Machine learning in predicting gastric cancer survival Presenting a novel decision support system model

Main Article Content

Ezgi Altinsoy
Batuhan Bakirarar
Serdar Culcu

Abstract

BACKGROUND: Gastric cancer is the 4th most frequent cause of cancer-related deaths, with a 5-year survival rate of less than 40%. In recent years, many artificial intelligence applications have been used in the field of gastric cancer through their effective computing and learning ability. In this study, we aim to develop a software that can accurately detect overall survival in gastric cancer cases with the help of artificial intelligence and machine learning.


METHODS: The study included 34417 patients’ data diagnosed with gastric cancer between 2010 and 2015. The main hypothesis in the study was overall survival (OS) in years, defined from the date of diagnosis to the date of death or, for living patients, the last control date. In addition to survival, other variables selected for the analyzes were age at diagnosis, race, gender, behavior, primary site, grade, histology, T stage, N stage, M stage and size of the tumor, vital status, and follow-up time (months).


RESULTS: The median overall survival of the patients was found to be 15.00±0.20 years. Median life expectancy was found to be 21.00±0.85 years for those younger than 50 years of age, 20.00±0.43 years for those aged 50-69 years, and 10.00±0.22 years for those aged 70 and over. Especially artificial intelligence techniques such as machine learning and deep learning lead to remarkable developments in the field of gastric cancer.


CONCLUSION: With the ability to compute and learn we think that use of artificial intelligence will be revolutionary in gastric cancer in terms of diagnosis and prognosis.

Article Details

How to Cite
Altinsoy, Ezgi, et al. “Machine Learning in Predicting Gastric Cancer Survival Presenting a Novel Decision Support System Model”. Annali Italiani Di Chirurgia, vol. 94, no. 6, Nov. 2023, pp. 631-8, https://annaliitalianidichirurgia.it/index.php/aic/article/view/3160.
Section
Article