Convolutional Neural Network: an overview and application in Radiology

Convolution neural network, a group of artificial neural networks that has become main feature in variety of applications computer vision tasks, is attracting interest on a variety of domains, including radiology. CNN is made to automatically and adaptive modify the spatial areas of features through back propagation by using number of building blocks, such as convolution layers, pooling layers, and fully connected layers. This research article offers a vision on the basics of CNN and its application to various radiological tasks, and discusses its difficulties for implementation and future scope in the field of radiology. Two challenges in using CNN to radio logical task, small database, will be get covered in this paper, also techniques to reduce them being introduced with the concepts and advantages, as well as limitations, of CNN is essential to leverage its potential in research radiology, with the aim of augmenting the features of radiologists and improving medical care.