Skip to main content

01.03.2018 | Gastrointestinal Oncology | Ausgabe 5/2018

Annals of Surgical Oncology 5/2018

Prediction of Overall Survival and Novel Classification of Patients with Gastric Cancer Using the Survival Recurrent Network

Annals of Surgical Oncology > Ausgabe 5/2018
MD Sung Eun Oh, MD, PhD Sung Wook Seo, MD Min-Gew Choi, MD Tae Sung Sohn, MD Jae Moon Bae, MD Sung Kim
Wichtige Hinweise

Electronic supplementary material

The online version of this article (https://​doi.​org/​10.​1245/​s10434-018-6343-7) contains supplementary material, which is available to authorized users.
Sung Eun Oh and Sung Wook Seo contributed equally to this work.



Artificial neural networks (ANNs) have been applied to many prediction and classification problems, and could also be used to develop a prediction model of survival outcomes for cancer patients.


The aim of this study is to develop a prediction model of survival outcomes for patients with gastric cancer using an ANN.


This study enrolled 1243 patients with stage IIA–IV gastric cancer who underwent D2 gastrectomy from January 2007 to June 2010. We used a recurrent neural network (RNN) to make the survival recurrent network (SRN), and patients were randomly sorted into a training set (80%) and a test set (20%). Fivefold cross-validation was performed with the training set, and the optimized model was evaluated with the test set. Receiver operating characteristic (ROC) curves and area under the curves (AUCs) were evaluated, and we compared the survival curves of the American Joint Committee on Cancer (AJCC) 8th stage groups with those of the groups classified by the SRN-predicted survival probability.


The test data showed that the ROC AUC of the SRN was 0.81 at the fifth year. The SRN-predicted survival corresponded closely with the actual survival in the calibration curve, and the survival outcome could be more discriminately classified by using the SRN than by using the AJCC staging system.


SRN was a more powerful tool for predicting the survival rates of gastric cancer patients than conventional TNM staging, and may also provide a more flexible and expandable method when compared with fixed prediction models such as nomograms.

Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten

e.Med Interdisziplinär

Mit e.Med Interdisziplinär erhalten Sie Zugang zu allen CME-Fortbildungen und Fachzeitschriften auf Zusätzlich können Sie eine Zeitschrift Ihrer Wahl in gedruckter Form beziehen – ohne Aufpreis.

Weitere Produktempfehlungen anzeigen
Supplementary material 1 (TIFF 18952 kb)
Supplementary material 2 (DOCX 15 kb)
Über diesen Artikel

Weitere Artikel der Ausgabe 5/2018

Annals of Surgical Oncology 5/2018 Zur Ausgabe
  1. Das kostenlose Testabonnement läuft nach 14 Tagen automatisch und formlos aus. Dieses Abonnement kann nur einmal getestet werden.

  2. Das kostenlose Testabonnement läuft nach 14 Tagen automatisch und formlos aus. Dieses Abonnement kann nur einmal getestet werden.

Neu im Fachgebiet Chirurgie

Mail Icon II Newsletter

Bestellen Sie unseren kostenlosen Newsletter Update Chirurgie und bleiben Sie gut informiert – ganz bequem per eMail.