Loading…

CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification

Taxonomy illustrates that natural creatures can be classified with a hierarchy. The connections between species are explicit and objective and can be organized into a knowledge graph (KG). It is a challenging task to mine features of known categories from KG and to reason on unknown categories. Grap...

Full description

Saved in:
Bibliographic Details
Published in:Sensors (Basel, Switzerland) Switzerland), 2022-12, Vol.22 (24), p.9980
Main Authors: Li, Liangwei, Liu, Lin, Du, Xiaohui, Wang, Xiangzhou, Zhang, Ziruo, Zhang, Jing, Zhang, Ping, Liu, Juanxiu
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13
cites cdi_FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13
container_end_page
container_issue 24
container_start_page 9980
container_title Sensors (Basel, Switzerland)
container_volume 22
creator Li, Liangwei
Liu, Lin
Du, Xiaohui
Wang, Xiangzhou
Zhang, Ziruo
Zhang, Jing
Zhang, Ping
Liu, Juanxiu
description Taxonomy illustrates that natural creatures can be classified with a hierarchy. The connections between species are explicit and objective and can be organized into a knowledge graph (KG). It is a challenging task to mine features of known categories from KG and to reason on unknown categories. Graph Convolutional Network (GCN) has recently been viewed as a potential approach to zero-shot learning. GCN enables knowledge transfer by sharing the statistical strength of nodes in the graph. More layers of graph convolution are stacked in order to aggregate the hierarchical information in the KG. However, the Laplacian over-smoothing problem will be severe as the number of GCN layers deepens, which leads the features between nodes toward a tendency to be similar and degrade the performance of zero-shot image classification tasks. We consider two parts to mitigate the Laplacian over-smoothing problem, namely reducing the invalid node aggregation and improving the discriminability among nodes in the deep graph network. We propose a top-k graph pooling method based on the self-attention mechanism to control specific node aggregation, and we introduce a dual structural symmetric knowledge graph additionally to enhance the representation of nodes in the latent space. Finally, we apply these new concepts to the recently widely used contrastive learning framework and propose a novel Contrastive Graph U-Net with two Attention-based graph pooling (Att-gPool) layers, CGUN-2A, which explicitly alleviates the Laplacian over-smoothing problem. To evaluate the performance of the method on complex real-world scenes, we test it on the large-scale zero-shot image classification dataset. Extensive experiments show the positive effect of allowing nodes to perform specific aggregation, as well as homogeneous graph comparison, in our deep graph network. We show how it significantly boosts zero-shot image classification performance. The Hit@1 accuracy is 17.5% relatively higher than the baseline model on the ImageNet21K dataset.
doi_str_mv 10.3390/s22249980
format article
fullrecord <record><control><sourceid>gale_doaj_</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_0f2483ad6e1947c884df8a2d4f9ca452</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><galeid>A746535441</galeid><doaj_id>oai_doaj_org_article_0f2483ad6e1947c884df8a2d4f9ca452</doaj_id><sourcerecordid>A746535441</sourcerecordid><originalsourceid>FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13</originalsourceid><addsrcrecordid>eNpdkk1vEzEQQFcIREvhwB9AlrjAYYvXH7teDkhRgBApKofSCxdrYo83Dpt1am-C-Pc4TYla5IOtmednezxF8bqil5y39ENijIm2VfRJcV4JJkrFGH36YH1WvEhpTSnjnKvnxRmvZU25rM6L2-ns5qpkk4_kM-KWzCJsV2Qahn3od6MPA_TkCsffIf4iew-HzBghjX6PZIEQBz90xIVIFhA7LK8N9Eh-Ygzl9SqMZL6BDsm0h5S88wYOxpfFMwd9wlf380Vx8_XLj-m3cvF9Np9OFqWRVI0lb4UQlbG1bDgFh4YxBVwslxwoWkapsXaJkkmDlHPXOmpa1lLWYMOohYpfFPOj1wZY6230G4h_dACv7wIhdhri6E2PmjomFAdbY9WKxiglrFPArHCtASFZdn06ura75QatwUMV-kfSx5nBr3QX9rptFJOVyoJ394IYbneYRr3xyWDfw4BhlzRrpKpozbjM6Nv_0HXYxfwRd1SdhVzUmbo8Ul2uuPaDC_lck4fFjTdhQOdzfNKIWnKZC5k3vD9uMDGkFNGdbl9RfegifeqizL55-NwT-a9t-F8N28Dw</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2756782346</pqid></control><display><type>article</type><title>CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification</title><source>Open