Loading…
Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms
Objectives To validate an AI system for standalone breast cancer detection on an entire screening population in comparison to first-reading breast radiologists. Materials and methods All mammography screenings performed between August 4, 2014, and August 15, 2018, in the Region of Southern Denmark w...
Saved in:
| Published in: | European radiology 2024-06, Vol.34 (6), p.3935-3946 |
|---|---|
| Main Authors: | , , , , , , , |
| 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-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3 |
|---|---|
| cites | cdi_FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3 |
| container_end_page | 3946 |
| container_issue | 6 |
| container_start_page | 3935 |
| container_title | European radiology |
| container_volume | 34 |
| creator | Kühl, Johanne Elhakim, Mohammad Talal Stougaard, Sarah Wordenskjold Rasmussen, Benjamin Schnack Brandt Nielsen, Mads Gerke, Oke Larsen, Lisbet Brønsro Graumann, Ole |
| description | Objectives
To validate an AI system for standalone breast cancer detection on an entire screening population in comparison to first-reading breast radiologists.
Materials and methods
All mammography screenings performed between August 4, 2014, and August 15, 2018, in the Region of Southern Denmark with follow-up within 24 months were eligible. Screenings were assessed as normal or abnormal by breast radiologists through double reading with arbitration. For an AI decision of normal or abnormal, two AI-score cut-off points were applied by matching at mean sensitivity (AI
sens
) and specificity (AI
spec
) of first readers. Accuracy measures were sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and recall rate (RR).
Results
The sample included 249,402 screenings (149,495 women) and 2033 breast cancers (72.6% screen-detected cancers, 27.4% interval cancers). AI
sens
had lower specificity (97.5% vs 97.7%;
p
|
| doi_str_mv | 10.1007/s00330-023-10423-7 |
| format | article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_11166831</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2887474858</sourcerecordid><originalsourceid>FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3</originalsourceid><addsrcrecordid>eNp9UcFu1TAQtBCIlsIPcECRuHAJ2F7HTk4IVUCRKsEBztbW3qSuEvthJ634-7p9pRQOXOxd7czsjoaxl4K_FZybd4VzAN5yCa3gqr7mETsUCmRte_X4QX3AnpVywTkfhDJP2QGYAXro9SE7_5Z224xrSLG9Cp4ausR5u-2bNDaY1zAGF3BuQlxpnsNE0VGD0TcZfUhzmkJZGyyFSlkorjes4jJRDHFqFlyWNGVcynP2ZMS50Iu7_4j9-PTx-_FJe_r185fjD6etU6ZbWzQjaOGJlCQPxpjOOUUDgPTogeQgjebGAa8lF464VlKhH5yBM98BwRF7v9fdbWcLeVdPyjjbXQ4L5l82YbB_T2I4t1O6tEIIrXsQVeHNnUJOPzcqq11CcdU7RkpbsbLvjTKq7_oKff0P9CJtOVZ_FrjW3WC00BUl9yiXUymZxvtrBLc3Sdp9krYmaW-TtKaSXj30cU_5HV0FwB5Q6ihOlP_s_o_sNVfjq-0</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3066597616</pqid></control><display><type>article</type><title>Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms</title><source>Springer Nature</source><creator>Kühl, Johanne ; Elhakim, Mohammad Talal ; Stougaard, Sarah Wordenskjold ; Rasmussen, Benjamin Schnack Brandt ; Nielsen, Mads ; Gerke, Oke ; Larsen, Lisbet Brønsro ; Graumann, Ole</creator><creatorcontrib>Kühl, Johanne ; Elhakim, Mohammad Talal ; Stougaard, Sarah Wordenskjold ; Rasmussen, Benjamin Schnack Brandt ; Nielsen, Mads ; Gerke, Oke ; Larsen, Lisbet Brønsro ; Graumann, Ole</creatorcontrib><description>Objectives
To validate an AI system for standalone breast cancer detection on an entire screening population in comparison to first-reading breast radiologists.
