Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis
Objectives The aim of the present systematic review and meta-analysis is to assess the accuracy of automated landmarking using deep learning in comparison with manual tracing for cephalometric analysis of 3D medical images. Methods PubMed/Medline, IEEE Xplore, Scopus and ArXiv electronic databases w...
Gespeichert in:
| Veröffentlicht in: | Radiologia medica 2023-05, Vol.128 (5), p.544-555 |
|---|---|
| Hauptverfasser: | , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 555 |
|---|---|
| container_issue | 5 |
| container_start_page | 544 |
| container_title | Radiologia medica |
| container_volume | 128 |
| creator | Serafin, Marco Baldini, Benedetta Cabitza, Federico Carrafiello, Gianpaolo Baselli, Giuseppe Del Fabbro, Massimo Sforza, Chiarella Caprioglio, Alberto Tartaglia, Gianluca M. |
| description | Objectives
The aim of the present systematic review and meta-analysis is to assess the accuracy of automated landmarking using deep learning in comparison with manual tracing for cephalometric analysis of 3D medical images.
Methods
PubMed/Medline, IEEE Xplore, Scopus and ArXiv electronic databases were searched. Selection criteria were: ex vivo and in vivo volumetric data images suitable for 3D landmarking (Problem), a minimum of five automated landmarking performed by deep learning method (Intervention), manual landmarking (Comparison), and mean accuracy, in mm, between manual and automated landmarking (Outcome). QUADAS-2 was adapted for quality analysis. Meta-analysis was performed on studies that reported as outcome mean values and standard deviation of the difference (error) between manual and automated landmarking. Linear regression plots were used to analyze correlations between mean accuracy and year of publication.
Results
The initial electronic screening yielded 252 papers published between 2020 and 2022. A total of 15 studies were included for the qualitative synthesis, whereas 11 studies were used for the meta-analysis. Overall random effect model revealed a mean value of 2.44 mm, with a high heterogeneity (
I
2
= 98.13%,
τ
2
= 1.018,
p
-value |
| doi_str_mv | 10.1007/s11547-023-01629-2 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_10181977</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2813001378</sourcerecordid><originalsourceid>FETCH-LOGICAL-c475t-3ead039d4d0d2e1b4f22661ebea884de8945699905db605d8a463c94b9a1f45d3</originalsourceid><addsrcrecordid>eNp9kU1v1DAQhi0EoqXwBzggS1y4BMYfSWwuqCq0IFXiAmfLsSe7Lkm8tZNW-fd42VIKBy62pXnmsccvIS8ZvGUA7bvMWC3bCriogDVcV_wROWaKN1WjlXj84HxEnuV8BSCBgX5KjkQLWgjRHpPrU-eWZN1KY0_tMsfRzuip-Egd7rZ2iCPOKTg62MmPNv3ItFupR9zRAW2awrShdtjEFObtmN_TvOYZi6J0JLwJeEtLHy0OW9nJDmsO-Tl50tsh44u7_YR8P__07exzdfn14svZ6WXlZFvPlUDrQWgvPXiOrJM9503DsEOrlPSotKwbrTXUvmvKoqxshNOy05b1svbihHw4eHdLN6J3OM3JDmaXQpljNdEG83dlCluziTeGAVNMt20xvLkzpHi9YJ7NGLLDofwFxiUbrqCumeSaF_T1P-hVXFKZeE8xAcBEqwrFD5RLMeeE_f1rGJh9pOYQqSmRml-Rmr361cM57lt-Z1gAcQByKU0bTH_u_o_2J8_orl8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2813001378</pqid></control><display><type>article</type><title>Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis</title><source>MEDLINE</source><source>SpringerLink Journals</source><creator>Serafin, Marco ; Baldini, Benedetta ; Cabitza, Federico ; Carrafiello, Gianpaolo ; Baselli, Giuseppe ; Del Fabbro, Massimo ; Sforza, Chiarella ; Caprioglio, Alberto ; Tartaglia, Gianluca M.</creator><creatorcontrib>Serafin, Marco ; Baldini, Benedetta ; Cabitza, Federico ; Carrafiello, Gianpaolo ; Baselli, Giuseppe ; Del Fabbro, Massimo ; Sforza, Chiarella ; Caprioglio, Alberto ; Tartaglia, Gianluca M.</creatorcontrib><description>Objectives
The aim of the present systematic review and meta-analysis is to assess the accuracy of automated landmarking using deep learning in comparison with manual tracing for cephalometric analysis of 3D medical images.
