A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection
Objectives To develop a machine learning (ML)‐assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings. Patients and Methods In all, 248 patients treated...
Gespeichert in:
| Veröffentlicht in: | BJU international 2019-12, Vol.124 (6), p.972-983 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 983 |
|---|---|
| container_issue | 6 |
| container_start_page | 972 |
| container_title | BJU international |
| container_volume | 124 |
| creator | Hou, Ying Bao, Mei‐Ling Wu, Chen‐Jiang Zhang, Jing Zhang, Yu‐Dong Shi, Hai‐Bin |
| description | Objectives
To develop a machine learning (ML)‐assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings.
Patients and Methods
In all, 248 patients treated with radical prostatectomy and ePLND or PLND were included. ML‐assisted models were developed from 18 integrated features using logistic regression (LR), support vector machine (SVM), and random forests (RFs). The models were compared to the Memorial SloanKettering Cancer Center (MSKCC) nomogram using receiver operating characteristic‐derived area under the curve (AUC) calibration plots and decision curve analysis (DCA).
Results
A total of 59/248 (23.8%) lymph node invasions (LNIs) were identified at surgery. The predictive accuracy of the ML‐based models, with (+) or without (−) MRI‐reported LNI, yielded similar AUCs (RFs+/RFs−: 0.906/0.885; SVM+/SVM−: 0.891/0.868; LR+/LR−: 0.886/0.882) and were higher than the MSKCC nomogram (0.816; P |
| doi_str_mv | 10.1111/bju.14892 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2270006245</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2317759025</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3532-57a26dcec6460f5b222dc92a46845d313fb955a9661db6b0e4f5da83ff2645293</originalsourceid><addsrcrecordid>eNp1kcFOFTEUhhuiAQQWvoBp4gYXF9pO2zuzRKKoIXEDCbum057x9mamM7Qd4e58BNc-nk_CwQsuTOymJydfvp7Tn5DXnJ1wPKftej7hsm7EDtnnUsuF5OzmxXPNGr1HXuW8ZgwbWu2SvYpXjahZvU9-ndHBulWIQHuwKYb47fePnzbnkAt46sGFHMaIvTxP05gKHUYPPS0jbaEUSDR4iCV0GzrZErDM9C6UFZ3SmIstQJ2NDrEEt3NIqKc2UrgvED36J-i_B0f7zTCtaEQz9SFncAXfPCQvO9tnOHq6D8j1xw9X558Wl18vPp-fXS5cpSqxUEsrtHfgtNSsU60QwrtGWKlrqTxu2rWNUrbRmvtWtwxkp7ytq64TWirRVAfkeOvFkW9nyMUMITvoexthnLMRYskY00IqRN_-g67HOUWczoiKL5eqYeKRerelHP5BTtCZKYXBpo3hzDwGZjAw8ycwZN88Ged2AP-XfE4IgdMtcBd62PzfZN5_ud4qHwAaY6SN</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2317759025</pqid></control><display><type>article</type><title>A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection</title><source>Wiley Journals</source><creator>Hou, Ying ; Bao, Mei‐Ling ; Wu, Chen‐Jiang ; Zhang, Jing ; Zhang, Yu‐Dong ; Shi, Hai‐Bin</creator><creatorcontrib>Hou, Ying ; Bao, Mei‐Ling ; Wu, Chen‐Jiang ; Zhang, Jing ; Zhang, Yu‐Dong ; Shi, Hai‐Bin</creatorcontrib><description>Objectives
To develop a machine learning (ML)‐assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings.
Patients and Methods
In all, 248 patients treated with radical prostatectomy and ePLND or PLND were included. ML‐assisted models were developed from 18 integrated features using logistic regression (LR), support vector machine (SVM), and random forests (RFs). The models were compared to the Memorial SloanKettering Cancer Center (MSKCC) nomogram using receiver operating characteristic‐derived area under the curve (AUC) calibration plots and decision curve analysis (DCA).