Materials and methods
All mammography screenings performed between August 4, 2014, and August 15, 2018, in the Region of Southern Denmark with follow-up within 24 months were eligible. Screenings were assessed as normal or abnormal by breast radiologists through double reading with arbitration. For an AI decision of normal or abnormal, two AI-score cut-off points were applied by matching at mean sensitivity (AI
sens
) and specificity (AI
spec
) of first readers. Accuracy measures were sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and recall rate (RR).
Results
The sample included 249,402 screenings (149,495 women) and 2033 breast cancers (72.6% screen-detected cancers, 27.4% interval cancers). AI
sens
had lower specificity (97.5% vs 97.7%;
p
< 0.0001) and PPV (17.5% vs 18.7%;
p
= 0.01) and a higher RR (3.0% vs 2.8%;
p
< 0.0001) than first readers. AI
spec
was comparable to first readers in terms of all accuracy measures. Both AI
sens
and AI
spec
detected significantly fewer screen-detected cancers (1166 (AI
sens
), 1156 (AI
spec
) vs 1252;
p
< 0.0001) but found more interval cancers compared to first readers (126 (AI
sens
), 117 (AI
spec
) vs 39;
p
< 0.0001) with varying types of cancers detected across multiple subgroups.
Conclusion
Standalone AI can detect breast cancer at an accuracy level equivalent to the standard of first readers when the AI threshold point was matched at first reader specificity. However, AI and first readers detected a different composition of cancers.
Clinical relevance statement
Replacing first readers with AI with an appropriate cut-off score could be feasible. AI-detected cancers not detected by radiologists suggest a potential increase in the number of cancers detected if AI is implemented to support double reading within screening, although the clinicopathological characteristics of detected cancers would not change significantly.
Key Points
•
Standalone AI cancer detection was compared to first readers in a double-read mammography screening population.
•
Standalone AI matched at first reader specificity showed no statistically significant difference in overall accuracy but detected different cancers.
•
With an appropriate threshold, AI-integrated screening can increase the number of detected cancers with similar clinicopathological characteristics.</description><identifier>ISSN: 1432-1084</identifier><identifier>ISSN: 0938-7994</identifier><identifier>EISSN: 1432-1084</identifier><identifier>DOI: 10.1007/s00330-023-10423-7</identifier><identifier>PMID: 37938386</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Accuracy ; Aged ; Arbitration ; Artificial Intelligence ; Breast ; Breast cancer ; Breast Neoplasms - diagnostic imaging ; Denmark ; Diagnostic Radiology ; Early Detection of Cancer - methods ; Female ; Humans ; Imaging ; Internal Medicine ; Interventional Radiology ; Mammography ; Mammography - methods ; Mass Screening - methods ; Medicine ; Medicine & Public Health ; Middle Aged ; Neuroradiology ; Population (statistical) ; Radiologists - statistics & numerical data ; Radiology ; Sensitivity ; Sensitivity and Specificity ; Statistical analysis ; Subgroups ; Ultrasound</subject><ispartof>European radiology, 2024-06, Vol.34 (6), p.3935-3946</ispartof><rights>The Author(s) 2023</rights><rights>2023. The Author(s).</rights><rights>The Author(s) 2023. 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><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3</citedby><cites>FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3</cites><orcidid>0000-0002-2673-2352</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27923,27924</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/37938386$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Kühl, Johanne</creatorcontrib><creatorcontrib>Elhakim, Mohammad Talal</creatorcontrib><creatorcontrib>Stougaard, Sarah Wordenskjold</creatorcontrib><creatorcontrib>Rasmussen, Benjamin Schnack Brandt</creatorcontrib><creatorcontrib>Nielsen, Mads</creatorcontrib><creatorcontrib>Gerke, Oke</creatorcontrib><creatorcontrib>Larsen, Lisbet Brønsro</creatorcontrib><creatorcontrib>Graumann, Ole</creatorcontrib><title>Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms</title><title>European radiology</title><addtitle>Eur Radiol</addtitle><addtitle>Eur Radiol</addtitle><description>Objectives
To validate an AI system for standalone breast cancer detection on an entire screening population in comparison to first-reading breast radiologists.
