Determining of prognostic factors in gastric cancer patients using artificial neural networks

Abstract

Background & Objectives: The aim of this study is to determine diagnostic factors for Iranian gastric cancer patients and their importance using artificial neural network and Weibull regression models. Methods: This study was a historical cohort study with data gathered from 436 registered gastric cancer patients who underwent surgery between 2002 and 2007 at the Taleghani Hospital (a referral center for gastrointestinal cancers), Tehran, Iran. In order to determine risk factors and their importance, neural network and Weibull regression models were used. Results: The Weibull regression analysis showed that lymph node metastasis and histopathology of tumor were selected as important variables. Based on the neural network model, staging, lymph node metastasis, histopathology of tumor, metastasis, and age at diagnosis were selected as important variables. The true prediction of neural network was 82.6%, and for the Weibull regression model, 75.7%. Conclusion: The present study showed that the neural network model is a more powerful tool in determining the important variables for gastric cancer patients compared to Weibull regression model. Therefore, this model is recommended for determining of risk factors of such patients.