Handling Missing Annotations for Semantic Segmentation with Deep ConvNets

Abstract : Annotation of medical images for semantic segmentation is a very time consuming and difficult task. Moreover, clinical experts often focus on specific 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 incomplete ground truth. SMILE aims to identify ambiguous labels in order 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.
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02471187
Contributor : Olivier Petit <>
Submitted on : Friday, February 7, 2020 - 6:19:49 PM
Last modification on : Thursday, February 13, 2020 - 1:26:45 AM

File

handling_missing_annotations_D...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02471187, version 1

Collections

Citation

Olivier Petit, Nicolas Thome, Arnaud Charnoz, Alexandre Hostettler, Luc Soler. Handling Missing Annotations for Semantic Segmentation with Deep ConvNets. MICCAI 2018 workshop DLMIA, Sep 2018, Grenade, Spain. ⟨hal-02471187⟩

Share

Metrics

Record views

7

Files downloads

7