QuReBrain: Quantum Data Re-Uploaded VQC for MRI Brain Tumor Classification
A Reflection by Ratul Hasan Rabbi
Every researcher remembers their first major stepping stone. Below is a personal update from Ratul regarding his journey contributing to our recent publication at QPAIN 2026.
I am a bit late in sharing this update. I initially hesitated, overthinking whether to post this at all, but I quickly realized that sharing these early steps is an incredibly important part of participating in the broader research community.
With that said, Alhamdulillah, I am very pleased to announce our latest research paper: "QuReBrain: Quantum Data Re-Uploaded VQC for MRI Brain Tumor Classification." It was successfully accepted and presented at the 2026 IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN).
This publication marks my very first formal involvement in academic research. Serving as a co-author and supporting the rigorous research process was a profound learning experience, one that has firmly solidified my dedication to the academic and research track.
Technical Highlights
Our paper explores the fascinating and critical intersection of quantum computing and medical imaging. We developed a hybrid quantum-classical framework specifically designed for efficiency:
- Feature Extraction: Utilizing a lightweight classical CNN.
- Decision-Making: Powered by a Variational Quantum Classifier (VQC) using a data re-uploading strategy.
The Impact: Our model achieves a highly competitive precision rate of 91.48%, but more impressively, it operates with only 26,694 trainable parameters. This drastically reduces computational latency compared to traditional deep learning networks, paving the way for faster, edge-deployable medical diagnostic tools.
🙌 Acknowledgments
I want to dedicate this milestone to, and extend my absolute deepest gratitude towards, Moniruzzaman Sir. He was the true driving force behind this project and did the heavy lifting. As an exceptional mentor, he patiently taught me the complex, foundational concepts of this domain, guided the entire research methodology, and generously gave me the invaluable opportunity to be a part of this work. Without his leadership and willingness to guide a first-time researcher, this would simply not have been possible.
You can view the full paper, including a detailed technical breakdown of the architecture and our findings, in the document viewer below.