A framework for CT image segmentation inspired by the clinical environment Marie Kloenne 1,2⋆ , Sebastian Niehaus 1,3⋆ , Leonie Lampe 1 , Alberto Merola 1 , Janis Reinelt 1 , and Nico Scherf 3,4 1 AICURA medical, Bessemerstrasse 22, 12103 Berlin, Germany firstname.lastname@aicura-medical.com 2 Technische Fakult¨ at, Universit¨ at Bielefeld, Universit¨ atsstrasse 25, 33615 Bielefeld, Germany 3 Institute for Medical Informatics and Biometry, Carl Gustav Carus Faculty of Medicine, Technische Universit¨at Dresden, Fetscherstrasse 74, 01307 Dresden, Germany 4 Max Planck Institute for Human Cognitive and Brain Sciences, Stephanstrasse 1a, 04103 Leipzig, Germany Abstract. Computed tomography (CT) data poses many challenges to medical image segmentation based on convolutional neural networks (CNNs). The main issues arise during feature extraction, due to the large dynamic range of intensities and the varying number of recorded slices of CT volumes. In this paper we address these issues with a frame- work that combines domain-specific data pre-processing and augmenta- tion with state-of-the-art CNN architectures. The focus is not limited to score optimization, but also to stabilize the achieved prediction perfor- mance, since this is a mandatory requirement for use in automated and semi-automated workflows in the clinical environment. The framework is validated contextually to an architecture comparison to show CNN architecture independent effects of our framework func- tionality. This comparison includes a modified U-Net and a modified Mixed-Scale Dense Network (MS-D Net) to compare dilated convolu- tions for parallel multi-scale processing to the U-Net approach based on traditional scaling operations. Finally, in order to combine the superior recognition performance of 2D-CNN models with the more comprehen- sive spatial information of 3D-CNN models, we propose an ensemble model. The framework performs successfully when tested on a range of tasks such as liver and kidney segmentation, without significant differences in prediction performance on strongly differing volume sizes and varying slice thickness. Keywords: Medical image segmentation · Computed Tomography (CT) · Kidney tumor segmentation · Liver segmentation ⋆ The authors contributed equally to this paper. arXiv:1907.10132v2 [eess.IV] 27 Sep 2019