An Ensemble of 3D U-Net Based Models for Segmentation of Kidney and Masses in CT Scans

Alex Golts, Daniel Khapun, Daniel Shats, Yoel Shoshan, Flora Gilboa-Solomon

Keywords: Semantic segmentation, medical imaging, 3D U-Net, kidney tumor

Abstract: Automatic segmentation of renal tumors and surrounding anatomy in computed tomography (CT) scans is a promising tool in assisting radiologists and surgeons in their efforts to study these scans and improve the prospect of treating kidney cancer. In this paper we describe our approach to compete in the 2021 Kidney and Kidney Tumor Segmentation (KiTS21) challenge. Our approach is based on the successful 3D U-Net architecture with our added novelties including the use of transfer learning, an unsupervised regularized loss, custom postprocessing and multi-annotator ground truth that mimics the evaluation protocol.