Abstract In this dataset, we present supplementary material from the paper by Bujny et al. (2024). The material includes STL models of aorta segmented in contrast Computed Tomography (CT) scans by a medical expert and using the Machine Learning (ML) model described in the paper. A gallery of ML-based segmentations of the aortic root in non-contrast CT, with the corresponding scans and aortic valve meshes registered from contrast CT using the Iterative Closest Point (ICP) algorithm is provided, as well. Finally, we include the source code of the ICP-based accuracy evaluation method proposed in the paper, with an example in Jupyter Notebook. All of the data included in the dataset has been generated based on the CT scans from the openly available orCaScore dataset (Wolterink et al. 2016). Dataset organization The root folder contains 3 catalogs: contrast_aorta_stls – set of 19 Ground Truth (GT) segmentations of aorta in contrast CT, with the corresponding inferences of the contrast ML model described in the paper, for the scans from the open-source orCaScore dataset (Wolterink et al. 2016). icp_code – Python code of the ICP-based accuracy evaluation method proposed in the paper. icp_gallery – HTML gallery of aorta segmentations based on the non-contrast ML model described in the paper, with the corresponding CT scans and meshes of the aortic valves segmented in contrast CT, which were rigidly registered to the non-contrast scans using the proposed ICP-based approach. In the CT scans, blue and red contours were used to depict non-contrast-based and registered contrast-based ML segmentations, respectively. References M. Bujny, K. Jesionek, J. Nalepa, T. Bartczak, K. Miszalski-Jamka, M. Kostur, “Seeing the Invisible: On Aortic Valve Reconstruction in Non-Contrast CT,” 27th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Marrakesh, Morocco, 2024 [accepted]. J. M. Wolterink et al., “An evaluation of automatic coronary artery calcium scoring methods with cardiac CT using the orCaScore framework: Evaluation of cardiac CT-based automatic coronary calcium scoring,” Med. Phys., vol. 43, no. 5, pp. 2361–2373, 2016.