Noisy Image Segmentation With Soft-Dice

Noisy Image Segmentation With Soft-Dice

This paper presents a study on the soft-Dice loss, one of the most popular loss functions in medical image segmentation, for situations where noise is present in target labels. In particular, the set of optimal solutions are characterized and sharp bounds on the volume bias of these solutions are pr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-05
Hauptverfasser: Nordström, Marcus, Hult, Henrik, Maki, Atsuto, Löfman, Fredrik
Format: Artikel
Sprache:eng
Schlagworte:
Convergence
Image segmentation
Medical imaging
Optimization
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 Nordström, Marcus
Hult, Henrik
Maki, Atsuto
Löfman, Fredrik
description This paper presents a study on the soft-Dice loss, one of the most popular loss functions in medical image segmentation, for situations where noise is present in target labels. In particular, the set of optimal solutions are characterized and sharp bounds on the volume bias of these solutions are provided. It is further shown that a sequence of soft segmentations converging to optimal soft-Dice also converges to optimal Dice when converted to hard segmentations using thresholding. This is an important result because soft-Dice is often used as a proxy for maximizing the Dice metric. Finally, experiments confirming the theoretical results are provided.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2795863174</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2795863174</sourcerecordid><originalsourceid>FETCH-proquest_journals_27958631743</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mRQ98vPLK5U8MxNTE9VCE5Nz03NK0ksyczPUwjPLMlQCM5PK9F1yUxO5WFgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCNzS1MLM2NDcxNj4lQBADxaL1E</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2795863174</pqid></control><display><type>article</type><title>Noisy Image Segmentation With Soft-Dice</title><source>Free E- Journals</source><creator>Nordström, Marcus ; Hult, Henrik ; Maki, Atsuto ; Löfman, Fredrik</creator><creatorcontrib>Nordström, Marcus ; Hult, Henrik ; Maki, Atsuto ; Löfman, Fredrik</creatorcontrib><description>This paper presents a study on the soft-Dice loss, one of the most popular loss functions in medical image segmentation, for situations where noise is present in target labels. In particular, the set of optimal solutions are characterized and sharp bounds on the volume bias of these solutions are provided. It is further shown that a sequence of soft segmentations converging to optimal soft-Dice also converges to optimal Dice when converted to hard segmentations using thresholding. This is an important result because soft-Dice is often used as a proxy for maximizing the Dice metric. Finally, experiments confirming the theoretical results are provided.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Convergence ; Image segmentation ; Medical imaging ; Optimization</subject><ispartof>arXiv.org, 2023-05</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</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>776,780</link.rule.ids></links><search><creatorcontrib>Nordström, Marcus</creatorcontrib><creatorcontrib>Hult, Henrik</creatorcontrib><creatorcontrib>Maki, Atsuto</creatorcontrib><creatorcontrib>Löfman, Fredrik</creatorcontrib><title>Noisy Image Segmentation With Soft-Dice</title><title>arXiv.org</title><description>This paper presents a study on the soft-Dice loss, one of the most popular loss functions in medical image segmentation, for situations where noise is present in target labels. In particular, the set of optimal solutions are characterized and sharp bounds on the volume bias of these solutions are provided. It is further shown that a sequence of soft segmentations converging to optimal soft-Dice also converges to optimal Dice when converted to hard segmentations using thresholding. This is an important result because soft-Dice is often used as a proxy for maximizing the Dice metric. Finally, experiments confirming the theoretical results are provided.</description><subject>Convergence</subject><subject>Image segmentation</subject><subject>Medical imaging</subject><subject>Optimization</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mRQ98vPLK5U8MxNTE9VCE5Nz03NK0ksyczPUwjPLMlQCM5PK9F1yUxO5WFgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCNzS1MLM2NDcxNj4lQBADxaL1E</recordid><startdate>20230504</startdate><enddate>20230504</enddate><creator>Nordström, Marcus</creator><creator>Hult, Henrik</creator><creator>Maki, Atsuto</creator><creator>Löfman, Fredrik</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></search><sort><creationdate>20230504</creationdate><title>Noisy Image Segmentation With Soft-Dice</title><author>Nordström, Marcus ; Hult, Henrik ; Maki, Atsuto ; Löfman, Fredrik</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27958631743</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Convergence</topic><topic>Image segmentation</topic><topic>Medical imaging</topic><topic>Optimization</topic><toplevel>online_resources</toplevel><creatorcontrib>Nordström, Marcus</creatorcontrib><creatorcontrib>Hult, Henrik</creatorcontrib><creatorcontrib>Maki, Atsuto</creatorcontrib><creatorcontrib>Löfman, Fredrik</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>Publicly Available Content Database</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></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Nordström, Marcus</au><au>Hult, Henrik</au><au>Maki, Atsuto</au><au>Löfman, Fredrik</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Noisy Image Segmentation With Soft-Dice</atitle><jtitle>arXiv.org</jtitle><date>2023-05-04</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>This paper presents a study on the soft-Dice loss, one of the most popular loss functions in medical image segmentation, for situations where noise is present in target labels. In particular, the set of optimal solutions are characterized and sharp bounds on the volume bias of these solutions are provided. It is further shown that a sequence of soft segmentations converging to optimal soft-Dice also converges to optimal Dice when converted to hard segmentations using thresholding. This is an important result because soft-Dice is often used as a proxy for maximizing the Dice metric. Finally, experiments confirming the theoretical results are provided.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-05
issn 2331-8422
language eng
recordid cdi_proquest_journals_2795863174
source Free E- Journals
subjects Convergence
Image segmentation
Medical imaging
Optimization
title Noisy Image Segmentation With Soft-Dice
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-15T05%3A10%3A56IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Noisy%20Image%20Segmentation%20With%20Soft-Dice&rft.jtitle=arXiv.org&rft.au=Nordstr%C3%B6m,%20Marcus&rft.date=2023-05-04&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2795863174%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2795863174&rft_id=info:pmid/&rfr_iscdi=true