EQDA-Net: an enhanced quantum-inspired dragonfly algorithm integrated with efficientnet for brain tumor detection
Abstract
Synthesizing quantum-inspired computation with advanced swarm intelligence, this study presents an evolutionary framework for brain tumor detection in Magnetic Resonance Imaging (MRI). Beginning with our baseline Quantum-inspired Dragonfly Algorithm (QDA), which achieved 92.3% accuracy, we describe the enhancements that led to the Enhanced Quantum-inspired Dragonfly Algorithm (EQDA) and its integration with EfficientNet-B4 for classification within the BraTS 2020 evaluation setting. By combining quantum-inspired search mechanisms, multiswarm optimization, adaptive control, attention modules, and multiscale feature fusion, the proposed framework improves optimization efficiency for deep medical image analysis. The experimental results on BraTS 2020 indicate that the revised EQDA-Net configuration achieves strong performance, reaching approximately 97.5% accuracy under the reported split and cross-validation protocol while also reducing convergence iterations relative to the baseline variants considered in this study. These findings support the potential of quantum-inspired optimization for heterogeneous brain MRI analysis; however, independent external validation is still required before broader claims regarding generalizability or clinical applicability can be made.
Related
- Robust Android malware detection through EfficientNet–BERT fusion and Jaya optimization for behavioral analysis
- An EfficientNet hybrid deep learning detection framework of network intrusion with an optimized mixture-of-experts architecture
- AI-Assisted Triage for Diabetic Retinopathy Integrated into Routine Diabetes Care: A Real-World Diagnostic Accuracy Study of the DAIRET® Algorithm
- Validation of a Revised V-Slope Algorithm for Automated First Ventilatory Threshold Detection in Healthy Adults
- An integrated serum peptidomics and multi-algorithm machine learning approach for early diagnosis of lung cancer
BibTeX
@article{helmy2026eqdanet,
title = {{EQDA-Net: an enhanced quantum-inspired dragonfly algorithm integrated with efficientnet for brain tumor detection}},
author = {Salma Mostafa A. Helmy and Salsabil A. El-Regaily and Abeer M. Mahmoud and El-Sayed M. El-Horbaty},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-71478-2},
url = {https://www.nature.com/articles/s41598-026-71478-2},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - EQDA-Net: an enhanced quantum-inspired dragonfly algorithm integrated with efficientnet for brain tumor detection
AU - Salma Mostafa A. Helmy
AU - Salsabil A. El-Regaily
AU - Abeer M. Mahmoud
AU - El-Sayed M. El-Horbaty
PY - 2026
DA - 2026/10/08
JO - Scientific Reports
DO - 10.1038/s41598-026-71478-2
UR - https://www.nature.com/articles/s41598-026-71478-2
AB - Synthesizing quantum-inspired computation with advanced swarm intelligence, this study presents an evolutionary framework for brain tumor detection in Magnetic Resonance Imaging (MRI). Beginning with our baseline Quantum-inspired Dragonfly Algorithm (QDA), which achieved 92.3% accuracy, we describe the enhancements that led to the Enhanced Quantum-inspired Dragonfly Algorithm (EQDA) and its integration with EfficientNet-B4 for classification within the BraTS 2020 evaluation setting. By combining quantum-inspired search mechanisms, multiswarm optimization, adaptive control, attention modules, and multiscale feature fusion, the proposed framework improves optimization efficiency for deep medical image analysis. The experimental results on BraTS 2020 indicate that the revised EQDA-Net configuration achieves strong performance, reaching approximately 97.5% accuracy under the reported split and cross-validation protocol while also reducing convergence iterations relative to the baseline variants considered in this study. These findings support the potential of quantum-inspired optimization for heterogeneous brain MRI analysis; however, independent external validation is still required before broader claims regarding generalizability or clinical applicability can be made.
SN - 2045-2322
ER -