A comparison of artificial intelligence algorithms in diagnosing and predicting gastric cancer: a review study

Abstract

Today, artificial intelligence is considered a powerful tool that can help physicians identify and diagnose and predict diseases. Gastric cancer has been the fourth most common malignancy and the second leading cause of cancer mortality in the world. Thus, timely diagnosis of this type of cancer could effectively control it. This paper compares AI (artificial intelligence) algorithms in diagnosing and predicting gastric cancer based on types of AI algorithms, sample size, accuracy, sensitivity, and specificity. This narrative-review paper aims to explore AI algorithms in diagnosing and predicting gastric cancer. To achieve this goal, we reviewed English articles published between 2011 and 2021 in PubMed and Science direct databases. According to the reviews conducted on the published papers, the endoscopic method has been the most used method to collect and incorporate samples into designed models. Also, the SVM (support vector machine), convolutional neural network (CNN), and deep-type CNN have been used the most; therefore, we propose the usage of these algorithms in medical subjects, especially in gastric cancer.

Keywords

• Artificial Intelligence
• Neural Networks, Computer
• Stomach Neoplasms
• Support Vector Machine