Universität zu Köln

Jung, Jin-On ORCID: 0000-0001-6121-0114, Crnovrsanin, Nerma, Wirsik, Naita Maren, Nienhuser, Henrik, Peters, Leila, Popp, Felix, Schulze, Andre, Wagner, Martin ORCID: 0000-0002-9831-9110, Muller-Stich, Beat Peter, Buchler, Markus Wolfgang and Schmidt, Thomas (2023). Machine learning for optimized individual survival prediction in resectable upper gastrointestinal cancer. J. Cancer Res. Clin. Oncol., 149 (5). S. 1691 - 1703. NEW YORK: SPRINGER. ISSN 1432-1335

Full text not available from this repository.

Abstract

Purpose Surgical oncologists are frequently confronted with the question of expected long-term prognosis. The aim of this study was to apply machine learning algorithms to optimize survival prediction after oncological resection of gastroesophageal cancers. Methods Eligible patients underwent oncological resection of gastric or distal esophageal cancer between 2001 and 2020 at Heidelberg University Hospital, Department of General Surgery. Machine learning methods such as multi-task logistic regression and survival forests were compared with usual algorithms to establish an individual estimation. Results The study included 117 variables with a total of 1360 patients. The overall missingness was 1.3%. Out of eight machine learning algorithms, the random survival forest (RSF) performed best with a concordance index of 0.736 and an integrated Brier score of 0.166. The RSF demonstrated a mean area under the curve (AUC) of 0.814 over a time period of 10 years after diagnosis. The most important long-term outcome predictor was lymph node ratio with a mean AUC of 0.730. A numeric risk score was calculated by the RSF for each patient and three risk groups were defined accordingly. Median survival time was 18.8 months in the high-risk group, 44.6 months in the medium-risk group and above 10 years in the low-risk group. Conclusion The results of this study suggest that RSF is most appropriate to accurately answer the question of long-term prognosis. Furthermore, we could establish a compact risk score model with 20 input parameters and thus provide a clinical tool to improve prediction of oncological outcome after upper gastrointestinal surgery.

Item Type:	Journal Article
Creators:	Creators Email ORCID ORCID Put Code Jung, Jin-On UNSPECIFIED orcid.org/0000-0001-6121-0114 UNSPECIFIED Crnovrsanin, Nerma UNSPECIFIED UNSPECIFIED UNSPECIFIED Wirsik, Naita Maren UNSPECIFIED UNSPECIFIED UNSPECIFIED Nienhuser, Henrik UNSPECIFIED UNSPECIFIED UNSPECIFIED Peters, Leila UNSPECIFIED UNSPECIFIED UNSPECIFIED Popp, Felix UNSPECIFIED UNSPECIFIED UNSPECIFIED Schulze, Andre UNSPECIFIED UNSPECIFIED UNSPECIFIED Wagner, Martin UNSPECIFIED orcid.org/0000-0002-9831-9110 UNSPECIFIED Muller-Stich, Beat Peter UNSPECIFIED UNSPECIFIED UNSPECIFIED Buchler, Markus Wolfgang UNSPECIFIED UNSPECIFIED UNSPECIFIED Schmidt, Thomas UNSPECIFIED UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-664851
DOI:	10.1007/s00432-022-04063-5
Journal or Publication Title:	J. Cancer Res. Clin. Oncol.
Volume:	149
Number:	5
Page Range:	S. 1691 - 1703
Date:	2023
Publisher:	SPRINGER
Place of Publication:	NEW YORK
ISSN:	1432-1335
Language:	English
Faculty:	Unspecified
Divisions:	Unspecified
Subjects:	no entry
Uncontrolled Keywords:	Keywords Language SELECTION; MODELS; TRENDS Multiple languages Oncology Multiple languages
URI:	http://kups.ub.uni-koeln.de/id/eprint/66485