Comparison of regularization techniques for DCNN-based abdominal aortic aneurysm segmentation

Egileak: Karen López-Linares Román Nerea Lete Urzelai Luis Kabongo Mario Ceresa Grégory Maclair Ainhoa García-Familiar Iván Macía Oliver Miguel Ángel González

Data: 04.04.2018


Abstract

This study compares several state-of-the-art regularization methods applicable to aortic aneurysm segmentation likelihood maps provided by a Deep Convolutional Neural Network (DCNN). These algorithms vary from simple Otsu s thresholding and K-Means clustering, to more complex Level-sets and Conditional Random Fields. Experiments demonstrate that K-means yields the best results for the current application, which poses the question about the need to employ a more sophisticated approach for post-processing the output probability maps.

BIB_text

@Article {
title = {Comparison of regularization techniques for DCNN-based abdominal aortic aneurysm segmentation},
pages = {864-867},
keywds = {
K-means, Level-set, Otsu, aneurysm
}
abstract = {

This study compares several state-of-the-art regularization methods applicable to aortic aneurysm segmentation likelihood maps provided by a Deep Convolutional Neural Network (DCNN). These algorithms vary from simple Otsu s thresholding and K-Means clustering, to more complex Level-sets and Conditional Random Fields. Experiments demonstrate that K-means yields the best results for the current application, which poses the question about the need to employ a more sophisticated approach for post-processing the output probability maps.


}
isbn = {978-153863636-7},
doi = {10.1109/ISBI.2018.8363708},
date = {2018-04-04},
}
Vicomtech

Gipuzkoako Zientzia eta Teknologia Parkea,
Mikeletegi Pasealekua 57,
20009 Donostia / San Sebastián (Espainia)

+(34) 943 309 230

Zorrotzaurreko Erribera 2, Deusto,
48014 Bilbo (Espainia)

close overlay

Jokaeraren araberako publizitateko cookieak beharrezkoak dira eduki hau kargatzeko

Onartu jokaeraren araberako publizitateko cookieak