An Investigational Approach for the Prediction of Gastric Cancer Using Artificial Intelligence Techniques: A Systematic Review

Gastric cancer is characterized by the growth of cancerous cells within the lining of the stomach. Traditionally, this condition has been challenging to diagnose. However, today artificial intelligence (AI) is becoming more widely used across healthcare sectors because it offers significant improvements in the speed and accuracy of data discovery, extraction, and personalized recommendations for treatments. At present, AI research on the identification and treatment of gastric malignant growth, helicobacter pylori bacteria, and throat disease is advancing; the connections between these sub-fields and those of gastric tumors and oesophageal diseases indicates that AI can also be utilized for diagnosis in these areas. PRISMA standards were used to identify publications published between 2009 and 2021 on Web of Science, EBSCO, and EMBASE. This study conducted an efficient search and included research publications that used AI-based learning algorithms for gastric cancer prediction. A total of 110 studies are regarded as important for gastric cancer prediction using traditional machine and deep learning-based classifications. In this work, we offer a survey of the work currently performed on AI-enabled diagnosis at different stages of gastric cancers. We also outline the symptoms of gastric cancer, the various means of diagnosis of gastric cancers, and the roles played by conventional machine learning (ML) and the ML subset of deep learning (DL) models for early detection of gastric cancer. Furthermore, we summarize the work done by different researchers in AI techniques for early prediction of gastric cancers and compare their work by using parameters such as prediction rate, accuracy, sensitivity, specificity, the area under the curve, and F1-score.