Hybrid Quantum–Classical Neural Network for Brain Tumor Classification Using MRI: Methodological Design and Proposed Framework
Contributors
Dr. Matam Santoshi kumari
Keywords
Proceeding
Track
General Track
License
Copyright (c) 2026 Sustainable Global Societies Initiative

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Abstract
Brain tumor classification from magnetic resonance imaging (MRI) is an important medical image analysis task that supports early diagnosis and treatment planning. This paper presents a proposed hybrid quantum–classical neural network framework for multiclass brain tumor classification using MRI images. The framework integrates EfficientNet-B2 as a classical feature extractor, followed by dimensionality reduction from 512 to 64 features. The reduced 64-dimensional representation is processed through two parallel branches: a classical pathway that retains the 64-dimensional features and a quantum pathway that maps the reduced features into four quantum features using four-qubit angle encoding and a trainable Variational Quantum Circuit (VQC). The quantum measurement outputs are concatenated with the classical features to form a 68-dimensional feature representation (64 classical features + 4 quantum features), which is subsequently processed by a fully connected classifier with softmax activation to predict four categories: glioma, meningioma, pituitary tumor, and no tumor. The proposed methodology outlines the integration of classical deep learning and variational quantum processing for medical image classification. Performance evaluation is planned using accuracy, precision, recall, F1-score, and confusion matrix analysis. As this work is currently at the methodological design stage, experimental results and claims of quantum advantage are not presented. The framework provides a foundation for future implementation and comparative evaluation with conventional deep learning approaches.