Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training

When reading images, radiologists generate text reports describing the findings therein. Current state-of-the-art computer-aided diagnosis tools utilize a fixed set of predefined categories automatically extracted from these medical reports for training. This form of supervision limits the potential...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2022-05
Hauptverfasser: Seibold, Constantin, Reiß, Simon, M Saquib Sarfraz, Stiefelhagen, Rainer, Kleesiek, Jens
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Seibold, Constantin
Reiß, Simon
M Saquib Sarfraz
Stiefelhagen, Rainer
Kleesiek, Jens
description When reading images, radiologists generate text reports describing the findings therein. Current state-of-the-art computer-aided diagnosis tools utilize a fixed set of predefined categories automatically extracted from these medical reports for training. This form of supervision limits the potential usage of models as they are unable to pick up on anomalies outside of their predefined set, thus, making it a necessity to retrain the classifier with additional data when faced with novel classes. In contrast, we investigate direct text supervision to break away from this closed set assumption. By doing so, we avoid noisy label extraction via text classifiers and incorporate more contextual information. We employ a contrastive global-local dual-encoder architecture to learn concepts directly from unstructured medical reports while maintaining its ability to perform free form classification. We investigate relevant properties of open set recognition for radiological data and propose a method to employ currently weakly annotated data into training. We evaluate our approach on the large-scale chest X-Ray datasets MIMIC-CXR, CheXpert, and ChestX-Ray14 for disease classification. We show that despite using unstructured medical report supervision, we perform on par with direct label supervision through a sophisticated inference setting.
doi_str_mv 10.48550/arxiv.2205.07139
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2205_07139</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2665374828</sourcerecordid><originalsourceid>FETCH-LOGICAL-a528-b694fbd3e40b01d9e69ab1157d362334783000c394132481dbac0a2aa233da73</originalsourceid><addsrcrecordid>eNotj01Lw0AURQdBsNT-AFcOuE6cz2Sy1GKrUFBsF-7CS2eaTK2ZOJnU9t87tq4e3He4nIvQDSWpUFKSe_AHu08ZIzIlOeXFBRoxzmmiBGNXaNL3W0IIy3ImJR-hj0dv4NO2Nf6xocEzezAaL03AbxAat3P1Eb-btatbG6xrcWi8G-omZp3zIZkPVkd-6trgoQ92b_DKg21j3zW63MCuN5P_O0bL2dNq-pwsXucv04dFApKppMoKsak0N4JUhOrCZAVUlMpc8yxai1zxaLvmhaCcCUV1BWsCDCA-NeR8jG7PrafVZeftF_hj-be-PK2PxN2Z6Lz7Hkwfyq0bfBuVSpZlkudCMcV_AVndXPU</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2665374828</pqid></control><display><type>article</type><title>Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Seibold, Constantin ; Reiß, Simon ; M Saquib Sarfraz ; Stiefelhagen, Rainer ; Kleesiek, Jens</creator><creatorcontrib>Seibold, Constantin ; Reiß, Simon ; M Saquib Sarfraz ; Stiefelhagen, Rainer ; Kleesiek, Jens</creatorcontrib><description>When reading images, radiologists generate text reports describing the findings therein. Current state-of-the-art computer-aided diagnosis tools utilize a fixed set of predefined categories automatically extracted from these medical reports for training. This form of supervision limits the potential usage of models as they are unable to pick up on anomalies outside of their predefined set, thus, making it a necessity to retrain the classifier with additional data when faced with novel classes. In contrast, we investigate direct text supervision to break away from this closed set assumption. By doing so, we avoid noisy label extraction via text classifiers and incorporate more contextual information. We employ a contrastive global-local dual-encoder architecture to learn concepts directly from unstructured medical reports while maintaining its ability to perform free form classification. We investigate relevant properties of open set recognition for radiological data and propose a method to employ currently weakly annotated data into training. We evaluate our approach on the large-scale chest X-Ray datasets MIMIC-CXR, CheXpert, and ChestX-Ray14 for disease classification. We show that despite using unstructured medical report supervision, we perform on par with direct label supervision through a sophisticated inference setting.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2205.07139</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Anomalies ; Classification ; Classifiers ; Coders ; Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Learning ; Free form ; Object recognition ; Supervision ; Training</subject><ispartof>arXiv.org, 2022-05</ispartof><rights>2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27925</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.2205.07139$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1007/978-3-031-16443-9_66$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Seibold, Constantin</creatorcontrib><creatorcontrib>Reiß, Simon</creatorcontrib><creatorcontrib>M Saquib Sarfraz</creatorcontrib><creatorcontrib>Stiefelhagen, Rainer</creatorcontrib><creatorcontrib>Kleesiek, Jens</creatorcontrib><title>Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training</title><title>arXiv.org</title><description>When reading images, radiologists generate text reports describing the findings therein. Current state-of-the-art computer-aided diagnosis tools utilize a fixed set of predefined categories automatically extracted from these medical reports for