An Approach for Prostate Cancer Detection using Deep Quantum Neural Network
Prostate cancer (PCa) is extreme case malignant cancer between men. The premature detection of PCa is very important for determining whether the patient should take invasive biopsy with probable health impediments. Though, the existing segmentation-based cancer analysis approaches are achieved only less segmentation accuracy and attained improper classification results. To take both segmentation accuracy and classification into consideration, the ultimate aim of this paper is the development of Deep Quantum Neural Network (Deep QNN) for prostate cancer detection. For this purpose, the input image is fed into the preprocessing module using median filter in order to remove noises. The Spine-Generative Adversarial Neural Networks (Spine-GAN) is devised to segment the affected area from noise free images. The data augmentation process is explored to perform the processes, such as rotation, shearing, zooming, cropping, flipping and adjusting the brightness. Then, the Deep QNN is explored to classify whether the image is normal or abnormal. Moreover, the experimental outcome demonstrates that the classification performance of developed Deep QNN outperforms that the existing cancer detection approaches in terms of accuracy, sensitivity and specificity of 0.9483, 0.95 and 0.95.
Keywords - Spine-Generative Adversarial Neural Networks, Deep Quantum Neural Network, Data Augmentation, Prostate Cancer, Median filter.