Access: PubMed Central</source><source>Publicly Available Content Database</source><source>Coronavirus Research Database</source><creator>Li, Liangwei ; Liu, Lin ; Du, Xiaohui ; Wang, Xiangzhou ; Zhang, Ziruo ; Zhang, Jing ; Zhang, Ping ; Liu, Juanxiu</creator><creatorcontrib>Li, Liangwei ; Liu, Lin ; Du, Xiaohui ; Wang, Xiangzhou ; Zhang, Ziruo ; Zhang, Jing ; Zhang, Ping ; Liu, Juanxiu</creatorcontrib><description>Taxonomy illustrates that natural creatures can be classified with a hierarchy. The connections between species are explicit and objective and can be organized into a knowledge graph (KG). It is a challenging task to mine features of known categories from KG and to reason on unknown categories. Graph Convolutional Network (GCN) has recently been viewed as a potential approach to zero-shot learning. GCN enables knowledge transfer by sharing the statistical strength of nodes in the graph. More layers of graph convolution are stacked in order to aggregate the hierarchical information in the KG. However, the Laplacian over-smoothing problem will be severe as the number of GCN layers deepens, which leads the features between nodes toward a tendency to be similar and degrade the performance of zero-shot image classification tasks. We consider two parts to mitigate the Laplacian over-smoothing problem, namely reducing the invalid node aggregation and improving the discriminability among nodes in the deep graph network. We propose a top-k graph pooling method based on the self-attention mechanism to control specific node aggregation, and we introduce a dual structural symmetric knowledge graph additionally to enhance the representation of nodes in the latent space. Finally, we apply these new concepts to the recently widely used contrastive learning framework and propose a novel Contrastive Graph U-Net with two Attention-based graph pooling (Att-gPool) layers, CGUN-2A, which explicitly alleviates the Laplacian over-smoothing problem. To evaluate the performance of the method on complex real-world scenes, we test it on the large-scale zero-shot image classification dataset. Extensive experiments show the positive effect of allowing nodes to perform specific aggregation, as well as homogeneous graph comparison, in our deep graph network. We show how it significantly boosts zero-shot image classification performance. The Hit@1 accuracy is 17.5% relatively higher than the baseline model on the ImageNet21K dataset.</description><identifier>ISSN: 1424-8220</identifier><identifier>EISSN: 1424-8220</identifier><identifier>DOI: 10.3390/s22249980</identifier><identifier>PMID: 36560351</identifier><language>eng</language><publisher>Switzerland: MDPI AG</publisher><subject>Algorithms ; Classification ; contrastive learning ; Datasets ; Dietary Supplements ; Ecology ; Environmental monitoring ; Graph comparison ; graph convolutional network ; Graph representations ; Image classification ; Knowledge ; Knowledge management ; Learning ; machine learning ; Neural networks ; Nodes ; Plankton ; Records ; Semantics ; Smoothing ; Taxonomy ; Teaching methods ; zero-shot learning</subject><ispartof>Sensors (Basel, Switzerland), 2022-12, Vol.22 (24), p.9980</ispartof><rights>COPYRIGHT 2022 MDPI AG</rights><rights>2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>2022 by the authors. 2022</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13</citedby><cites>FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13</cites><orcidid>0000-0002-5965-109X ; 0000-0001-6052-4496</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.proquest.com/docview/2756782346?pq-origsite=primo$$EPDF$$P50$$Gproquest$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.proquest.com/docview/2756782346?