Related Experiment Video
Updated: Jun 27, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.8K
Multi-level brain tumor classification using hybrid coot flamingo search optimization Algorithm Enabled deep learning
Jayasri Kotti1, Manikandan Moovendran2, Mekala Kandasamy3
1Department of Information Technology, GMR Institute of Technology, Rajam, Andhra Pradesh, India.
Network (Bristol, England)
|April 26, 2024
Summary
This study introduces a deep learning method for brain tumor (BT) classification using optimized SpinalNet and CootFSOA-LinkNet for accurate segmentation and multi-type tumor identification.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Brain tumor (BT) detection and classification are critical for effective treatment planning.
- Accurate segmentation of tumor regions from MRI is essential for precise diagnosis.
- Existing deep learning models require optimization for improved performance in BT analysis.
Purpose of the Study:
- To propose an innovative multi-level BT classification approach using deep learning.
- To develop a novel segmentation method for isolating brain tumor areas.
- To optimize deep learning model structures and hyperparameters for enhanced accuracy.
Main Methods:
- Utilized Adaptive Kalman Filter (AKF) for MRI image denoising.
- Employed CootFSOA-LinkNet for tumor segmentation and LinkNet structural optimization.
- Implemented CootFSOA-SpinalNet for multi-level BT classification and hyperparameter tuning.
Main Results:
- Achieved superior performance in BT classification with an accuracy of 0.926.
- Demonstrated high efficacy with a True Positive Rate (TPR) of 0.931 and True Negative Rate (TNR) of 0.925.
- Successfully classified different brain tumor types including gliomas, pituitary tumors, and meningiomas.
Conclusions:
- The proposed CootFSOA-SpinalNet deep learning approach significantly improves brain tumor detection and classification accuracy.
- The novel segmentation technique effectively isolates tumor regions, aiding in precise diagnosis.
- This method offers a promising advancement in automated brain tumor analysis using medical imaging.
