@inproceedings{56df7789bc9f4e74916ed999f8c52e3e,
title = "Artificially enlarged training set in image segmentation",
abstract = "Due to small training sets, statistical shape models constrain often too much the deformation in medical image segmentation. Hence, an artificial enlargement of the training set has been proposed as a solution for the problem. In this paper, the error sources in the statistical shape model based segmentation were analyzed and the optimization processes were improved. The method was evaluated with 3D cardiac MR volume data. The enlargement method based on non-rigid movement produced good results – with 250 artificial modes, the average error for four-chamber model was 2.11 mm when evaluated using 25 subjects.",
keywords = "Statistical shape models, segmentation, MR images",
author = "Tuomas T{\"o}lli and Juha Koikkalainen and Kirsi Lauerma and Jyrki L{\"o}tj{\"o}nen",
year = "2006",
doi = "10.1007/11866565_10",
language = "English",
isbn = "978-3-540-44707-8",
volume = "1",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "75--82",
editor = "Rasmus Larsen and Mads Nielsen and Jon Sporring",
booktitle = "Medical Image Computing and Computer-Assisted Intervention – MICCAI 2006",
address = "Germany",
note = "9th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI2006) ; Conference date: 01-10-2006 Through 06-10-2006",
}