Creating a large-scale silver corpus from multiple algorithmic segmentations

Krenn, Markus (Computational Imaging Research (CIR) Lab, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria) ; Dorfer, Matthias (Department of Computational Perception, Johannes Kepler University (JKU),Linz, Austria) ; Müller, Henning (University of Applied Sciences and Arts Western Switzerland (HES-SO Valais-Wallis)) ; et al.

Currently, increasingly large medical imaging data sets become available for research and are analysed by a range of algorithms segmenting anatomical structures automatically and interactively. While they provide segmentations on a much larger scale than possible to achieve with expert annotators, they are typically less accurate than experts. We present and compare approaches to estimate segmentations on large imaging data sets based on a small number of expert annotated examples, and algorithmic segmentations on a much larger data set. Results demonstrate that combining algorithmic segmentations is reliably outperforming the average individual algorithm. Furthermore, injecting organ specic reliability assessments of algorithms based on expert annotations improves accuracy compared to standard label fusion algorithms. The proposed methods are particularly relevant in putting the results of large image analysis algorithm benchmarks to long-term use.


Mots-clés:
Type de conférence:
full paper
Faculté:
Economie et Services
Ecole:
HEG VS HES-SO Valais-Wallis - Haute Ecole de Gestion & Tourisme
Institut:
Institut Informatique de gestion
Classification:
Informatique
Adresse bibliogr.:
Munich, Germany, 5-9 October 2015
Date:
Munich, Germany
5-9 October 2015
2015
Pagination:
13 p.
Publié dans
Proceedings of MICCAI 2015
Le document apparaît dans:

Note Le statut de cette notice est : non diffusé

Note: The status of this file is: restricted


 Notice créée le 2016-05-03, modifiée le 2018-08-31

Fichiers:
Télécharger le document
PDF

Évaluer ce document:

Rate this document:
1
2
3
 
(Pas encore évalué)