Rechercher

[PTC18a] Handling Missing Annotations for Semantic Segmentation with Deep ConvNets.

Atelier, Poster ou Démonstration dans une Conférence Internationale : 4th Workshop on Deep Learning in Medical Image Analysis, MICCAI, September 2018, Spain,

Mots clés: Medical Images, Deep Learning, Convolutional Neural Networks, Incomplete Ground Truth Annotation, Noisy Labels, Missing Labels.

Résumé: Annotation of medical images for semantic segmentation is a very time consuming and dicult task. Moreover, clinical experts of- ten focus on speci c anatomical structures and thus, produce partially annotated images. In this paper, we introduce SMILE, a new deep con- volutional neural network which addresses the issue of learning with in- complete ground truth. SMILE aims to identify ambiguous labels in or- der to ignore them during training, and don't propagate incorrect or noisy information. A second contribution is SMILEr which uses SMILE as initialization for automatically relabeling missing annotations, using a curriculum strategy. Experiments on 3 organ classes (liver, stomach, pancreas) show the relevance of the proposed approach for semantic seg- mentation: with 70% of missing annotations, SMILEr performs similarly as a baseline trained with complete ground truth annotations.

BibTeX

@inproceedings {
PTC18a,
title="{Handling Missing Annotations for Semantic Segmentation with Deep ConvNets.}",
author=" O. Petit and N. Thome and A. Charnoz and A. Hostettler and L. Soler ",
booktitle="{4th Workshop on Deep Learning in Medical Image Analysis, MICCAI}",
year=2018,
month="September",
address=" Spain",
}