1Malte BARTH, 1Sarel COHEN, 1Gustavo CRIADO, 1Vanja DOSKOČ, 1Timofei KORNEV, 1Tobias FRIEDRICH and 2Raid SAABNI

1 University of Potsdam, Germany

2 The Academic College of Tel Aviv-Yafo, Israel

Abstract

The automatic segmentation of X-ray images of the lung is a crucial tool for medical professionals to have in order to effectively diagnose and treat illnesses. We propose a development environment that may make artificial intelligence available to everyone so that everyone will be able to utilize the tools that we provide. In addition to this, we show that transfer learning may be useful for tackling image segmentation challenges, and that using Gaussian noise can offer a boost to the performance of image segmentation algorithms.

Shares