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>230,314,727,780,784,885,25753,27924,27925,37012,37013,38516,43895,44590,53791,53793,74412,75126</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36560351$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Li, Liangwei</creatorcontrib><creatorcontrib>Liu, Lin</creatorcontrib><creatorcontrib>Du, Xiaohui</creatorcontrib><creatorcontrib>Wang, Xiangzhou</creatorcontrib><creatorcontrib>Zhang, Ziruo</creatorcontrib><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Zhang, Ping</creatorcontrib><creatorcontrib>Liu, Juanxiu</creatorcontrib><title>CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification</title><title>Sensors (Basel, Switzerland)</title><addtitle>Sensors (Basel)</addtitle><description>Taxonomy illustrates that natural creatures can be classified with a hierarchy. The connections between species are explicit and objective and can be organized into a knowledge graph (KG). It is a challenging task to mine features of known categories from KG and to reason on unknown categories. Graph Convolutional Network (GCN) has recently been viewed as a potential approach to zero-shot learning. GCN enables knowledge transfer by sharing the statistical strength of nodes in the graph. More layers of graph convolution are stacked in order to aggregate the hierarchical information in the KG. However, the Laplacian over-smoothing problem will be severe as the number of GCN layers deepens, which leads the features between nodes toward a tendency to be similar and degrade the performance of zero-shot image classification tasks. We consider two parts to mitigate the Laplacian over-smoothing problem, namely reducing the invalid node aggregation and improving the discriminability among nodes in the deep graph network. We propose a top-k graph pooling method based on the self-attention mechanism to control specific node aggregation, and we introduce a dual structural symmetric knowledge graph additionally to enhance the representation of nodes in the latent space. Finally, we apply these new concepts to the recently widely used contrastive learning framework and propose a novel Contrastive Graph U-Net with two Attention-based graph pooling (Att-gPool) layers, CGUN-2A, which explicitly alleviates the Laplacian over-smoothing problem. To evaluate the performance of the method on complex real-world scenes, we test it on the large-scale zero-shot image classification dataset. Extensive experiments show the positive effect of allowing nodes to perform specific aggregation, as well as homogeneous graph comparison, in our deep graph network. We show how it significantly boosts zero-shot image classification performance. The Hit@1 accuracy is 17.5% relatively higher than the baseline model on the ImageNet21K dataset.</description><subject>Algorithms</subject><subject>Classification</subject><subject>contrastive learning</subject><subject>Datasets</subject><subject>Dietary Supplements</subject><subject>Ecology</subject><subject>Environmental monitoring</subject><subject>Graph comparison</subject><subject>graph convolutional network</subject><subject>Graph representations</subject><subject>Image classification</subject><subject>Knowledge</subject><subject>Knowledge management</subject><subject>Learning</subject><subject>machine learning</subject><subject>Neural networks</subject><subject>Nodes</subject><subject>Plankton</subject><subject>Records</subject><subject>Semantics</subject><subject>Smoothing</subject><subject>Taxonomy</subject><subject>Teaching methods</subject><subject>zero-shot