An open, clinically-validated database of 3D+t cine-MR images of the left ventricle with associated manual and automated segmentations Release 1.00 L. Najman 1 , J. Cousty 1 , M. Couprie 1 , H. Talbot 1 , S. Cl ´ ement-Guinaudeau 2,3,4 , T. Goissen 2,3,4 and J. Garot 2,3,4 June 28, 2007 1 Institut Gaspard-Monge, Laboratoire A2SI, Groupe ESIEE, Cit´ e Descartes, BP99, 93162 Noisy-le-Grand Cedex France 2 F´ ed´ eration de Cardiologie 3 Unit´ e INSERM 660, Facult´ e de M´ edecine de Cr´ eteil 4 Mondor University Hospital, Assistance-Publique-Hˆ opitaux de Paris, Cr´ eteil, France. Abstract In this paper, we describe a database of cine-MR (3D+t) images of the left ventricle. This database contains the voxel data, one automated and two manual segmentations for each sequence of images. The segmentations are validated from a clinical point of view. We detail how the images were obtained, as well as how the associated segmentations were performed. We also provide the data clinical validation process. This database, including tools to compute quantitative measures and the software package used to obtain the automated segmentation, is freely available for research purposes. The address of the site is http://laurentnajman.org/heart . Contents 1 Introduction 2 2 MR images acquisition 2 3 Methodologies for segmentation 3 3.1 Preprocessing .......................................... 3 3.2 Automated 4D segmentation .................................. 3 3.3 Manual segmentation ...................................... 4 4 Validation of the segmentations 4 4.1 Quantitative assessment ..................................... 4 Accuracy: point-to-surface measurement ............................ 4