Enhanced Variational Quantum Networks for EarlyAlzheimer’s Detection: a Quantum-Classical Approach

The population aged 60 years is rapid increased to 20% of the total Indian population according to the statistics. Alzheimer is a disease which is a neurogenerative disorder. It is considered as the major cause of dementia. India is in critical scenario that faces increase in the elder population suffering from dementia. Hence, there arise a need for early diagnosis of Alzheimer disease (AD) in order to improve the patient’s mental health condition. Though, the existing diagnostic methods are accurate and fast computing, there arise a need to propose a model that aids in improved accuracy and reducing the computational time for early prediction of AD.The proposed model uses Enhanced variational Quantum Network(VAQ) that aids in improving the potential of AD in terms of accuracy and time complexity is designed to decide the feature extraction by processing the huge voluminous images and then fed the outcome into the classifier model. Alzheimer disease neuroimaging stage I and stage II are the datasets considered for applying VAQ for AD prediction. Both stage I and stage II dataset were combined together in order to process the resource efficiently for the diagnosis performs feature extraction and then the results been fed as input to the Quantum classifier. The classifier processes the data and classify the images as no dement, mild dement, average dement, very mild dement. The results are compared with already existing Convolutional Neural Network model (CNN) and Ensemble Classifier.