The first conference on machine learning for tomographic reconstruction, known as the Deep Learning Workshop, was held at Rensselaer Polytechnic Institute from November 18-19, 2017. This was organized by Prof. Ge Wang’s laboratory, sponsored by the Center for Biotechnology and Interdisciplinary Studies and a NIH/NIGMS Training Program. Among over 50 participants, 17 speakers presented their cutting-edge results. They are from both major research universities and General Electric Global Research, a leading industrial center for tomographic medical imaging.

Deep learning has important applications. Widely-known examples are computer vision and image analysis, for which deep learning improves machine’s capability of recognizing objects in images and generating features from images (images to features). Tomographic imaging is an emerging area of deep learning, for which deep learning transforms externally measured features/data about an object/patient into sectional/volumetric images of the internal structures (features to images). For more details, please take a look at the website: http://www.fully3d.org/deep-reconstruction-workshop.