Loading...
Loading...

Go to the content (press return)

Cost-effective active learning for melanoma segmentation

Author
Górriz, M.; Giro, X.; Carlier, A.; Faure, E.
Type of activity
Presentation of work at congresses
Name of edition
Machine Learning for Health Workshop at NIPS 2017
Date of publication
2017
Presentation's date
2017-12-08
Book of congress proceedings
ML4H: Machine Learning for Health NIPS, Workshop at NIPS 2017
First page
1
Last page
5
Project funding
Multimodal Signal Processing and Machine Learning on Graphs
Repository
http://hdl.handle.net/2117/113801 Open in new window
https://arxiv.org/abs/1711.09168 Open in new window
Abstract
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.
Citation
Górriz, M., Giro, X., Carlier, A., Faure, E. Cost-effective active learning for melanoma segmentation. A: Machine Learning for Health Workshop at NIPS. "ML4H: Machine Learning for Health NIPS, Workshop at NIPS 2017". Long Beach, CA: 2017, p. 1-5.
Group of research
GPI - Image and Video Processing Group
IDEAI-UPC - Intelligent Data Science and Artificial Intelligence Research Center

Participants

  • Górriz, Marc  (author and speaker )
  • Giro Nieto, Xavier  (author and speaker )
  • Carlier, Axel  (author and speaker )
  • Faure, Emmanuel  (author and speaker )

Attachments