Implementing QKD-Encrypted Hybrid Quantum Machine Learning for Dementia Detection
By Ryan Kim
Magnetic resonance imaging (MRI) is a common technique to scan brains for strokes, tumors, and other abnormalities that cause forms of dementia. However, correctly diagnosing forms of dementia from MRIs is difficult, leading to misdiagnoses, an issue neural networks can rectify. The performance of these neural networks has been shown to be improved by applying quantum algorithms. This proposed approach takes advantage of a classical convolutional neural network (CNN) to extract features, and then uses a quantum support vector machine (QSVM) to classify a given instance. This study hopes to improve the accuracy and efficiency of classical CNNs. With hospitals beginning to adopt machine learning applications for biomedical image detection, this proposed architecture could improve accuracies and prevent more misdiagnoses. Additionally, to address privacy concerns over sensitive patient data, quantum key distribution (QKD) for image encryption is employed before classification. Furthermore, the proposed architecture is flexible and can be used for transfer-learning tasks, saving time and resources.