training. This form of supervision limits the potential usage of models as they are unable to pick up on anomalies outside of their predefined set, thus, making it a necessity to retrain the classifier with additional data when faced with novel classes. In contrast, we investigate direct text supervision to break away from this closed set assumption. By doing so, we avoid noisy label extraction via text classifiers and incorporate more contextual information. We employ a contrastive global-local dual-encoder architecture to learn concepts directly from unstructured medical reports while maintaining its ability to perform free form classification. We investigate relevant properties of open set recognition for radiological data and propose a method to employ currently weakly annotated data into training. We evaluate our approach on the large-scale chest X-Ray datasets MIMIC-CXR, CheXpert, and ChestX-Ray14 for disease classification. We show that despite using unstructured medical report supervision, we perform on par with direct label supervision through a sophisticated inference setting.</description><subject>Anomalies</subject><subject>Classification</subject><subject>Classifiers</subject><subject>Coders</subject><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Learning</subject><subject>Free form</subject><subject>Object recognition</subject><subject>Supervision</subject><subject>Training</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotj01Lw0AURQdBsNT-AFcOuE6cz2Sy1GKrUFBsF-7CS2eaTK2ZOJnU9t87tq4e3He4nIvQDSWpUFKSe_AHu08ZIzIlOeXFBRoxzmmiBGNXaNL3W0IIy3ImJR-hj0dv4NO2Nf6xocEzezAaL03AbxAat3P1Eb-btatbG6xrcWi8G-omZp3zIZkPVkd-6trgoQ92b_DKg21j3zW63MCuN5P_O0bL2dNq-pwsXucv04dFApKppMoKsak0N4JUhOrCZAVUlMpc8yxai1zxaLvmhaCcCUV1BWsCDCA-NeR8jG7PrafVZeftF_hj-be-PK2PxN2Z6Lz7Hkwfyq0bfBuVSpZlkudCMcV_AVndXPU</recordid><startdate>20220514</startdate><enddate>20220514</enddate><creator>Seibold, Constantin</creator><creator>Reiß, Simon</creator><creator>M Saquib Sarfraz</creator><creator>Stiefelhagen, Rainer</creator><creator>Kleesiek, Jens</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220514</creationdate><title>Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training</title><author>Seibold, Constantin ; Reiß, Simon ; M Saquib Sarfraz ; Stiefelhagen, Rainer ; Kleesiek, Jens</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a528-b694fbd3e40b01d9e69ab1157d362334783000c394132481dbac0a2aa233da73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Anomalies</topic><topic>Classification</topic><topic>Classifiers</topic><topic>Coders</topic><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Learning</topic><topic>Free form</topic><topic>Object recognition</topic><topic>Supervision</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Seibold, Constantin</creatorcontrib><creatorcontrib>Reiß, Simon</creatorcontrib><creatorcontrib>M Saquib Sarfraz</creatorcontrib><creatorcontrib>Stiefelhagen, Rainer</creatorcontrib><creatorcontrib>Kleesiek, Jens</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Access via ProQuest (Open Access)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Seibold, Constantin</au><au>Reiß, Simon</au><au>M Saquib Sarfraz</au><au>Stiefelhagen, Rainer</au><au>Kleesiek, Jens</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training</atitle><jtitle>arXiv.org</jtitle><date>2022-05-14</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>When reading images, radiologists generate text reports describing the findings therein. Current state-of-the-art computer-aided diagnosis tools utilize a fixed set of predefined categories automatically extracted from these medical reports for training. This form of supervision limits the potential usage of models as they are unable to pick up on anomalies outside of their predefined set, thus, making it a necessity to retrain the classifier with additional data when faced with novel classes. In contrast, we investigate direct text supervision to break away from this closed set assumption. By doing so, we avoid noisy label extraction via text classifiers and incorporate more contextual information. We employ a contrastive global-local dual-encoder architecture to learn concepts directly from unstructured medical reports while maintaining its ability to perform free form classification. We investigate relevant properties of open set recognition for radiological data and propose a method to employ currently weakly annotated data into training. We evaluate our approach on the large-scale chest X-Ray datasets MIMIC-CXR, CheXpert, and ChestX-Ray14 for disease classification. We show that despite using unstructured medical report supervision, we perform on par with direct label supervision through a sophisticated inference setting.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2205.07139</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2022-05
issn 2331-8422
language eng
recordid cdi_arxiv_primary_2205_07139
source arXiv.org; Free E- Journals
subjects Anomalies
Classification
Classifiers
Coders
Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
Free form
Object recognition
Supervision
Training
title Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T22%3A19%3A26IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Breaking%20with%20Fixed%20Set%20Pathology%20Recognition%20through%20Report-Guided%20Contrastive%20Training&rft.jtitle=arXiv.org&rft.au=Seibold,%20Constantin&rft.date=2022-05-14&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2205.07139&rft_dat=%3Cproquest_arxiv%3E2665374828%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2665374828&rft_id=info:pmid/&rfr_iscdi=true