learning</subject><issn>1424-8220</issn><issn>1424-8220</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>COVID</sourceid><sourceid>PIMPY</sourceid><sourceid>DOA</sourceid><recordid>eNpdkk1vEzEQQFcIREvhwB9AlrjAYYvXH7teDkhRgBApKofSCxdrYo83Dpt1am-C-Pc4TYla5IOtmednezxF8bqil5y39ENijIm2VfRJcV4JJkrFGH36YH1WvEhpTSnjnKvnxRmvZU25rM6L2-ns5qpkk4_kM-KWzCJsV2Qahn3od6MPA_TkCsffIf4iew-HzBghjX6PZIEQBz90xIVIFhA7LK8N9Eh-Ygzl9SqMZL6BDsm0h5S88wYOxpfFMwd9wlf380Vx8_XLj-m3cvF9Np9OFqWRVI0lb4UQlbG1bDgFh4YxBVwslxwoWkapsXaJkkmDlHPXOmpa1lLWYMOohYpfFPOj1wZY6230G4h_dACv7wIhdhri6E2PmjomFAdbY9WKxiglrFPArHCtASFZdn06ura75QatwUMV-kfSx5nBr3QX9rptFJOVyoJ394IYbneYRr3xyWDfw4BhlzRrpKpozbjM6Nv_0HXYxfwRd1SdhVzUmbo8Ul2uuPaDC_lck4fFjTdhQOdzfNKIWnKZC5k3vD9uMDGkFNGdbl9RfegifeqizL55-NwT-a9t-F8N28Dw</recordid><startdate>20221218</startdate><enddate>20221218</enddate><creator>Li, Liangwei</creator><creator>Liu, Lin</creator><creator>Du, Xiaohui</creator><creator>Wang, Xiangzhou</creator><creator>Zhang, Ziruo</creator><creator>Zhang, Jing</creator><creator>Zhang, Ping</creator><creator>Liu, Juanxiu</creator><general>MDPI AG</general><general>MDPI</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>3V.</scope><scope>7X7</scope><scope>7XB</scope><scope>88E</scope><scope>8FI</scope><scope>8FJ</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>CCPQU</scope><scope>COVID</scope><scope>DWQXO</scope><scope>FYUFA</scope><scope>GHDGH</scope><scope>K9.</scope><scope>M0S</scope><scope>M1P</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>7X8</scope><scope>5PM</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-5965-109X</orcidid><orcidid>https://orcid.org/0000-0001-6052-4496</orcidid></search><sort><creationdate>20221218</creationdate><title>CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification</title><author>Li, Liangwei ; Liu, Lin ; Du, Xiaohui ; Wang, Xiangzhou ; Zhang, Ziruo ; Zhang, Jing ; Zhang, Ping ; Liu, Juanxiu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Algorithms</topic><topic>Classification</topic><topic>contrastive learning</topic><topic>Datasets</topic><topic>Dietary Supplements</topic><topic>Ecology</topic><topic>Environmental monitoring</topic><topic>Graph comparison</topic><topic>graph convolutional network</topic><topic>Graph representations</topic><topic>Image classification</topic><topic>Knowledge</topic><topic>Knowledge management</topic><topic>Learning</topic><topic>machine learning</topic><topic>Neural networks</topic><topic>Nodes</topic><topic>Plankton</topic><topic>Records</topic><topic>Semantics</topic><topic>Smoothing</topic><topic>Taxonomy</topic><topic>Teaching methods</topic><topic>zero-shot learning</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Li, Liangwei</creatorcontrib><creatorcontrib>Liu, Lin</creatorcontrib><creatorcontrib>Du, Xiaohui</creatorcontrib><creatorcontrib>Wang, Xiangzhou</creatorcontrib><creatorcontrib>Zhang, Ziruo</creatorcontrib><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Zhang, Ping</creatorcontrib><creatorcontrib>Liu, Juanxiu</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Health &amp; Medical Collection (Proquest)</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Medical Database (Alumni Edition)</collection><collection>Hospital Premium Collection</collection><collection>Hospital Premium Collection (Alumni Edition)</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>ProQuest One Community College</collection><collection>Coronavirus Research Database</collection><collection>ProQuest Central</collection><collection>Health Research Premium Collection</collection><collection>Health Research Premium Collection (Alumni)</collection><collection>ProQuest Health &amp; Medical Complete (Alumni)</collection><collection>Health &amp; Medical Collection (Alumni Edition)</collection><collection>PML(ProQuest Medical Library)</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>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><collection>Directory of Open Access Journals</collection><jtitle>Sensors (Basel, Switzerland)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Li, Liangwei</au><au>Liu, Lin</au><au>Du, Xiaohui</au><au>Wang, Xiangzhou</au><au>Zhang, Ziruo</au><au>Zhang, Jing</au><au>Zhang, Ping</au><au>Liu, Juanxiu</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification</atitle><jtitle>Sensors (Basel, Switzerland)</jtitle><addtitle>Sensors (Basel)</addtitle><date>2022-12-18</date><risdate>2022</risdate><volume>22</volume><issue>24</issue><spage>9980</spage><pages>9980-</pages><issn>1424-8220</issn><eissn>1424-8220</eissn><abstract>Taxonomy illustrates that natural creatures can be classified with a hierarchy. The connections between species are explicit and objective and can be organized into a knowledge graph (KG). It is a challenging task to mine features of known categories from KG and to reason on unknown categories. Graph Convolutional Network (GCN) has recently been viewed as a potential approach to zero-shot learning. GCN enables knowledge transfer by sharing the statistical strength of nodes in the graph. More layers of graph convolution are stacked in order to aggregate the hierarchical information in the KG. However, the Laplacian over-smoothing problem will be severe as the number of GCN layers deepens, which leads the features between nodes toward a tendency to be similar and degrade the performance of zero-shot image classification tasks. We consider two parts to mitigate the Laplacian over-smoothing problem, namely reducing the invalid node aggregation and improving the discriminability among nodes in the deep graph network. We propose a top-k graph pooling method based on the self-attention mechanism to control specific node aggregation, and we introduce a dual structural symmetric knowledge graph additionally to enhance the representation of nodes in the latent space. Finally, we apply these new concepts to the recently widely used contrastive learning framework and propose a novel Contrastive Graph U-Net with two Attention-based graph pooling (Att-gPool) layers, CGUN-2A, which explicitly alleviates the Laplacian over-smoothing problem. To evaluate the performance of the method on complex real-world scenes, we test it on the large-scale zero-shot image classification dataset. Extensive experiments show the positive effect of allowing nodes to perform specific aggregation, as well as homogeneous graph comparison, in our deep graph network. We show how it significantly boosts zero-shot image classification performance. The Hit@1 accuracy is 17.5% relatively higher than the baseline model on the ImageNet21K dataset.</abstract><cop>Switzerland</cop><pub>MDPI AG</pub><pmid>36560351</pmid><doi>10.3390/s22249980</doi><orcidid>https://orcid.org/0000-0002-5965-109X</orcidid><orcidid>https://orcid.org/0000-0001-6052-4496</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1424-8220
ispartof Sensors (Basel, Switzerland), 2022-12, Vol.22 (24), p.9980
issn 1424-8220
1424-8220
language eng
recordid cdi_doaj_primary_oai_doaj_org_article_0f2483ad6e1947c884df8a2d4f9ca452
source Open Access: PubMed Central; Publicly Available Content Database; Coronavirus Research Database
subjects Algorithms
Classification
contrastive learning
Datasets
Dietary Supplements
Ecology
Environmental monitoring
Graph comparison
graph convolutional network
Graph representations
Image classification
Knowledge
Knowledge management
Learning
machine learning
Neural networks
Nodes
Plankton
Records
Semantics
Smoothing
Taxonomy
Teaching methods
zero-shot learning
title CGUN-2A: Deep Graph Convolutional Network via Contrastive Learning for Large-Scale Zero-Shot Image Classification
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T16%3A55%3A17IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-gale_doaj_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=CGUN-2A:%20Deep%20Graph%20Convolutional%20Network%20via%20Contrastive%20Learning%20for%20Large-Scale%20Zero-Shot%20Image%20Classification&rft.jtitle=Sensors%20(Basel,%20Switzerland)&rft.au=Li,%20Liangwei&rft.date=2022-12-18&rft.volume=22&rft.issue=24&rft.spage=9980&rft.pages=9980-&rft.issn=1424-8220&rft.eissn=1424-8220&rft_id=info:doi/10.3390/s22249980&rft_dat=%3Cgale_doaj_%3EA746535441%3C/gale_doaj_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c508t-394441cd65730afec228a34bb3a0ed200cddbe525ce033f9f0c929027e720da13%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2756782346&rft_id=info:pmid/36560351&rft_galeid=A746535441&rfr_iscdi=true