Results
A total of 59/248 (23.8%) lymph node invasions (LNIs) were identified at surgery. The predictive accuracy of the ML‐based models, with (+) or without (−) MRI‐reported LNI, yielded similar AUCs (RFs+/RFs−: 0.906/0.885; SVM+/SVM−: 0.891/0.868; LR+/LR−: 0.886/0.882) and were higher than the MSKCC nomogram (0.816; P < 0.001). The calibration of the MSKCC nomogram tended to underestimate LNI risk across the entire range of predicted probabilities compared to the ML‐assisted models. The DCA showed that the ML‐assisted models significantly improved risk prediction at a risk threshold of ≤80% compared to the MSKCC nomogram. If ePLNDs missed was controlled at <3%, both RFs+ and RFs− resulted in a higher positive predictive value (51.4%/49.6% vs 40.3%), similar negative predictive value (97.2%/97.8% vs 97.2%), and higher number of ePLNDs spared (56.9%/54.4% vs 43.9%) compared to the MSKCC nomogram.
Conclusions
Our ML‐based model, with a 5–15% cutoff, is superior to the MSKCC nomogram, sparing ≥50% of ePLNDs with a risk of missing <3% of LNIs.</description><identifier>ISSN: 1464-4096</identifier><identifier>EISSN: 1464-410X</identifier><identifier>DOI: 10.1111/bju.14892</identifier><identifier>PMID: 31392808</identifier><language>eng</language><publisher>England: Wiley Subscription Services, Inc</publisher><subject>Artificial intelligence ; Biopsy ; Cancer surgery ; Learning algorithms ; logistic regression ; Lymph nodes ; Lymphatic system ; Machine learning ; Magnetic resonance imaging ; NMR ; Nomograms ; Nuclear magnetic resonance ; Patients ; PCSM ; pelvic lymph node invasion ; Prostate cancer ; ProstateCancer ; Prostatectomy ; random forests ; support vector machine ; Surgery ; Urological surgery</subject><ispartof>BJU international, 2019-12, Vol.124 (6), p.972-983</ispartof><rights>2019 The Authors BJU International © 2019 BJU International Published by John Wiley & Sons Ltd</rights><rights>2019 The Authors BJU International © 2019 BJU International Published by John Wiley & Sons Ltd.</rights><rights>BJUI © 2019 BJU International</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3532-57a26dcec6460f5b222dc92a46845d313fb955a9661db6b0e4f5da83ff2645293</citedby><cites>FETCH-LOGICAL-c3532-57a26dcec6460f5b222dc92a46845d313fb955a9661db6b0e4f5da83ff2645293</cites><orcidid>0000-0002-2811-7513</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1111%2Fbju.14892$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1111%2Fbju.14892$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>314,780,784,1417,27924,27925,45574,45575</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/31392808$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Hou, Ying</creatorcontrib><creatorcontrib>Bao, Mei‐Ling</creatorcontrib><creatorcontrib>Wu, Chen‐Jiang</creatorcontrib><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Zhang, Yu‐Dong</creatorcontrib><creatorcontrib>Shi, Hai‐Bin</creatorcontrib><title>A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection</title><title>BJU international</title><addtitle>BJU Int</addtitle><description>Objectives
To develop a machine learning (ML)‐assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings.
Patients and Methods
In all, 248 patients treated with radical prostatectomy and ePLND or PLND were included. ML‐assisted models were developed from 18 integrated features using logistic regression (LR), support vector machine (SVM), and random forests (RFs). The models were compared to the Memorial SloanKettering Cancer Center (MSKCC) nomogram using receiver operating characteristic‐derived area under the curve (AUC) calibration plots and decision curve analysis (DCA).
Results
A total of 59/248 (23.8%) lymph node invasions (LNIs) were identified at surgery. The predictive accuracy of the ML‐based models, with (+) or without (−) MRI‐reported LNI, yielded similar AUCs (RFs+/RFs−: 0.906/0.885; SVM+/SVM−: 0.891/0.868; LR+/LR−: 0.886/0.882) and were higher than the MSKCC nomogram (0.816; P < 0.001). The calibration of the MSKCC nomogram tended to underestimate LNI risk across the entire range of predicted probabilities compared to the ML‐assisted models. The DCA showed that the ML‐assisted models significantly improved risk prediction at a risk threshold of ≤80% compared to the MSKCC nomogram. If ePLNDs missed was controlled at <3%, both RFs+ and RFs− resulted in a higher positive predictive value (51.4%/49.6% vs 40.3%), similar negative predictive value (97.2%/97.8% vs 97.2%), and higher number of ePLNDs spared (56.9%/54.4% vs 43.9%) compared to the MSKCC nomogram.