Methods
PubMed/Medline, IEEE Xplore, Scopus and ArXiv electronic databases were searched. Selection criteria were: ex vivo and in vivo volumetric data images suitable for 3D landmarking (Problem), a minimum of five automated landmarking performed by deep learning method (Intervention), manual landmarking (Comparison), and mean accuracy, in mm, between manual and automated landmarking (Outcome). QUADAS-2 was adapted for quality analysis. Meta-analysis was performed on studies that reported as outcome mean values and standard deviation of the difference (error) between manual and automated landmarking. Linear regression plots were used to analyze correlations between mean accuracy and year of publication.
Results
The initial electronic screening yielded 252 papers published between 2020 and 2022. A total of 15 studies were included for the qualitative synthesis, whereas 11 studies were used for the meta-analysis. Overall random effect model revealed a mean value of 2.44 mm, with a high heterogeneity (
I
2
= 98.13%,
τ
2
= 1.018,
p
-value < 0.001); risk of bias was high due to the presence of issues for several domains per study. Meta-regression indicated a significant relation between mean error and year of publication (
p
value = 0.012).
Conclusion
Deep learning algorithms showed an excellent accuracy for automated 3D cephalometric landmarking. In the last two years promising algorithms have been developed and improvements in landmarks annotation accuracy have been done.</description><identifier>ISSN: 1826-6983</identifier><identifier>ISSN: 0033-8362</identifier><identifier>EISSN: 1826-6983</identifier><identifier>DOI: 10.1007/s11547-023-01629-2</identifier><identifier>PMID: 37093337</identifier><language>eng</language><publisher>Milan: Springer Milan</publisher><subject>Accuracy ; Algorithms ; Anatomic Landmarks ; Annotations ; Automation ; Cephalometry - methods ; Computed Tomography ; Deep Learning ; Diagnostic Radiology ; Heterogeneity ; Humans ; Imaging ; Imaging, Three-Dimensional - methods ; Interventional Radiology ; Machine learning ; Medical imaging ; Medicine ; Medicine & Public Health ; Meta-analysis ; Neuroradiology ; Qualitative analysis ; Radiology ; Reproducibility of Results ; Systematic review ; Ultrasound</subject><ispartof>Radiologia medica, 2023-05, Vol.128 (5), p.544-555</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-3ead039d4d0d2e1b4f22661ebea884de8945699905db605d8a463c94b9a1f45d3</citedby><cites>FETCH-LOGICAL-c475t-3ead039d4d0d2e1b4f22661ebea884de8945699905db605d8a463c94b9a1f45d3</cites><orcidid>0000-0001-7062-5143 ; 0000-0001-7144-0984 ; 0000-0001-6532-6464 ; 0000-0003-4680-0989 ; 0000-0003-4365-4998 ; 0000-0003-1448-2808 ; 0000-0002-4065-3415 ; 0000-0003-2978-1704 ; 0000-0002-8264-7320</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s11547-023-01629-2$$EPDF$$P50$$Gspringer$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s11547-023-01629-2$$EHTML$$P50$$Gspringer$$Hfree_for_read</linktohtml><link.rule.ids>230,314,776,780,881,27901,27902,41464,42533,51294</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/37093337$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Serafin, Marco</creatorcontrib><creatorcontrib>Baldini, Benedetta</creatorcontrib><creatorcontrib>Cabitza, Federico</creatorcontrib><creatorcontrib>Carrafiello, Gianpaolo</creatorcontrib><creatorcontrib>Baselli, Giuseppe</creatorcontrib><creatorcontrib>Del Fabbro, Massimo</creatorcontrib><creatorcontrib>Sforza, Chiarella</creatorcontrib><creatorcontrib>Caprioglio, Alberto</creatorcontrib><creatorcontrib>Tartaglia, Gianluca M.</creatorcontrib><title>Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis</title><title>Radiologia medica</title><addtitle>Radiol med</addtitle><addtitle>Radiol Med</addtitle><description>Objectives
The aim of the present systematic review and meta-analysis is to assess the accuracy of automated landmarking using deep learning in comparison with manual tracing for cephalometric analysis of 3D medical images.
