Deep Learning-Enhanced Hybrid Metaheuristic Fusion Model for Biomedical Image Analysis

View Sample PDF

Author(s): A. V. Geetha (Sri Ramachandra Institute of Higher Education and Research, India), M. Lakshmi (SRM Institute of Science and Technology, India)and A. V. Kalpana (SRM Institute of Science and Technology, India)
Copyright: 2026
Pages: 22
Source title: Metaheuristic Algorithms and Optimizing Neural Networks for Biomedical Image Processing
Source Author(s)/Editor(s): Prasanalakshmi Balaji (King Khalid University, Saudi Arabia), K. Martin Sagayam (Karunya Institute of Technology and Sciences, India), Aditi Sharma (Symbiosis International University, India)and Korhen Cengiz (University of Fujairah, UAE)
DOI: 10.4018/979-8-3373-0523-3.ch009

Keywords: Computer Science & IT / Medical Information Science Reference / Neural Networks / Radiology and Imaging

Purchase

View Deep Learning-Enhanced Hybrid Metaheuristic Fusion Model for Biomedical Image Analysis on the publisher's website for pricing and purchasing information.

Abstract

This study introduces the Hybrid Deep Learning with Brain Tumor Analysis (HDLB-BTA), a novel automated biomedical image classification method. HDLB-BTA enhances image quality through preprocessing, segments images using Swin-UNet, and applies a fusion-based feature extraction approach. The classification model's parameters are fine-tuned using a Hybrid Firefly Optimization with Simulated Annealing (HFSA) technique. Gated Recurrent Units (GRUs) are used for robust image classification. Evaluation on BraTS2017 and EBTA datasets demonstrated superior performance, achieving 94.51% and 95.38% accuracy on ISIC 2017 and ISIC 2020 datasets, respectively. Future work will explore scalability, multi-modal data integration, and model interpretability for clinical applications.

The IRMA Community

Research IRM

Deep Learning-Enhanced Hybrid Metaheuristic Fusion Model for Biomedical Image Analysis

Purchase

Abstract

Related Content

IRMA Sponsors