Brain Tumor Detection Using Quantum Neural Network

According to American Society of Clinical Oncology, in 2020, around 300 thousand people were diagnosed with brain or spinal cord tumor(s) worldwide. For effective treatment of such ailments, timely detection is of utmost importance. Currently, doctors detect tumors manually by analyzing the MRI images of a patient. For the most effective treatment, it is expected that doctors crosscheck and re-verify the accuracy of their diagnosis. In the era of machine learning, mathematical models are being used to simplify decision making. In the presented work, a hybrid quantum convolution neural network is being proposed for the detection of brain tumors. The presented work uses publicly available Kaggle dataset(s) of Magnetic Resonance Imaging (MRI) images for the binary classification of brain tumors using hybrid quantum convolution neural networks (HQCNN). There are two types of layers used in the presented model. The quantum layer is used for faster computation and the classical layer takes the output of the quantum layer as input and processes it to produce classification output. The proposed model has achieved an accuracy of about 94%.