Methods
PubMed/Medline, IEEE Xplore, Scopus and ArXiv electronic databases were searched. Selection criteria were: ex vivo and in vivo volumetric data images suitable for 3D landmarking (Problem), a minimum of five automated landmarking performed by deep learning method (Intervention), manual landmarking (Comparison), and mean accuracy, in mm, between manual and automated landmarking (Outcome). QUADAS-2 was adapted for quality analysis. Meta-analysis was performed on studies that reported as outcome mean values and standard deviation of the difference (error) between manual and automated landmarking. Linear regression plots were used to analyze correlations between mean accuracy and year of publication.
Results
The initial electronic screening yielded 252 papers published between 2020 and 2022. A total of 15 studies were included for the qualitative synthesis, whereas 11 studies were used for the meta-analysis. Overall random effect model revealed a mean value of 2.44 mm, with a high heterogeneity (
I
2
= 98.13%,
τ
2
= 1.018,
p
-value < 0.001); risk of bias was high due to the presence of issues for several domains per study. Meta-regression indicated a significant relation between mean error and year of publication (
p
value = 0.012).
Conclusion
Deep learning algorithms showed an excellent accuracy for automated 3D cephalometric landmarking. In the last two years promising algorithms have been developed and improvements in landmarks annotation accuracy have been done.</description><subject>Accuracy</subject><subject>Algorithms</subject><subject>Anatomic Landmarks</subject><subject>Annotations</subject><subject>Automation</subject><subject>Cephalometry - methods</subject><subject>Computed Tomography</subject><subject>Deep Learning</subject><subject>Diagnostic Radiology</subject><subject>Heterogeneity</subject><subject>Humans</subject><subject>Imaging</subject><subject>Imaging, Three-Dimensional - methods</subject><subject>Interventional Radiology</subject><subject>Machine learning</subject><subject>Medical imaging</subject><subject>Medicine</subject><subject>Medicine & Public Health</subject><subject>Meta-analysis</subject><subject>Neuroradiology</subject><subject>Qualitative analysis</subject><subject>Radiology</subject><subject>Reproducibility of Results</subject><subject>Systematic review</subject><subject>Ultrasound</subject><issn>1826-6983</issn><issn>0033-8362</issn><issn>1826-6983</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>C6C</sourceid><sourceid>EIF</sourceid><recordid>eNp9kU1v1DAQhi0EoqXwBzggS1y4BMYfSWwuqCq0IFXiAmfLsSe7Lkm8tZNW-fd42VIKBy62pXnmsccvIS8ZvGUA7bvMWC3bCriogDVcV_wROWaKN1WjlXj84HxEnuV8BSCBgX5KjkQLWgjRHpPrU-eWZN1KY0_tMsfRzuip-Egd7rZ2iCPOKTg62MmPNv3ItFupR9zRAW2awrShdtjEFObtmN_TvOYZi6J0JLwJeEtLHy0OW9nJDmsO-Tl50tsh44u7_YR8P__07exzdfn14svZ6WXlZFvPlUDrQWgvPXiOrJM9503DsEOrlPSotKwbrTXUvmvKoqxshNOy05b1svbihHw4eHdLN6J3OM3JDmaXQpljNdEG83dlCluziTeGAVNMt20xvLkzpHi9YJ7NGLLDofwFxiUbrqCumeSaF_T1P-hVXFKZeE8xAcBEqwrFD5RLMeeE_f1rGJh9pOYQqSmRml-Rmr361cM57lt-Z1gAcQByKU0bTH_u_o_2J8_orl8</recordid><startdate>20230501</startdate><enddate>20230501</enddate><creator>Serafin, Marco</creator><creator>Baldini, Benedetta</creator><creator>Cabitza, Federico</creator><creator>Carrafiello, Gianpaolo</creator><creator>Baselli, Giuseppe</creator><creator>Del Fabbro, Massimo</creator><creator>Sforza, Chiarella</creator><creator>Caprioglio, Alberto</creator><creator>Tartaglia, Gianluca M.