Materials and methods
All mammography screenings performed between August 4, 2014, and August 15, 2018, in the Region of Southern Denmark with follow-up within 24 months were eligible. Screenings were assessed as normal or abnormal by breast radiologists through double reading with arbitration. For an AI decision of normal or abnormal, two AI-score cut-off points were applied by matching at mean sensitivity (AI
sens
) and specificity (AI
spec
) of first readers. Accuracy measures were sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and recall rate (RR).
Results
The sample included 249,402 screenings (149,495 women) and 2033 breast cancers (72.6% screen-detected cancers, 27.4% interval cancers). AI
sens
had lower specificity (97.5% vs 97.7%;
p
< 0.0001) and PPV (17.5% vs 18.7%;
p
= 0.01) and a higher RR (3.0% vs 2.8%;
p
< 0.0001) than first readers. AI
spec
was comparable to first readers in terms of all accuracy measures. Both AI
sens
and AI
spec
detected significantly fewer screen-detected cancers (1166 (AI
sens
), 1156 (AI
spec
) vs 1252;
p
< 0.0001) but found more interval cancers compared to first readers (126 (AI
sens
), 117 (AI
spec
) vs 39;
p
< 0.0001) with varying types of cancers detected across multiple subgroups.
Conclusion
Standalone AI can detect breast cancer at an accuracy level equivalent to the standard of first readers when the AI threshold point was matched at first reader specificity. However, AI and first readers detected a different composition of cancers.
Clinical relevance statement
Replacing first readers with AI with an appropriate cut-off score could be feasible. AI-detected cancers not detected by radiologists suggest a potential increase in the number of cancers detected if AI is implemented to support double reading within screening, although the clinicopathological characteristics of detected cancers would not change significantly.
Key Points
•
Standalone AI cancer detection was compared to first readers in a double-read mammography screening population.
•
Standalone AI matched at first reader specificity showed no statistically significant difference in overall accuracy but detected different cancers.
•
With an appropriate threshold, AI-integrated screening can increase the number of detected cancers with similar clinicopathological characteristics.</description><subject>Accuracy</subject><subject>Aged</subject><subject>Arbitration</subject><subject>Artificial Intelligence</subject><subject>Breast</subject><subject>Breast cancer</subject><subject>Breast Neoplasms - diagnostic imaging</subject><subject>Denmark</subject><subject>Diagnostic Radiology</subject><subject>Early Detection of Cancer - methods</subject><subject>Female</subject><subject>Humans</subject><subject>Imaging</subject><subject>Internal Medicine</subject><subject>Interventional Radiology</subject><subject>Mammography</subject><subject>Mammography - methods</subject><subject>Mass Screening - methods</subject><subject>Medicine</subject><subject>Medicine & Public Health</subject><subject>Middle Aged</subject><subject>Neuroradiology</subject><subject>Population (statistical)</subject><subject>Radiologists - statistics & numerical data</subject><subject>Radiology</subject><subject>Sensitivity</subject><subject>Sensitivity and Specificity</subject><subject>Statistical analysis</subject><subject>Subgroups</subject><subject>Ultrasound</subject><issn>1432-1084</issn><issn>0938-7994</issn><issn>1432-1084</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp9UcFu1TAQtBCIlsIPcECRuHAJ2F7HTk4IVUCRKsEBztbW3qSuEvthJ634-7p9pRQOXOxd7czsjoaxl4K_FZybd4VzAN5yCa3gqr7mETsUCmRte_X4QX3AnpVywTkfhDJP2QGYAXro9SE7_5Z224xrSLG9Cp4ausR5u-2bNDaY1zAGF3BuQlxpnsNE0VGD0TcZfUhzmkJZGyyFSlkorjes4jJRDHFqFlyWNGVcynP2ZMS50Iu7_4j9-PTx-_FJe_r185fjD6etU6ZbWzQjaOGJlCQPxpjOOUUDgPTogeQgjebGAa8lF464VlKhH5yBM98BwRF7v9fdbWcLeVdPyjjbXQ4L5l82YbB_T2I4t1O6tEIIrXsQVeHNnUJOPzcqq11CcdU7RkpbsbLvjTKq7_oKff0P9CJtOVZ_FrjW3WC00BUl9yiXUymZxvtrBLc3Sdp9krYmaW-TtKaSXj30cU_5HV0FwB5Q6ihOlP_s_o_sNVfjq-0</recordid><startdate>20240601</startdate><enddate>20240601</enddate><creator>Kühl, Johanne</creator><creator>Elhakim, Mohammad