Hippocampus segmentation masks from brain MRIs
Segmentation masks of the Hippocampus from 23 randomly selected images from the OASIS dataset
To offer the means and to promote fair comparison between different methods, we encourage other groups to use the same dataset for the evaluation of their segmentation algorithms. Thus, we hereby provide all necessary information about the dataset and manual segmentations used in the evaluation of our segmentation algorithms.
The dataset used during the performance assessment of our algorithms, consists of 23 T1-weighted MR image volumes chosen randomly from the OASIS database (OASIS – Cross-sectional) to include the entire age range of both sexes. The MRI volumes were acquired through the use of a 1.5T Vision scanner and were produced based on the average of four scans of the same individual. The original MRI volumes are publicly available and can be downloaded for free at http://www.oasis-brains.org/. The manual segmentations of the hippocampus , provided to us by a professional radiologist (Angelos Mpaltatzidis M.D., Radiologist) and used for training and experimentally evaluating our algorithms, are also available to download for free after submitting a request to:
An .xls file which contains information on the selected subjects is also available. This file makes matching the original MRI volumes to the corresponding masks easier. The first part of each mask filename contains the SESSION_ID of the subject as defined in the original MRI volume. Every mask is a volume with 2 values representing the hippocampus (HC value=2) and the background (value=0). The radiologist provided us with one side of the structures (either left or right). As the majority of masks were of the right HC, we mirrored those that correspond to the left HC to appear as the right HC. The corresponding MRI volume can be found (after downloading the database) in the SESSION_ID\PROCESSED\MPRAGE\T88_111. To avoid confusion, the names of the original MRI volumes are also provided in the .xls file.
D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras, “Accurate and fully Automatic Hippocampus Segmentation using subject-specific 3D Optimal Local Maps into a hybrid Active Contour Model“, IEEE Journal of Translational Engineering in Health and Medicine, accepted for publication.
D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras, “Gradient based Reliability Maps for ACM based Segmentation of Hippocampus“, IEEE Transactions on BioMedical Engineering, accepted for publication.
D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras, “Segmentation through a local and adaptive weighting scheme, for contour-based blending of image and prior information“, The 25th IEEE International Symposium on Computer-Based Medical Systems (CBMS 2012), Rome, Italy, 20-22 June 2012
D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras, “Hippocampus segmentation by optimizing the local contribution of image and prior terms, through graph cuts and multi-atlas“, IEEE International Symposium on Biomedical Imaging (ISBI 2012), Barcelona, Spain, 2-5 May 2012
D. Zarpalas, A. Zafeiropoulos, P. Daras, N. Maglaveras, “Hippocampus Segmentation using a Local Prior Model on its Boundary“, International Conference on Machine Vision, Image Processing, and Pattern Analysis (ICMVIPPA 2011), Venice, Italy, November 28-30, 2011
D. Zarpalas, A. Zafeiropoulos, P. Daras, N. Maglaveras, M. G. Strintzis, “Brain Structures Segmentation using Optimum Global and Local Weights on Mixing Active Contours and Neighboring Constraints“, International Symposium on Applied Sciences in Biomedical and Communication Technologies, ISABEL 2011, Barcelona, Spain, October 26-29, 2011