</creator><general>Springer Milan</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>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0001-7062-5143</orcidid><orcidid>https://orcid.org/0000-0001-7144-0984</orcidid><orcidid>https://orcid.org/0000-0001-6532-6464</orcidid><orcidid>https://orcid.org/0000-0003-4680-0989</orcidid><orcidid>https://orcid.org/0000-0003-4365-4998</orcidid><orcidid>https://orcid.org/0000-0003-1448-2808</orcidid><orcidid>https://orcid.org/0000-0002-4065-3415</orcidid><orcidid>https://orcid.org/0000-0003-2978-1704</orcidid><orcidid>https://orcid.org/0000-0002-8264-7320</orcidid></search><sort><creationdate>20230501</creationdate><title>Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis</title><author>Serafin, Marco ; Baldini, Benedetta ; Cabitza, Federico ; Carrafiello, Gianpaolo ; Baselli, Giuseppe ; Del Fabbro, Massimo ; Sforza, Chiarella ; Caprioglio, Alberto ; Tartaglia, Gianluca M.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c475t-3ead039d4d0d2e1b4f22661ebea884de8945699905db605d8a463c94b9a1f45d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Accuracy</topic><topic>Algorithms</topic><topic>Anatomic Landmarks</topic><topic>Annotations</topic><topic>Automation</topic><topic>Cephalometry - methods</topic><topic>Computed Tomography</topic><topic>Deep Learning</topic><topic>Diagnostic Radiology</topic><topic>Heterogeneity</topic><topic>Humans</topic><topic>Imaging</topic><topic>Imaging, Three-Dimensional - methods</topic><topic>Interventional Radiology</topic><topic>Machine learning</topic><topic>Medical imaging</topic><topic>Medicine</topic><topic>Medicine & Public Health</topic><topic>Meta-analysis</topic><topic>Neuroradiology</topic><topic>Qualitative analysis</topic><topic>Radiology</topic><topic>Reproducibility of Results</topic><topic>Systematic review</topic><topic>Ultrasound</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Serafin, Marco</creatorcontrib><creatorcontrib>Baldini, Benedetta</creatorcontrib><creatorcontrib>Cabitza, Federico</creatorcontrib><creatorcontrib>Carrafiello, Gianpaolo</creatorcontrib><creatorcontrib>Baselli, Giuseppe</creatorcontrib><creatorcontrib>Del Fabbro, Massimo</creatorcontrib><creatorcontrib>Sforza, Chiarella</creatorcontrib><creatorcontrib>Caprioglio, Alberto</creatorcontrib><creatorcontrib>Tartaglia, Gianluca M.</creatorcontrib><collection>Springer Nature OA/Free Journals</collection><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><collection>PubMed Central (Full Participant titles)</collection><jtitle>Radiologia medica</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Serafin, Marco</au><au>Baldini, Benedetta</au><au>Cabitza, Federico</au><au>Carrafiello, Gianpaolo</au><au>Baselli, Giuseppe</au><au>Del Fabbro, Massimo</au><au>Sforza, Chiarella</au><au>Caprioglio, Alberto</au><au>Tartaglia, Gianluca M.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis</atitle><jtitle>Radiologia medica</jtitle><stitle>Radiol med</stitle><addtitle>Radiol Med</addtitle><date>2023-05-01</date><risdate>2023</risdate><volume>128</volume><issue>5</issue><spage>544</spage><epage>555</epage><pages>544-555</pages><issn>1826-6983</issn><issn>0033-8362</issn><eissn>1826-6983</eissn><abstract>Objectives
The aim of the present systematic review and meta-analysis is to assess the accuracy of automated landmarking using deep learning in comparison with manual tracing for cephalometric analysis of 3D medical images.