Talal</creator><creator>Stougaard, Sarah Wordenskjold</creator><creator>Rasmussen, Benjamin Schnack Brandt</creator><creator>Nielsen, Mads</creator><creator>Gerke, Oke</creator><creator>Larsen, Lisbet Brønsro</creator><creator>Graumann, Ole</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><scope>C6C</scope><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>7QO</scope><scope>8FD</scope><scope>FR3</scope><scope>K9.</scope><scope>NAPCQ</scope><scope>P64</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-2673-2352</orcidid></search><sort><creationdate>20240601</creationdate><title>Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms</title><author>Kühl, Johanne ; Elhakim, Mohammad Talal ; Stougaard, Sarah Wordenskjold ; Rasmussen, Benjamin Schnack Brandt ; Nielsen, Mads ; Gerke, Oke ; Larsen, Lisbet Brønsro ; Graumann, Ole</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Aged</topic><topic>Arbitration</topic><topic>Artificial Intelligence</topic><topic>Breast</topic><topic>Breast cancer</topic><topic>Breast Neoplasms - diagnostic imaging</topic><topic>Denmark</topic><topic>Diagnostic Radiology</topic><topic>Early Detection of Cancer - methods</topic><topic>Female</topic><topic>Humans</topic><topic>Imaging</topic><topic>Internal Medicine</topic><topic>Interventional Radiology</topic><topic>Mammography</topic><topic>Mammography - methods</topic><topic>Mass Screening - methods</topic><topic>Medicine</topic><topic>Medicine & Public Health</topic><topic>Middle Aged</topic><topic>Neuroradiology</topic><topic>Population (statistical)</topic><topic>Radiologists - statistics & numerical data</topic><topic>Radiology</topic><topic>Sensitivity</topic><topic>Sensitivity and Specificity</topic><topic>Statistical analysis</topic><topic>Subgroups</topic><topic>Ultrasound</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kühl, Johanne</creatorcontrib><creatorcontrib>Elhakim, Mohammad Talal</creatorcontrib><creatorcontrib>Stougaard, Sarah Wordenskjold</creatorcontrib><creatorcontrib>Rasmussen, Benjamin Schnack Brandt</creatorcontrib><creatorcontrib>Nielsen, Mads</creatorcontrib><creatorcontrib>Gerke, Oke</creatorcontrib><creatorcontrib>Larsen, Lisbet Brønsro</creatorcontrib><creatorcontrib>Graumann, Ole</creatorcontrib><collection>SpringerOpen (Open Access)</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Biotechnology Research Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Nursing & Allied Health Premium</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>European radiology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kühl, Johanne</au><au>Elhakim, Mohammad Talal</au><au>Stougaard, Sarah Wordenskjold</au><au>Rasmussen, Benjamin Schnack Brandt</au><au>Nielsen, Mads</au><au>Gerke, Oke</au><au>Larsen, Lisbet Brønsro</au><au>Graumann, Ole</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms</atitle><jtitle>European radiology</jtitle><stitle>Eur Radiol</stitle><addtitle>Eur Radiol</addtitle><date>2024-06-01</date><risdate>2024</risdate><volume>34</volume><issue>6</issue><spage>3935</spage><epage>3946</epage><pages>3935-3946</pages><issn>1432-1084</issn><issn>0938-7994</issn><eissn>1432-1084</eissn><abstract>Objectives
To validate an AI system for standalone breast cancer detection on an entire screening population in comparison to first-reading breast radiologists.