Conclusions
Our ML‐based model, with a 5–15% cutoff, is superior to the MSKCC nomogram, sparing ≥50% of ePLNDs with a risk of missing <3% of LNIs.</description><subject>Artificial intelligence</subject><subject>Biopsy</subject><subject>Cancer surgery</subject><subject>Learning algorithms</subject><subject>logistic regression</subject><subject>Lymph nodes</subject><subject>Lymphatic system</subject><subject>Machine learning</subject><subject>Magnetic resonance imaging</subject><subject>NMR</subject><subject>Nomograms</subject><subject>Nuclear magnetic resonance</subject><subject>Patients</subject><subject>PCSM</subject><subject>pelvic lymph node invasion</subject><subject>Prostate cancer</subject><subject>ProstateCancer</subject><subject>Prostatectomy</subject><subject>random forests</subject><subject>support vector machine</subject><subject>Surgery</subject><subject>Urological surgery</subject><issn>1464-4096</issn><issn>1464-410X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNp1kcFOFTEUhhuiAQQWvoBp4gYXF9pO2zuzRKKoIXEDCbum057x9mamM7Qd4e58BNc-nk_CwQsuTOymJydfvp7Tn5DXnJ1wPKftej7hsm7EDtnnUsuF5OzmxXPNGr1HXuW8ZgwbWu2SvYpXjahZvU9-ndHBulWIQHuwKYb47fePnzbnkAt46sGFHMaIvTxP05gKHUYPPS0jbaEUSDR4iCV0GzrZErDM9C6UFZ3SmIstQJ2NDrEEt3NIqKc2UrgvED36J-i_B0f7zTCtaEQz9SFncAXfPCQvO9tnOHq6D8j1xw9X558Wl18vPp-fXS5cpSqxUEsrtHfgtNSsU60QwrtGWKlrqTxu2rWNUrbRmvtWtwxkp7ytq64TWirRVAfkeOvFkW9nyMUMITvoexthnLMRYskY00IqRN_-g67HOUWczoiKL5eqYeKRerelHP5BTtCZKYXBpo3hzDwGZjAw8ycwZN88Ged2AP-XfE4IgdMtcBd62PzfZN5_ud4qHwAaY6SN</recordid><startdate>201912</startdate><enddate>201912</enddate><creator>Hou, Ying</creator><creator>Bao, Mei‐Ling</creator><creator>Wu, Chen‐Jiang</creator><creator>Zhang, Jing</creator><creator>Zhang, Yu‐Dong</creator><creator>Shi, Hai‐Bin</creator><general>Wiley Subscription Services, Inc</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7QP</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-2811-7513</orcidid></search><sort><creationdate>201912</creationdate><title>A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection</title><author>Hou, Ying ; Bao, Mei‐Ling ; Wu, Chen‐Jiang ; Zhang, Jing ; Zhang, Yu‐Dong ; Shi, Hai‐Bin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3532-57a26dcec6460f5b222dc92a46845d313fb955a9661db6b0e4f5da83ff2645293</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Artificial intelligence</topic><topic>Biopsy</topic><topic>Cancer surgery</topic><topic>Learning algorithms</topic><topic>logistic regression</topic><topic>Lymph nodes</topic><topic>Lymphatic system</topic><topic>Machine learning</topic><topic>Magnetic resonance imaging</topic><topic>NMR</topic><topic>Nomograms</topic><topic>Nuclear magnetic resonance</topic><topic>Patients</topic><topic>PCSM</topic><topic>pelvic lymph node invasion</topic><topic>Prostate cancer</topic><topic>ProstateCancer</topic><topic>Prostatectomy</topic><topic>random forests</topic><topic>support vector machine</topic><topic>Surgery</topic><topic>Urological surgery</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Hou, Ying</creatorcontrib><creatorcontrib>Bao, Mei‐Ling</creatorcontrib><creatorcontrib>Wu, Chen‐Jiang</creatorcontrib><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Zhang, Yu‐Dong</creatorcontrib><creatorcontrib>Shi, Hai‐Bin</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>Calcium & Calcified Tissue Abstracts</collection><collection>MEDLINE - Academic</collection><jtitle>BJU international</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Hou, Ying</au><au>Bao, Mei‐Ling</au><au>Wu, Chen‐Jiang</au><au>Zhang, Jing</au><au>Zhang, Yu‐Dong</au><au>Shi, Hai‐Bin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection</atitle><jtitle>BJU international</jtitle><addtitle>BJU Int</addtitle><date>2019-12</date><risdate>2019</risdate><volume>124</volume><issue>6</issue><spage>972</spage><epage>983</epage><pages>972-983</pages><issn>1464-4096</issn><eissn>1464-410X</eissn><abstract>Objectives
To develop a machine learning (ML)‐assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings.
Patients and Methods
In all, 248 patients treated with radical prostatectomy and ePLND or PLND were included. ML‐assisted models were developed from 18 integrated features using logistic regression (LR), support vector machine (SVM), and random forests (RFs). The models were compared to the Memorial SloanKettering Cancer Center (MSKCC) nomogram using receiver operating characteristic‐derived area under the curve (AUC) calibration plots and decision curve analysis (DCA).
Results
A total of 59/248 (23.8%) lymph node invasions (LNIs) were identified at surgery. The predictive accuracy of the ML‐based models, with (+) or without (−) MRI‐reported LNI, yielded similar AUCs (RFs+/RFs−: 0.906/0.885; SVM+/SVM−: 0.891/0.868; LR+/LR−: 0.886/0.882) and were higher than the MSKCC nomogram (0.816; P < 0.001). The calibration of the MSKCC nomogram tended to underestimate LNI risk across the entire range of predicted probabilities compared to the ML‐assisted models. The DCA showed that the ML‐assisted models significantly improved risk prediction at a risk threshold of ≤80% compared to the MSKCC nomogram. If ePLNDs missed was controlled at <3%, both RFs+ and RFs− resulted in a higher positive predictive value (51.4%/49.6% vs 40.3%), similar negative predictive value (97.2%/97.8% vs 97.2%), and higher number of ePLNDs spared (56.9%/54.4% vs 43.9%) compared to the MSKCC nomogram.
Conclusions
Our ML‐based model, with a 5–15% cutoff, is superior to the MSKCC nomogram, sparing ≥50% of ePLNDs with a risk of missing <3% of LNIs.</abstract><cop>England</cop><pub>Wiley Subscription Services, Inc</pub><pmid>31392808</pmid><doi>10.1111/bju.14892</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0002-2811-7513</orcidid></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1464-4096 |
| ispartof | BJU international, 2019-12, Vol.124 (6), p.972-983 |
| issn | 1464-4096 1464-410X |
| language | eng |
| recordid | cdi_proquest_miscellaneous_2270006245 |
| source | Wiley Journals |
| subjects | Artificial intelligence Biopsy Cancer surgery Learning algorithms logistic regression Lymph nodes Lymphatic system Machine learning Magnetic resonance imaging NMR Nomograms Nuclear magnetic resonance Patients PCSM pelvic lymph node invasion Prostate cancer ProstateCancer Prostatectomy random forests support vector machine Surgery Urological surgery |
| title | A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T23%3A37%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20machine%20learning%E2%80%90assisted%20decision%E2%80%90support%20model%20to%20better%20identify%20patients%20with%20prostate%20cancer%20requiring%20an%20extended%20pelvic%20lymph%20node%20dissection&rft.jtitle=BJU%20international&rft.au=Hou,%20Ying&rft.date=2019-12&rft.volume=124&rft.issue=6&rft.spage=972&rft.epage=983&rft.pages=972-983&rft.issn=1464-4096&rft.eissn=1464-410X&rft_id=info:doi/10.1111/bju.14892&rft_dat=%3Cproquest_cross%3E2317759025%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2317759025&rft_id=info:pmid/31392808&rfr_iscdi=true |