File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Machine learning models for predicting the outcomes of surgical treatment of colorectal liver metastases

TitleMachine learning models for predicting the outcomes of surgical treatment of colorectal liver metastases
Authors
Issue Date19-Jan-2023
PublisherLippincott, Williams & Wilkins
Citation
Journal of The American College of Surgeons, 2023, v. 236, n. 4, p. 884-893 How to Cite?
Abstract

BACKGROUND: 

Surgical intervention remains the cornerstone of a multidisciplinary approach in the treatment of colorectal liver metastases (CLM). Nevertheless, patient outcomes vary greatly. While predictive tools can assist decision-making and patient counseling, decades of efforts have yet to result in generating a universally adopted tool in clinical practice.

STUDY DESIGN: 

An international collaborative database of CLM patients who underwent surgical therapy between 2000 and 2018 was used to select 1,004 operations for this study. Two different machine learning methods were applied to construct 2 predictive models for recurrence and death, using 128 clinicopathologic variables: gradient-boosted trees (GBTs) and logistic regression with bootstrapping (LRB) in a leave-one-out cross-validation.

RESULTS: 

Median survival after resection was 47.2 months, and disease-free survival was 19.0 months, with a median follow-up of 32.0 months in the cohort. Both models had good predictive power, with GBT demonstrating a superior performance in predicting overall survival (area under the receiver operating curve [AUC] 0.773, 95% CI 0.743 to 0.801 vs LRB: AUC 0.648, 95% CI 0.614 to 0.682) and recurrence (AUC 0.635, 95% CI 0.599 to 0.669 vs LRB: AUC 0.570, 95% CI 0.535 to 0.601). Similarly, better performances were observed predicting 3- and 5-year survival, as well as 3- and 5-year recurrence, with GBT methods generating higher AUCs.

CONCLUSIONS: 

Machine learning provides powerful tools to create predictive models of survival and recurrence after surgery for CLM. The effectiveness of both machine learning models varies, but on most occasions, GBT outperforms LRB. Prospective validation of these models lays the groundwork to adopt them in clinical practice.


Persistent Identifierhttp://hdl.handle.net/10722/331237
ISSN
2023 Impact Factor: 3.8
2023 SCImago Journal Rankings: 1.419
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorMoaven, O-
dc.contributor.authorTavolara, TE-
dc.contributor.authorValenzuela, CD-
dc.contributor.authorCheung, TT-
dc.contributor.authorCorvera, CU-
dc.contributor.authorCha, CH-
dc.contributor.authorStauffer, JA-
dc.contributor.authorNiazi, MKK-
dc.contributor.authorGurcan, MN-
dc.contributor.authorShen, P-
dc.date.accessioned2023-09-21T06:53:57Z-
dc.date.available2023-09-21T06:53:57Z-
dc.date.issued2023-01-19-
dc.identifier.citationJournal of The American College of Surgeons, 2023, v. 236, n. 4, p. 884-893-
dc.identifier.issn1072-7515-
dc.identifier.urihttp://hdl.handle.net/10722/331237-
dc.description.abstract<h3>BACKGROUND: </h3><p>Surgical intervention remains the cornerstone of a multidisciplinary approach in the treatment of colorectal liver metastases (CLM). Nevertheless, patient outcomes vary greatly. While predictive tools can assist decision-making and patient counseling, decades of efforts have yet to result in generating a universally adopted tool in clinical practice.</p><h3>STUDY DESIGN: </h3><p>An international collaborative database of CLM patients who underwent surgical therapy between 2000 and 2018 was used to select 1,004 operations for this study. Two different machine learning methods were applied to construct 2 predictive models for recurrence and death, using 128 clinicopathologic variables: gradient-boosted trees (GBTs) and logistic regression with bootstrapping (LRB) in a leave-one-out cross-validation.</p><h3>RESULTS: </h3><p>Median survival after resection was 47.2 months, and disease-free survival was 19.0 months, with a median follow-up of 32.0 months in the cohort. Both models had good predictive power, with GBT demonstrating a superior performance in predicting overall survival (area under the receiver operating curve [AUC] 0.773, 95% CI 0.743 to 0.801 vs LRB: AUC 0.648, 95% CI 0.614 to 0.682) and recurrence (AUC 0.635, 95% CI 0.599 to 0.669 vs LRB: AUC 0.570, 95% CI 0.535 to 0.601). Similarly, better performances were observed predicting 3- and 5-year survival, as well as 3- and 5-year recurrence, with GBT methods generating higher AUCs.</p><h3>CONCLUSIONS: </h3><p>Machine learning provides powerful tools to create predictive models of survival and recurrence after surgery for CLM. The effectiveness of both machine learning models varies, but on most occasions, GBT outperforms LRB. Prospective validation of these models lays the groundwork to adopt them in clinical practice.</p>-
dc.languageeng-
dc.publisherLippincott, Williams & Wilkins-
dc.relation.ispartofJournal of The American College of Surgeons-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.titleMachine learning models for predicting the outcomes of surgical treatment of colorectal liver metastases-
dc.typeArticle-
dc.identifier.doi10.1097/XCS.0000000000000573-
dc.identifier.scopuseid_2-s2.0-85150311807-
dc.identifier.volume236-
dc.identifier.issue4-
dc.identifier.spage884-
dc.identifier.epage893-
dc.identifier.isiWOS:000960891300031-
dc.identifier.issnl1072-7515-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats