Biological data annotation via a human-augmenting AI-based labeling system

Abstract Biology has become a prime area for the deployment of deep learning and artificial intelligence (AI), enabled largely by the massive data sets that the field can generate. Key to most AI tasks is the availability of a sufficiently large, labeled data set with which to train AI models. In th...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Douwe van der Wal, Iny Jhun, Israa Laklouk, Jeff Nirschl, Lara Richer, Rebecca Rojansky, Talent Theparee, Joshua Wheeler, Jörg Sander, Felix Feng, Osama Mohamad, Silvio Savarese, Richard Socher, Andre Esteva
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Computer applications to medicine. Medical informatics R858-859.7
Acceso en línea:	https://doaj.org/article/d6164252a0d94935879eacf682d4bdaf
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d6164252a0d94935879eacf682d4bdaf
record_format	dspace
spelling	oai:doaj.org-article:d6164252a0d94935879eacf682d4bdaf2021-12-02T17:13:14ZBiological data annotation via a human-augmenting AI-based labeling system10.1038/s41746-021-00520-62398-6352https://doaj.org/article/d6164252a0d94935879eacf682d4bdaf2021-10-01T00:00:00Zhttps://doi.org/10.1038/s41746-021-00520-6https://doaj.org/toc/2398-6352Abstract Biology has become a prime area for the deployment of deep learning and artificial intelligence (AI), enabled largely by the massive data sets that the field can generate. Key to most AI tasks is the availability of a sufficiently large, labeled data set with which to train AI models. In the context of microscopy, it is easy to generate image data sets containing millions of cells and structures. However, it is challenging to obtain large-scale high-quality annotations for AI models. Here, we present HALS (Human-Augmenting Labeling System), a human-in-the-loop data labeling AI, which begins uninitialized and learns annotations from a human, in real-time. Using a multi-part AI composed of three deep learning models, HALS learns from just a few examples and immediately decreases the workload of the annotator, while increasing the quality of their annotations. Using a highly repetitive use-case—annotating cell types—and running experiments with seven pathologists—experts at the microscopic analysis of biological specimens—we demonstrate a manual work reduction of 90.60%, and an average data-quality boost of 4.34%, measured across four use-cases and two tissue stain types.Douwe van der WalIny JhunIsraa LakloukJeff NirschlLara RicherRebecca RojanskyTalent ThepareeJoshua WheelerJörg SanderFelix FengOsama MohamadSilvio SavareseRichard SocherAndre EstevaNature PortfolioarticleComputer applications to medicine. Medical informaticsR858-859.7ENnpj Digital Medicine, Vol 4, Iss 1, Pp 1-7 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Computer applications to medicine. Medical informatics R858-859.7
spellingShingle	Computer applications to medicine. Medical informatics R858-859.7 Douwe van der Wal Iny Jhun Israa Laklouk Jeff Nirschl Lara Richer Rebecca Rojansky Talent Theparee Joshua Wheeler Jörg Sander Felix Feng Osama Mohamad Silvio Savarese Richard Socher Andre Esteva Biological data annotation via a human-augmenting AI-based labeling system
description	Abstract Biology has become a prime area for the deployment of deep learning and artificial intelligence (AI), enabled largely by the massive data sets that the field can generate. Key to most AI tasks is the availability of a sufficiently large, labeled data set with which to train AI models. In the context of microscopy, it is easy to generate image data sets containing millions of cells and structures. However, it is challenging to obtain large-scale high-quality annotations for AI models. Here, we present HALS (Human-Augmenting Labeling System), a human-in-the-loop data labeling AI, which begins uninitialized and learns annotations from a human, in real-time. Using a multi-part AI composed of three deep learning models, HALS learns from just a few examples and immediately decreases the workload of the annotator, while increasing the quality of their annotations. Using a highly repetitive use-case—annotating cell types—and running experiments with seven pathologists—experts at the microscopic analysis of biological specimens—we demonstrate a manual work reduction of 90.60%, and an average data-quality boost of 4.34%, measured across four use-cases and two tissue stain types.
format	article
author	Douwe van der Wal Iny Jhun Israa Laklouk Jeff Nirschl Lara Richer Rebecca Rojansky Talent Theparee Joshua Wheeler Jörg Sander Felix Feng Osama Mohamad Silvio Savarese Richard Socher Andre Esteva
author_facet	Douwe van der Wal Iny Jhun Israa Laklouk Jeff Nirschl Lara Richer Rebecca Rojansky Talent Theparee Joshua Wheeler Jörg Sander Felix Feng Osama Mohamad Silvio Savarese Richard Socher Andre Esteva
author_sort	Douwe van der Wal
title	Biological data annotation via a human-augmenting AI-based labeling system
title_short	Biological data annotation via a human-augmenting AI-based labeling system
title_full	Biological data annotation via a human-augmenting AI-based labeling system
title_fullStr	Biological data annotation via a human-augmenting AI-based labeling system
title_full_unstemmed	Biological data annotation via a human-augmenting AI-based labeling system
title_sort	biological data annotation via a human-augmenting ai-based labeling system
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/d6164252a0d94935879eacf682d4bdaf
work_keys_str_mv	AT douwevanderwal biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT inyjhun biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT israalaklouk biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT jeffnirschl biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT lararicher biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT rebeccarojansky biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT talenttheparee biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT joshuawheeler biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT jorgsander biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT felixfeng biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT osamamohamad biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT silviosavarese biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT richardsocher biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem AT andreesteva biologicaldataannotationviaahumanaugmentingaibasedlabelingsystem
_version_	1718381409325809664

Biological data annotation via a human-augmenting AI-based labeling system

Ejemplares similares