Materials and methods
All mammography screenings performed between August 4, 2014, and August 15, 2018, in the Region of Southern Denmark with follow-up within 24 months were eligible. Screenings were assessed as normal or abnormal by breast radiologists through double reading with arbitration. For an AI decision of normal or abnormal, two AI-score cut-off points were applied by matching at mean sensitivity (AI
sens
) and specificity (AI
spec
) of first readers. Accuracy measures were sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and recall rate (RR).
Results
The sample included 249,402 screenings (149,495 women) and 2033 breast cancers (72.6% screen-detected cancers, 27.4% interval cancers). AI
sens
had lower specificity (97.5% vs 97.7%;
p
< 0.0001) and PPV (17.5% vs 18.7%;
p
= 0.01) and a higher RR (3.0% vs 2.8%;
p
< 0.0001) than first readers. AI
spec
was comparable to first readers in terms of all accuracy measures. Both AI
sens
and AI
spec
detected significantly fewer screen-detected cancers (1166 (AI
sens
), 1156 (AI
spec
) vs 1252;
p
< 0.0001) but found more interval cancers compared to first readers (126 (AI
sens
), 117 (AI
spec
) vs 39;
p
< 0.0001) with varying types of cancers detected across multiple subgroups.
Conclusion
Standalone AI can detect breast cancer at an accuracy level equivalent to the standard of first readers when the AI threshold point was matched at first reader specificity. However, AI and first readers detected a different composition of cancers.
Clinical relevance statement
Replacing first readers with AI with an appropriate cut-off score could be feasible. AI-detected cancers not detected by radiologists suggest a potential increase in the number of cancers detected if AI is implemented to support double reading within screening, although the clinicopathological characteristics of detected cancers would not change significantly.
Key Points
•
Standalone AI cancer detection was compared to first readers in a double-read mammography screening population.
•
Standalone AI matched at first reader specificity showed no statistically significant difference in overall accuracy but detected different cancers.
•
With an appropriate threshold, AI-integrated screening can increase the number of detected cancers with similar clinicopathological characteristics.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><pmid>37938386</pmid><doi>10.1007/s00330-023-10423-7</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0002-2673-2352</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1432-1084 |
| ispartof | European radiology, 2024-06, Vol.34 (6), p.3935-3946 |
| issn | 1432-1084 0938-7994 1432-1084 |
| language | eng |
| recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_11166831 |
| source | Springer Nature |
| subjects | Accuracy Aged Arbitration Artificial Intelligence Breast Breast cancer Breast Neoplasms - diagnostic imaging Denmark Diagnostic Radiology Early Detection of Cancer - methods Female Humans Imaging Internal Medicine Interventional Radiology Mammography Mammography - methods Mass Screening - methods Medicine Medicine & Public Health Middle Aged Neuroradiology Population (statistical) Radiologists - statistics & numerical data Radiology Sensitivity Sensitivity and Specificity Statistical analysis Subgroups Ultrasound |
| title | Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-09T02%3A57%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Population-wide%20evaluation%20of%20artificial%20intelligence%20and%20radiologist%20assessment%20of%20screening%20mammograms&rft.jtitle=European%20radiology&rft.au=K%C3%BChl,%20Johanne&rft.date=2024-06-01&rft.volume=34&rft.issue=6&rft.spage=3935&rft.epage=3946&rft.pages=3935-3946&rft.issn=1432-1084&rft.eissn=1432-1084&rft_id=info:doi/10.1007/s00330-023-10423-7&rft_dat=%3Cproquest_pubme%3E2887474858%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c475t-a7f361dee42ed37775cc4e9332dad3e2927607c30e2901ce06424ad9c73bd53e3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=3066597616&rft_id=info:pmid/37938386&rfr_iscdi=true |