About: A Weakly Supervised Region-Based Active Learning Method for COVID-19 Segmentation in CT Images

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: A Weakly Supervised Region-Based Active Learning Method for COVID-19 Segmentation in CT Images Goto Sponge NotDistinct Permalink

An Entity of Type : schema:ScholarlyArticle, within Data Space : wasabi.inria.fr associated with source document(s)

Attributes	Values
type	Academic Article research paper schema:ScholarlyArticle
isDefinedBy	Covid-on-the-Web dataset
title	A Weakly Supervised Region-Based Active Learning Method for COVID-19 Segmentation in CT Images
Creator	Parker, William Com, Ai, Element Ai, Xtract Atighehchian, Parmida Branchaud-Charron, Frederic Laradji, Issam Lensink, Keegan Nowrouzezahrai, Derek Rodriguez, Pau Sapienml, Vazquez, David
source	ArXiv
abstract	One of the key challenges in the battle against the Coronavirus (COVID-19) pandemic is to detect and quantify the severity of the disease in a timely manner. Computed tomographies (CT) of the lungs are effective for assessing the state of the infection. Unfortunately, labeling CT scans can take a lot of time and effort, with up to 150 minutes per scan. We address this challenge introducing a scalable, fast, and accurate active learning system that accelerates the labeling of CT scan images. Conventionally, active learning methods require the labelers to annotate whole images with full supervision, but that can lead to wasted efforts as many of the annotations could be redundant. Thus, our system presents the annotator with unlabeled regions that promise high information content and low annotation cost. Further, the system allows annotators to label regions using point-level supervision, which is much cheaper to acquire than per-pixel annotations. Our experiments on open-source COVID-19 datasets show that using an entropy-based method to rank unlabeled regions yields to significantly better results than random labeling of these regions. Also, we show that labeling small regions of images is more efficient than labeling whole images. Finally, we show that with only 7/% of the labeling effort required to label the whole training set gives us around 90/% of the performance obtained by training the model on the fully annotated training set. Code is available at: /url{https://github.com/IssamLaradji/covid19_active_learning}.
has issue date	2020-07-07(xsd:dateTime)
has license	arxiv
sha1sum (hex)	0ce1fa5894959f4cdf4db416dd0d3e0dc5571809
resource representing a document's title	A Weakly Supervised Region-Based Active Learning Method for COVID-19 Segmentation in CT Images
resource representing a document's body	covid:0ce1fa5894959f4cdf4db416dd0d3e0dc5571809#body_text
is schema:about of	named entity 'key' named entity 'accelerates' named entity 'images' named entity 'https' named entity 'rank' named entity 'Conventionally' named entity 'performance' named entity 'BUT' named entity '150' named entity 'RANDOM' named entity 'ALLOWS' named entity 'BATTLE' named entity 'PERFORMANCE' named entity 'PROMISE' named entity 'HTTPS' named entity 'ACTIVE' covid:arg/0ce1fa5894959f4cdf4db416dd0d3e0dc5571809 named entity 'github.com' named entity 'datasets' named entity 'labeling' named entity 'time' named entity 'Thus' named entity 'annotated' named entity 'annotations' named entity 'lead' named entity 'One' named entity 'scan' named entity 'content' named entity 'labeling' named entity 'fast' named entity 'pixel' named entity 'COVID-19' named entity 'active learning' named entity 'CT scan' named entity 'training set' named entity 'COVID-19' named entity 'probability' named entity 'entropy' named entity 'learning rate' named entity 'COVID-19' named entity 'pixel' named entity 'active learning' named entity 'active learning' named entity 'CTs' named entity 'training set' named entity 'multi-tasking' named entity 'heuristic' named entity 'infection' named entity 'U-Net' named entity 'training set' named entity 'DICOM' named entity 'Active learning' named entity 'infection' named entity 'pixel' named entity 'jYj' named entity 'semantic' named entity 'edge detection' named entity 'active learning' named entity 'entropy' named entity 'pixel' named entity 'deep learning' named entity 'pixel map' named entity 'pancreas' named entity 'similar efforts' named entity 'pixel' named entity 'training set' named entity 'pixel' named entity 'heuristic' named entity 'lung' named entity 'active learning' named entity 'SARS-CoV-2' named entity 'Severe Acute Respiratory Syndrome Coronavirus 2' named entity 'intensive care unit' named entity 'cross-entropy' named entity 'entropy' named entity 'Active learning'

Faceted Search & Find service v1.13.91 as of Mar 24 2020

Alternative Linked Data Documents: Sponger | ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3229 as of Jul 10 2020, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (94 GB total memory)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2025 OpenLink Software