Methods
PubMed/Medline, IEEE Xplore, Scopus and ArXiv electronic databases were searched. Selection criteria were: ex vivo and in vivo volumetric data images suitable for 3D landmarking (Problem), a minimum of five automated landmarking performed by deep learning method (Intervention), manual landmarking (Comparison), and mean accuracy, in mm, between manual and automated landmarking (Outcome). QUADAS-2 was adapted for quality analysis. Meta-analysis was performed on studies that reported as outcome mean values and standard deviation of the difference (error) between manual and automated landmarking. Linear regression plots were used to analyze correlations between mean accuracy and year of publication.
Results
The initial electronic screening yielded 252 papers published between 2020 and 2022. A total of 15 studies were included for the qualitative synthesis, whereas 11 studies were used for the meta-analysis. Overall random effect model revealed a mean value of 2.44 mm, with a high heterogeneity (
I
2
= 98.13%,
τ
2
= 1.018,
p
-value < 0.001); risk of bias was high due to the presence of issues for several domains per study. Meta-regression indicated a significant relation between mean error and year of publication (
p
value = 0.012).
Conclusion
Deep learning algorithms showed an excellent accuracy for automated 3D cephalometric landmarking. In the last two years promising algorithms have been developed and improvements in landmarks annotation accuracy have been done.</abstract><cop>Milan</cop><pub>Springer Milan</pub><pmid>37093337</pmid><doi>10.1007/s11547-023-01629-2</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0001-7062-5143</orcidid><orcidid>https://orcid.org/0000-0001-7144-0984</orcidid><orcidid>https://orcid.org/0000-0001-6532-6464</orcidid><orcidid>https://orcid.org/0000-0003-4680-0989</orcidid><orcidid>https://orcid.org/0000-0003-4365-4998</orcidid><orcidid>https://orcid.org/0000-0003-1448-2808</orcidid><orcidid>https://orcid.org/0000-0002-4065-3415</orcidid><orcidid>https://orcid.org/0000-0003-2978-1704</orcidid><orcidid>https://orcid.org/0000-0002-8264-7320</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1826-6983 |
| ispartof | Radiologia medica, 2023-05, Vol.128 (5), p.544-555 |
| issn | 1826-6983 0033-8362 1826-6983 |
| language | eng |
| recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_10181977 |
| source | MEDLINE; SpringerLink Journals |
| subjects | Accuracy Algorithms Anatomic Landmarks Annotations Automation Cephalometry - methods Computed Tomography Deep Learning Diagnostic Radiology Heterogeneity Humans Imaging Imaging, Three-Dimensional - methods Interventional Radiology Machine learning Medical imaging Medicine Medicine & Public Health Meta-analysis Neuroradiology Qualitative analysis Radiology Reproducibility of Results Systematic review Ultrasound |
| title | Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: systematic review and meta-analysis |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T06%3A50%3A15IST&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=Accuracy%20of%20automated%203D%20cephalometric%20landmarks%20by%20deep%20learning%20algorithms:%20systematic%20review%20and%20meta-analysis&rft.jtitle=Radiologia%20medica&rft.au=Serafin,%20Marco&rft.date=2023-05-01&rft.volume=128&rft.issue=5&rft.spage=544&rft.epage=555&rft.pages=544-555&rft.issn=1826-6983&rft.eissn=1826-6983&rft_id=info:doi/10.1007/s11547-023-01629-2&rft_dat=%3Cproquest_pubme%3E2813001378%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2813001378&rft_id=info:pmid/37093337&rfr_iscdi=true |