Building medical image classifiers with very limited data using segmentation networks

With the availability of big data and GPU computing, deep learning has become a powerful technique which raises the benchmarks of classification and detection challenges in computer vision (Schmidhuber, 2015). Using huge databases such as ImageNet which has more than a million annotated images (Deng et  al., 2009), deep convolutional neural networks (CNNs) such as AlexNet (Krizhevsky et al., 2012), VGGNet (Simonyan and Zisserman, 2014), and GoogLeNet (Szegedy et al., 2015) were proposed with impressive performance in visual pattern recognition.

Source: Medical Image Analysis - August 3, 2018 Category: Radiology Authors: Ken C.L. Wong, Tanveer Syeda-Mahmood, Mehdi Moradi Source Type: research