Meeting Banner
Abstract #4639

Training Labels for Hippocampal Segmentation Based on the EADC-ADNI Harmonized Hippocampal Protocol

Marina Boccardi 1 , Martina Bocchetta 1 , Felix Morency 2 , Masami Nishikawa 3 , Rossana Ganzola 4 , Michel Grothe 5 , Dominik Wolf 6 , Alberto Redolfi 1 , Michela Pievani 1 , Andreas Fellgiebel 6 , Hiroshi Matsuda 3 , Stefan Teipel 5 , Simon Duchesne 7 , Clifford R. Jack Jr. 8 , Giovanni B. Frisoni 1 , and Luigi Antelmi 1

1 IRCCS Fatebenefratelli, Brescia, Italy, 2 IMEKA, Sherbrooke, Quebec, Canada, 3 Kawamura Gakuen Woman's University, Abiko-City, Japan, 4 Universite Laval, Quebec, Canada, 5 German Center for Neurodegenerative Diseases, Rostock, Germany, 6 Johannes Gutenberg-Universitt, Mainz, Germany, 7 Universite Laval, Quebec, Quebec, Canada, 8 Mayo Clinic, Rochester, Minnesota, United States

The EADC-ADNI Harmonized Protocol (HarP) is a Delphi definition of manual hippocampal segmentation from MRI. Training manual raters, and especially automated segmentation algorithms, requires large datasets representative of wide physiological variability. This work produced a set of 270 hippocampal labels certified to comply with standard HarP segmentation allowing proper training of tracers and algorithms alike.

This abstract and the presentation materials are available to members only; a login is required.

Join Here