DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images

Neovascular age-related macular degeneration (nAMD) and polypoidal choroidal vasculopathy (PCV) present many similar clinical features. However, there are significant differences in the progression of nAMD and PCV. and it is crucial to make accurate diagnosis for treatment. In this paper, we propose...

Full description

Saved in:
Bibliographic Details
Published in:Physics in medicine & biology 2024-04, Vol.69 (7), p.75012
Main Authors: Shen, Erwei, Wang, Zhenmao, Lin, Tian, Meng, Qingquan, Zhu, Weifang, Shi, Fei, Chen, Xinjian, Chen, Haoyu, Xiang, Dehui
Format: Article
Language:English
Subjects:
Choroid - blood supply
Fluorescein Angiography - methods
Humans
Image classification
OCT
Polypoidal Choroidal Vasculopathy
Radiomics
Retrospective Studies
Tomography, Optical Coherence - methods
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites cdi_FETCH-LOGICAL-c321t-c260d90c0b9416afae92b0b07605130553c8229f4f719380dc66d27921a15cb63
container_end_page
container_issue 7
container_start_page 75012
container_title Physics in medicine & biology
container_volume 69
creator Shen, Erwei
Wang, Zhenmao
Lin, Tian
Meng, Qingquan
Zhu, Weifang
Shi, Fei
Chen, Xinjian
Chen, Haoyu
Xiang, Dehui
description Neovascular age-related macular degeneration (nAMD) and polypoidal choroidal vasculopathy (PCV) present many similar clinical features. However, there are significant differences in the progression of nAMD and PCV. and it is crucial to make accurate diagnosis for treatment. In this paper, we propose a structure-radiomic fusion network (DRFNet) to differentiate PCV and nAMD in optical coherence tomography (OCT) images. The subnetwork (RIMNet) is designed to automatically segment the lesion of nAMD and PCV. Another subnetwork (StrEncoder) is designed to extract deep structural features of the segmented lesion. The subnetwork (RadEncoder) is designed to extract radiomic features from the segmented lesions based on radiomics. 305 eyes (155 with nAMD and 150 with PCV) are included and manually annotated CNV region in this study. The proposed method was trained and evaluated by 4-fold cross validation using the collected data and was compared with the advanced differentiation methods. The proposed method achieved high classification performace of nAMD/PCV differentiation in OCT images, which was an improvement of 4.68 compared with other best method. . The presented structure-radiomic fusion network (DRFNet) has great performance of diagnosing nAMD and PCV and high clinical value by using OCT instead of indocyanine green angiography.
doi_str_mv 10.1088/1361-6560/ad2ca0
format article
fullrecord <record><control><sourceid>proquest_iop_j</sourceid><recordid>TN_cdi_iop_journals_10_1088_1361_6560_ad2ca0</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2973104357</sourcerecordid><originalsourceid>FETCH-LOGICAL-c321t-c260d90c0b9416afae92b0b07605130553c8229f4f719380dc66d27921a15cb63</originalsourceid><addsrcrecordid>eNp9kE1v1DAQhq2Kql1K75yQb3Ag3Rk7dmJu1Za2SIVCVbhajj-QyyYOdiLEvyerLT0hTiONnvfVzEPIS4QzhLZdI5dYSSFhbRyzBg7I6mn1jKwAOFYKhTgmz0t5AEBsWX1EjnnLVS0buSJfLu4uP_npHTXUeT_SbFxMfbQ0zCWmgQ5--pXyDxpSpsP5x4v158036mIIPvthimbaQXGgt5t7Gnvz3ZcX5DCYbfGnj_OEfL18f7-5rm5urz5szm8qyxlOlWUSnAILnapRmmC8Yh100EgQyEEIblvGVKhDg4q34KyUjjWKoUFhO8lPyJt975jTz9mXSfexWL_dmsGnuWimGo5Qc9EsKOxRm1Mp2Qc95uXY_Fsj6J1IvbOmd9b0XuQSefXYPne9d0-Bv-YW4O0eiGnUD2nOw_Ls__pe_wMf-05LpRsNjQBkenSB_wFx44ci</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2973104357</pqid></control><display><type>article</type><title>DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images</title><source>Institute of Physics:Jisc Collections:IOP Publishing Read and Publish 2024-2025 (Reading List)</source><creator>Shen, Erwei ; Wang, Zhenmao ; Lin, Tian ; Meng, Qingquan ; Zhu, Weifang ; Shi, Fei ; Chen, Xinjian ; Chen, Haoyu ; Xiang, Dehui</creator><creatorcontrib>Shen, Erwei ; Wang, Zhenmao ; Lin, Tian ; Meng, Qingquan ; Zhu, Weifang ; Shi, Fei ; Chen, Xinjian ; Chen, Haoyu ; Xiang, Dehui</creatorcontrib><description>Neovascular age-related macular degeneration (nAMD) and polypoidal choroidal vasculopathy (PCV) present many similar clinical features. However, there are significant differences in the progression of nAMD and PCV. and it is crucial to make accurate diagnosis for treatment. In this paper, we propose a structure-radiomic fusion network (DRFNet) to differentiate PCV and nAMD in optical coherence tomography (OCT) images. The subnetwork (RIMNet) is designed to automatically segment the lesion of nAMD and PCV. Another subnetwork (StrEncoder) is designed to extract deep structural features of the segmented lesion. The subnetwork (RadEncoder) is designed to extract radiomic features from the segmented lesions based on radiomics. 305 eyes (155 with nAMD and 150 with PCV) are included and manually annotated CNV region in this study. The proposed method was trained and evaluated by 4-fold cross validation using the collected data and was compared with the advanced differentiation methods. The proposed method achieved high classification performace of nAMD/PCV differentiation in OCT images, which was an improvement of 4.68 compared with other best method. . The presented structure-radiomic fusion network (DRFNet) has great performance of diagnosing nAMD and PCV and high clinical value by using OCT instead of indocyanine green angiography.</description><identifier>ISSN: 0031-9155</identifier><identifier>EISSN: 1361-6560</identifier><identifier>DOI: 10.1088/1361-6560/ad2ca0</identifier><identifier>PMID: 38394676</identifier><identifier>CODEN: PHMBA7</identifier><language>eng</language><publisher>England: IOP Publishing</publisher><subject>Choroid - blood supply ; Fluorescein Angiography - methods ; Humans ; Image classification ; OCT ; Polypoidal Choroidal Vasculopathy ; Radiomics ; Retrospective Studies ; Tomography, Optical Coherence - methods</subject><ispartof>Physics in medicine &amp; biology, 2024-04, Vol.69 (7), p.75012</ispartof><rights>2024 Institute of Physics and Engineering in Medicine</rights><rights>2024 Institute of Physics and Engineering in Medicine.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c321t-c260d90c0b9416afae92b0b07605130553c8229f4f719380dc66d27921a15cb63</cites><orcidid>0000-0001-7873-9778 ; 0000-0003-0676-4610</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>315,786,790,27957,27958</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/38394676$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Shen, Erwei</creatorcontrib><creatorcontrib>Wang, Zhenmao</creatorcontrib><creatorcontrib>Lin, Tian</creatorcontrib><creatorcontrib>Meng, Qingquan</creatorcontrib><creatorcontrib>Zhu, Weifang</creatorcontrib><creatorcontrib>Shi, Fei</creatorcontrib><creatorcontrib>Chen, Xinjian</creatorcontrib><creatorcontrib>Chen, Haoyu</creatorcontrib><creatorcontrib>Xiang, Dehui</creatorcontrib><title>DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images</title><title>Physics in medicine &amp; biology</title><addtitle>PMB</addtitle><addtitle>Phys. Med. Biol</addtitle><description>Neovascular age-related macular degeneration (nAMD) and polypoidal choroidal vasculopathy (PCV) present many similar clinical features. However, there are significant differences in the progression of nAMD and PCV. and it is crucial to make accurate diagnosis for treatment. In this paper, we propose a structure-radiomic fusion network (DRFNet) to differentiate PCV and nAMD in optical coherence tomography (OCT) images. The subnetwork (RIMNet) is designed to automatically segment the lesion of nAMD and PCV. Another subnetwork (StrEncoder) is designed to extract deep structural features of the segmented lesion. The subnetwork (RadEncoder) is designed to extract radiomic features from the segmented lesions based on radiomics. 