Hippocampus Segmentation

Automatic segmentation of deep brain structures in MR images has attracted considerable scientific attention due to the widespread use of MRI and to the principal role of some structures in various mental disorders. More specifically, morphological analysis of the hippocampus (HC) and amygdala (AG) is considered a key requirement for the assessment, treatment and follow-up of various mental disorders, including Major Depressive Disorder (MDD), Post-Traumatic Stress Disorder (PTSD), schizophrenia (SD), Alzheimer’s Disease, Bipolar disorder (BD), etc. In the literature, there exist a substantial amount of work relying on deformable models incorporating prior knowledge about structures’ anatomy and shape information. However,  shape priors capture global shape characteristics and thus fail to model boundaries of varying properties; hippocampus’ boundaries present rich, poor and missing gradient regions. On top of that, shape prior knowledge is blended with image information in the evolution process, through global weighting of the two terms, again neglecting the spatially varying boundary properties, causing segmentation faults. Our work aims to achieve highly accurate HC and AG segmentation in MR images, based on the modeling of boundary properties at each anatomical location and the inclusion of appropriate image information for each of those. Hence, blending of image information and prior knowledge is based on a local weighting map, which mixes gradient information, regional and whole brain statistical information with a multi-atlas based spatial distribution map of the structure’s labels.



Relevant Projects

Internal Research


A dataset of manually segmented hippocampi

Relevant Publications

  1. D. Ataloglou, A. Dimou, D. Zarpalas, P. Daras, "Fast and Precise Hippocampus Segmentation Through Deep Convolutional Neural Network Ensembles and Transfer Learning". Neuroinform (2019). DOI: https://doi.org/10.1007/s12021-019-09417-y
  2. 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, Vol. 2, June 2014, doi: 10.1109/JTEHM.2014.2297953
  3. D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras"Gradient based Reliability Maps for ACM based Segmentation of Hippocampus", IEEE Transactions on BioMedical Engineering, Vol: 61, Issue: 4, April 2014, doi: 10.1109/TBME.2013.2293023
  4. D. Zarpalas, P. Gkontra, P. Daras, N. Maglaveras"Hippocampus segmentation through gradient based reliability maps for local blending of ACM energy terms", IEEE International Symposium on Biomedical Imaging (ISBI), San Francisco, USA, 7-11 April, 2013
  5. 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.
  6. 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
  7. 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.
  8. 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.