Cost-Effective Active Learning for Melanoma Segmentation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

Cost-Effective Active Learning for Melanoma Segmentation

Résumé

We propose a novel Active Learning framework capable to train effectively a convolutional neural network for semantic segmentation of medical imaging, with a limited amount of training labeled data. Our contribution is a practical Cost-Effective Active Learning approach using dropout at test time as Monte Carlo sampling to model the pixel-wise uncertainty and to analyze the image information to improve the training performance. The source code of this project is available at this https URL :https://marc-gorriz.github.io/CEAL-Medical-Image-Segmentation/.
Fichier principal
Vignette du fichier
gorriz_22238.pdf (226.65 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02871320 , version 1 (17-06-2020)

Identifiants

  • HAL Id : hal-02871320 , version 1
  • OATAO : 22238

Citer

Marc Gorriz Blanch, Xavier Giro I Nieto, Axel Carlier, Emmanuel Faure. Cost-Effective Active Learning for Melanoma Segmentation. 31st Conference on Machine Learning for Health: Workshop at NIPS 2017 (ML4H 2017), Dec 2017, Long Beach, California, United States. pp.1-5. ⟨hal-02871320⟩
49 Consultations
58 Téléchargements

Partager

Gmail Facebook X LinkedIn More