses significant challenges due to the complex and varied characteristics of tumors, including irregular shapes, diverse sizes, and subtle textural differences. Traditional convolutional neural network (CNN) models, whether handcrafted or pretrained, frequently fall short in capturing these intricate details comprehensively. To address this complexity, an automated approach employing Particle Swarm Optimization (PSO) has been applied to create a CNN architecture specifically adapted for MRI-based brain tumor classification. PSO systematically searches for an optimal configuration of architectural parameters-such as the types and numbers of layers, filter quantities and sizes, and neuron numbers in fully connected layers-with the objective of enhancing classification accuracy. This performance-driven method avoids the inefficiencies of manual design and iterative trial and error. Experimental results indicate that the PSO-optimized CNN achieves a classification accuracy of 99.19%, demonstrating significant potential for improving diagnostic precision in complex medical imaging applications and underscoring the value of automated architecture search in advancing critical healthcare technology.
Insights
Particle Swarm Optimization (PSO) automated the design of a Convolutional Neural Network (CNN) for brain tumor classification from MRI scans, achieving 99.19% accuracy. This optimizes diagnostic precision in medical imaging.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Neuroscience
Background:
- Accurate brain tumor classification from MRI scans is crucial for diagnosis and treatment planning.
- Traditional Convolutional Neural Networks (CNNs) struggle with the complex variations in tumor characteristics like shape, size, and texture.
- Manual CNN architecture design is inefficient and prone to errors.
Purpose of the Study:
- To develop an automated approach for designing a CNN architecture tailored for MRI-based brain tumor classification.
- To enhance the accuracy and efficiency of brain tumor classification using optimized CNN models.
Main Methods:
- Employed Particle Swarm Optimization (PSO) to automatically search for optimal CNN architectural parameters.
- Optimized parameters included layer types/numbers, filter sizes/quantities, and fully connected layer neuron counts.
- Evaluated the PSO-optimized CNN's performance on MRI-based brain tumor classification tasks.
Main Results:
- The PSO-optimized CNN achieved a highly accurate classification rate of 99.19%.
- Demonstrated superior performance compared to traditional or manually designed CNN models.
- Validated the effectiveness of automated architecture search for medical imaging applications.
Conclusions:
- Automated CNN architecture design using PSO significantly improves brain tumor classification accuracy from MRI data.
- This approach offers a powerful tool for enhancing diagnostic precision in complex medical imaging.
- Highlights the potential of AI-driven methods in advancing healthcare technology.
