ASO Author Reflections: Use of the Survival Recurrent Network for Prediction of Overall Survival in Patients with Gastric Cancer

Excerpt

It is important to stratify high-risk patients who are diagnosed with gastric cancer, a common cause of cancer mortality in East Asia. There are several ways to predict this prognosis, such as TNM staging and nomogram. However, the pathophysiologic process in the human body is not simple but chaotic, and the mechanism cannot be fully reflected through the conventional statistical method.1 To overcome this limitation of the classical linear analyzing method, artificial neural networks (ANNs) have been developed and introduced to medicine to solve this problem with complex data of patients.2 The ANN is formed with an extensive network of nodes, similar to human brain neurons. Because of this feature, the ANN is superior for analyzing many variables at the same time, as opposed to general statistical methods, which test only a few statistically significant variables. In this research, we developed a stronger survival prediction model using artificial intelligence. …