Quantum Methods for Neural Networks and Application to Medical Image Classification

Quantum machine learning techniques have been proposed as a way to potentially enhance performance in machine learn-ing applications. In this paper, we intro-duce two new quantum methods for neural networks. The first method is quantum -assisted neural networks, where a quan-tum computer is used to perform inner product estimation for inference and train-ing of classical neural networks. The sec-ond is a quantum orthogonal neural net-work, which is based on a quantum pyra-midal circuit as the building block for im-plementing orthogonal matrix multiplica-tion. We provide an efficient way for train-ing such orthogonal neural networks; novel algorithms are detailed for both classical and quantum hardware, where both are proven to scale asymptotically better than previously known training algorithms. Extensive experiments applied to medi-cal image classification tasks using current state of the art quantum hardware are presented, where different quantum meth-ods are compared with classical ones, on both real quantum hardware and simula-tors. Our results show that the proposed quantum networks generate similar level of accuracy compared with classical neu-ral networks while demonstrating compet-itive scalability, supporting the promise that quantum methods can be useful in solving visual tasks, given the advent of better quantum hardware.