305 eyes (155 with nAMD and 150 with PCV) are included and manually annotated CNV region in this study. The proposed method was trained and evaluated by 4-fold cross validation using the collected data and was compared with the advanced differentiation methods. The proposed method achieved high classification performace of nAMD/PCV differentiation in OCT images, which was an improvement of 4.68 compared with other best method. . The presented structure-radiomic fusion network (DRFNet) has great performance of diagnosing nAMD and PCV and high clinical value by using OCT instead of indocyanine green angiography.</description><subject>Choroid - blood supply</subject><subject>Fluorescein Angiography - methods</subject><subject>Humans</subject><subject>Image classification</subject><subject>OCT</subject><subject>Polypoidal Choroidal Vasculopathy</subject><subject>Radiomics</subject><subject>Retrospective Studies</subject><subject>Tomography, Optical Coherence - methods</subject><issn>0031-9155</issn><issn>1361-6560</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp9kE1v1DAQhq2Kql1K75yQb3Ag3Rk7dmJu1Za2SIVCVbhajj-QyyYOdiLEvyerLT0hTiONnvfVzEPIS4QzhLZdI5dYSSFhbRyzBg7I6mn1jKwAOFYKhTgmz0t5AEBsWX1EjnnLVS0buSJfLu4uP_npHTXUeT_SbFxMfbQ0zCWmgQ5--pXyDxpSpsP5x4v158036mIIPvthimbaQXGgt5t7Gnvz3ZcX5DCYbfGnj_OEfL18f7-5rm5urz5szm8qyxlOlWUSnAILnapRmmC8Yh100EgQyEEIblvGVKhDg4q34KyUjjWKoUFhO8lPyJt975jTz9mXSfexWL_dmsGnuWimGo5Qc9EsKOxRm1Mp2Qc95uXY_Fsj6J1IvbOmd9b0XuQSefXYPne9d0-Bv-YW4O0eiGnUD2nOw_Ls__pe_wMf-05LpRsNjQBkenSB_wFx44ci</recordid><startdate>20240407</startdate><enddate>20240407</enddate><creator>Shen, Erwei</creator><creator>Wang, Zhenmao</creator><creator>Lin, Tian</creator><creator>Meng, Qingquan</creator><creator>Zhu, Weifang</creator><creator>Shi, Fei</creator><creator>Chen, Xinjian</creator><creator>Chen, Haoyu</creator><creator>Xiang, Dehui</creator><general>IOP Publishing</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0001-7873-9778</orcidid><orcidid>https://orcid.org/0000-0003-0676-4610</orcidid></search><sort><creationdate>20240407</creationdate><title>DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images</title><author>Shen, Erwei ; Wang, Zhenmao ; Lin, Tian ; Meng, Qingquan ; Zhu, Weifang ; Shi, Fei ; Chen, Xinjian ; Chen, Haoyu ; Xiang, Dehui</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c321t-c260d90c0b9416afae92b0b07605130553c8229f4f719380dc66d27921a15cb63</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Choroid - blood supply</topic><topic>Fluorescein Angiography - methods</topic><topic>Humans</topic><topic>Image classification</topic><topic>OCT</topic><topic>Polypoidal Choroidal Vasculopathy</topic><topic>Radiomics</topic><topic>Retrospective Studies</topic><topic>Tomography, Optical Coherence - methods</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shen, Erwei</creatorcontrib><creatorcontrib>Wang, Zhenmao</creatorcontrib><creatorcontrib>Lin, Tian</creatorcontrib><creatorcontrib>Meng, Qingquan</creatorcontrib><creatorcontrib>Zhu, Weifang</creatorcontrib><creatorcontrib>Shi, Fei</creatorcontrib><creatorcontrib>Chen, Xinjian</creatorcontrib><creatorcontrib>Chen, Haoyu</creatorcontrib><creatorcontrib>Xiang, Dehui</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Physics in medicine &amp; biology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shen, Erwei</au><au>Wang, Zhenmao</au><au>Lin, Tian</au><au>Meng, Qingquan</au><au>Zhu, Weifang</au><au>Shi, Fei</au><au>Chen, Xinjian</au><au>Chen, Haoyu</au><au>Xiang, Dehui</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images</atitle><jtitle>Physics in medicine &amp; biology</jtitle><stitle>PMB</stitle><addtitle>Phys. Med. Biol</addtitle><date>2024-04-07</date><risdate>2024</risdate><volume>69</volume><issue>7</issue><spage>75012</spage><pages>75012-</pages><issn>0031-9155</issn><eissn>1361-6560</eissn><coden>PHMBA7</coden><notes>PMB-116015.R2</notes><notes>ObjectType-Article-1</notes><notes>SourceType-Scholarly Journals-1</notes><notes>ObjectType-Feature-2</notes><notes>content type line 23</notes><abstract>Neovascular age-related macular degeneration (nAMD) and polypoidal choroidal vasculopathy (PCV) present many similar clinical features. However, there are significant differences in the progression of nAMD and PCV. and it is crucial to make accurate diagnosis for treatment. In this paper, we propose a structure-radiomic fusion network (DRFNet) to differentiate PCV and nAMD in optical coherence tomography (OCT) images. The subnetwork (RIMNet) is designed to automatically segment the lesion of nAMD and PCV. Another subnetwork (StrEncoder) is designed to extract deep structural features of the segmented lesion. The subnetwork (RadEncoder) is designed to extract radiomic features from the segmented lesions based on radiomics. 305 eyes (155 with nAMD and 150 with PCV) are included and manually annotated CNV region in this study. The proposed method was trained and evaluated by 4-fold cross validation using the collected data and was compared with the advanced differentiation methods. The proposed method achieved high classification performace of nAMD/PCV differentiation in OCT images, which was an improvement of 4.68 compared with other best method. . The presented structure-radiomic fusion network (DRFNet) has great performance of diagnosing nAMD and PCV and high clinical value by using OCT instead of indocyanine green angiography.</abstract><cop>England</cop><pub>IOP Publishing</pub><pmid>38394676</pmid><doi>10.1088/1361-6560/ad2ca0</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0001-7873-9778</orcidid><orcidid>https://orcid.org/0000-0003-0676-4610</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 0031-9155
ispartof Physics in medicine & biology, 2024-04, Vol.69 (7), p.75012
issn 0031-9155
1361-6560
language eng
recordid cdi_iop_journals_10_1088_1361_6560_ad2ca0
source Institute of Physics:Jisc Collections:IOP Publishing Read and Publish 2024-2025 (Reading List)
subjects Choroid - blood supply
Fluorescein Angiography - methods
Humans
Image classification
OCT
Polypoidal Choroidal Vasculopathy
Radiomics
Retrospective Studies
Tomography, Optical Coherence - methods
title DRFNet: a deep radiomic fusion network for nAMD/PCV differentiation in OCT images
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-22T14%3A35%3A52IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_iop_j&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=DRFNet:%20a%20deep%20radiomic%20fusion%20network%20for%20nAMD/PCV%20differentiation%20in%20OCT%20images&rft.jtitle=Physics%20in%20medicine%20&%20biology&rft.au=Shen,%20Erwei&rft.date=2024-04-07&rft.volume=69&rft.issue=7&rft.spage=75012&rft.pages=75012-&rft.issn=0031-9155&rft.eissn=1361-6560&rft.coden=PHMBA7&rft_id=info:doi/10.1088/1361-6560/ad2ca0&rft_dat=%3Cproquest_iop_j%3E2973104357%3C/proquest_iop_j%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c321t-c260d90c0b9416afae92b0b07605130553c8229f4f719380dc66d27921a15cb63%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2973104357&rft_id=info:pmid/38394676&rfr_iscdi=true