A multimodal metaheuristic-optimized deep learning framework for radiological image classification

doi:10.1016/j.array.2026.101055

A multimodal metaheuristic-optimized deep learning framework for radiological image classification

2026-07-03

0

OA

AI

Said Jadid Abdulkadir *

DOI：10.1016/j.array.2026.101055

Original

Save

View PDF

Abstract

En 中文

• A novel deep learning model, Covid_L3GBNet23, is proposed for radiological image classification. • A multimodal metaheuristic optimization approach (GA + BDA) is used for effective feature selection. • The framework improves feature representation through pretraining and deep feature extraction. • The proposed method outperforms traditional classifiers, including SVM and KNN variants. • The framework shows strong potential for reliable and automated medical image diagnosis.

Keywords:

Classification

Binary dragonfly algorithm

Genetic algorithm

COVID-19

CNN

Ultrasound

A multimodal metaheuristic-optimized deep learning framework for radiological image classification

Abstract

AI Summary

Journal

Organization