Medical Image Denoising using Convolutional Autoencoder with Shortcut Connections

In recent times, medical image analysis has gained immense attention among researchers for diagnosing and treating deadly diseases, e.g., Cancer. Denoising of Images is considered as a crucial process in medical image analysis. Several Deep Learning techniques are effectively used for image denoising including Autoencoders. In this paper, a convolutional autoencoders based approach with shortcut connections is proposed for medical image denoising. The proposed approach is evaluated by using three medical images datasets. The results demonstrated that the proposed approach outperforms the current cutting-edge methods of medical image denoising on all the three datasets.