Learning to estimate label uncertainty for automatic radiology report parsing

Bootstrapping labels from radiology reports has become the scalable alternative to provide inexpensive ground truth for medical imaging. Because of the domain specific nature, state-of-the-art report labeling tools are predominantly rule-based. These tools, however, typically yield a binary 0 or 1 p...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Olatunji, Tobi, Yao, Li
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Olatunji, Tobi
Yao, Li
description Bootstrapping labels from radiology reports has become the scalable alternative to provide inexpensive ground truth for medical imaging. Because of the domain specific nature, state-of-the-art report labeling tools are predominantly rule-based. These tools, however, typically yield a binary 0 or 1 prediction that indicates the presence or absence of abnormalities. These hard targets are then used as ground truth to train image models in the downstream, forcing models to express high degree of certainty even on cases where specificity is low. This could negatively impact the statistical efficiency of image models. We address such an issue by training a Bidirectional Long-Short Term Memory Network to augment heuristic-based discrete labels of X-ray reports from all body regions and achieve performance comparable or better than domain-specific NLP, but with additional uncertainty estimates which enable finer downstream image model training.
doi_str_mv 10.48550/arxiv.1910.00673
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1910_00673</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1910_00673</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-c01e5a81a1de0d6149566647e54ad4d8132a87b7e4acefb0a88a51de8db64a3c3</originalsourceid><addsrcrecordid>eNotj8tOwzAQRb1hgQofwKr-gRQbP-IuUUUBKaib7qOJPakspXE0cVHz95jC6kpzR2fmMPYkxUY7Y8Qz0DV-b-S2DISwtbpnXw0CjXE88Zw4zjmeISMfoMOBX0aPlCGOeeF9Ig6XnEodPScIMQ3ptHDCKVHmE9BcIA_srodhxsf_XLHj_u24-6iaw_vn7rWpoBytvJBowEmQAUWwUm-NtVbXaDQEHZxUL-DqrkYNHvtOgHNgyq4LndWgvFqx9R_25tNOVL6mpf31am9e6gf-WEn9</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Learning to estimate label uncertainty for automatic radiology report parsing</title><source>arXiv.org</source><creator>Olatunji, Tobi ; Yao, Li</creator><creatorcontrib>Olatunji, Tobi ; Yao, Li</creatorcontrib><description>Bootstrapping labels from radiology reports has become the scalable alternative to provide inexpensive ground truth for medical imaging. Because of the domain specific nature, state-of-the-art report labeling tools are predominantly rule-based. These tools, however, typically yield a binary 0 or 1 prediction that indicates the presence or absence of abnormalities. These hard targets are then used as ground truth to train image models in the downstream, forcing models to express high degree of certainty even on cases where specificity is low. This could negatively impact the statistical efficiency of image models. We address such an issue by training a Bidirectional Long-Short Term Memory Network to augment heuristic-based discrete labels of X-ray reports from all body regions and achieve performance comparable or better than domain-specific NLP, but with additional uncertainty estimates which enable finer downstream image model training.</description><identifier>DOI: 10.48550/arxiv.1910.00673</identifier><language>eng</language><subject>Computer Science - Computation and Language ; Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2019-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1910.00673$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1910.00673$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Olatunji, Tobi</creatorcontrib><creatorcontrib>Yao, Li</creatorcontrib><title>Learning to estimate label uncertainty for automatic radiology report parsing</title><description>Bootstrapping labels from radiology reports has become the scalable alternative to provide inexpensive ground truth for medical imaging. Because of the domain specific nature, state-of-the-art report labeling tools are predominantly rule-based. These tools, however, typically yield a binary 0 or 1 prediction that indicates the presence or absence of abnormalities. These hard targets are then used as ground truth to train image models in the downstream, forcing models to express high degree of certainty even on cases where specificity is low. This could negatively impact the statistical efficiency of image models. We address such an issue by training a Bidirectional Long-Short Term Memory Network to augment heuristic-based discrete labels of X-ray reports from all body regions and achieve performance comparable or better than domain-specific NLP, but with additional uncertainty estimates which enable finer downstream image model training.</description><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8tOwzAQRb1hgQofwKr-gRQbP-IuUUUBKaib7qOJPakspXE0cVHz95jC6kpzR2fmMPYkxUY7Y8Qz0DV-b-S2DISwtbpnXw0CjXE88Zw4zjmeISMfoMOBX0aPlCGOeeF9Ig6XnEodPScIMQ3ptHDCKVHmE9BcIA_srodhxsf_XLHj_u24-6iaw_vn7rWpoBytvJBowEmQAUWwUm-NtVbXaDQEHZxUL-DqrkYNHvtOgHNgyq4LndWgvFqx9R_25tNOVL6mpf31am9e6gf-WEn9</recordid><startdate>20191001</startdate><enddate>20191001</enddate><creator>Olatunji, Tobi</creator><creator>Yao, Li</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20191001</creationdate><title>Learning to estimate label uncertainty for automatic radiology report parsing</title><author>Olatunji, Tobi ; Yao, Li</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-c01e5a81a1de0d6149566647e54ad4d8132a87b7e4acefb0a88a51de8db64a3c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Olatunji, Tobi</creatorcontrib><creatorcontrib>Yao, Li</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Olatunji, Tobi</au><au>Yao, Li</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Learning to estimate label uncertainty for automatic radiology report parsing</atitle><date>2019-10-01</date><risdate>2019</risdate><abstract>Bootstrapping labels from radiology reports has become the scalable alternative to provide inexpensive ground truth for medical imaging. Because of the domain specific nature, state-of-the-art report labeling tools are predominantly rule-based. These tools, however, typically yield a binary 0 or 1 prediction that indicates the presence or absence of abnormalities. These hard targets are then used as ground truth to train image models in the downstream, forcing models to express high degree of certainty even on cases where specificity is low. This could negatively impact the statistical efficiency of image models. We address such an issue by training a Bidirectional Long-Short Term Memory Network to augment heuristic-based discrete labels of X-ray reports from all body regions and achieve performance comparable or better than domain-specific NLP, but with additional uncertainty estimates which enable finer downstream image model training.</abstract><doi>10.48550/arxiv.1910.00673</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1910.00673
ispartof
issn
language eng
recordid cdi_arxiv_primary_1910_00673
source arXiv.org
subjects Computer Science - Computation and Language
Computer Science - Computer Vision and Pattern Recognition
title Learning to estimate label uncertainty for automatic radiology report parsing
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T09%3A39%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Learning%20to%20estimate%20label%20uncertainty%20for%20automatic%20radiology%20report%20parsing&rft.au=Olatunji,%20Tobi&rft.date=2019-10-01&rft_id=info:doi/10.48550/arxiv.1910.00673&rft_dat=%3Carxiv_GOX%3E1910_00673%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true