Related Experiment Video
Updated: Aug 30, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.9K
Metaheuristic Optimization-Driven Novel Deep Learning Approach for Brain Tumor Segmentation
R Kalpana1, M Anto Bennet1, Abdul Wahab Rahmani2
1Department of Electronics and Communication Engineering, VelTech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, Tamil Nadu 600062, India.
Biomed Research International
|August 29, 2022
Summary
This study introduces an optimized DenseNet-169 model using the Procedure for Lightning Attachment Algorithm (PLA) for accurate brain tumor detection from MRI scans. The novel approach enhances classification accuracy, outperforming existing methods for identifying abnormal brain growths.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Brain tumors are a leading cause of mortality, necessitating accurate and early detection methods.
- Magnetic Resonance Imaging (MRI) is crucial for noninvasive brain tumor visualization and feature extraction.
- Current brain tumor detection algorithms face limitations in quality, sensitivity, and early-stage diagnosis.
Purpose of the Study:
- To develop an advanced system for accurate brain tumor classification using MRI.
- To enhance the performance of Convolutional Neural Network (CNN) models in brain tumor detection.
- To address the limitations of existing algorithms in terms of sensitivity and diagnostic accuracy.
Main Methods:
- Preprocessing of brain MRI images to remove outliers.
- Application of the Procedure for Lightning Attachment Algorithm (PLA) for optimization.
- Utilizing the DenseNet-169 CNN model for feature extraction and classification of brain tumors.
Main Results:
- The proposed system achieved satisfactory accuracy in classifying brain tumors on benchmarked datasets.
- The DenseNet-169 model effectively extracted relevant features from MRI scans.
- The optimized model demonstrated superior performance compared to several existing techniques.
Conclusions:
- The integration of PLA with DenseNet-169 offers a robust and accurate method for brain tumor detection.
- This approach shows significant potential for improving early diagnosis and patient outcomes.
- The validated algorithmic rule provides a reliable tool for identifying aberrant brain growths.
