Hybrid Quantum Neural Network and Imaging for Brain Tumor Classification

Accurate classification of brain tumors from MRI scans is critical for diagnosis and treatment planning, but distinguishing between tumor types can be challenging due to similarities in imaging features. Traditional deep learning models often require substantial computational resources and may struggle to efficiently capture complex feature relationships. As medical imaging datasets continue to grow, there is increasing interest in approaches that improve accuracy while reducing computational complexity. This creates a need for advanced diagnostic models that combine efficient processing with high-performance image analysis.

 

Researchers at Arizona State University have developed a novel hybrid quantum‑classical convolutional neural network (QCNN) for automated MRI‑based brain tumor classification. The system combines a classical convolutional neural network (CNN) with a quantum circuit that leverages quantum principles to enhance classification of glioma, meningioma, and pituitary tumors while improving computational efficiency. The model was trained and validated using a large annotated MRI dataset and evaluated using metrics including accuracy, precision, and recall. On a 3,064‑image T1‑weighted CE‑MRI dataset, the hybrid model achieved a 95% test accuracy, with precision and recall metrics comparable to leading classical CNNs. This integration of classical deep learning and quantum computing offers a promising framework for advanced medical image analysis.

 

This novel hybrid classical-quantum neural network is able to classify brain tumors from MRI scans with high accuracy and improved computational efficiency.