Implementation of Autoencoder for Super Resolution of 3D MRI Imaging using Convolution Neural Network

In today’s scenario, acquiring high dimensional image modalities is still extensive area of medicinal and clinical research in healthcare field. The need of increasing the quality of medical imaging for the examination by healthcare domain, medicines or clinical analysis. The elucidation can be the provided by super resolution image to demarcate as higher resolution image structure formed from three-dimensional medical image. The recreation of higher dimensional medical image structure to increase resolution can be done through the notion of convolutional neural networks. The proposed methodology will implement the TensorFlow framework for construction of super resolution MRI (Magnetic resonance image) using convolutional neural network (CNN) in order to minimize loss in image. The understanding of deep learning technique using autoencoder approach for higher resolution will provide the wide-ranging anatomical information of threedimensional image for precise clinical analysis for curative handling.