Bibliometrics of gastric cancer prediction models.

Abstract

This paper analyzes the manuscripts in the field of gastric cancer (GC) prediction, guiding clinical work and prevention of GC. Using a search strategy, we retrieved research articles related to GC prognosis from the Web of Science (WOS) core database: topic search (TS) = ((gastric cancer OR stomach cancer) AND (survival rate OR survival analysis OR prognosis) AND (predict model)). We set the language to English, the document type to article and review, and completed the search on July 1, 2023. We obtained 1,598 relevant articles, and two researchers screened the search results again, excluding irrelevant, misclassified, and retracted articles. Any controversial articles were reviewed by a third researcher to make the final decision on the required literature. We finally selected 1,056 articles, excluding 542 articles, and extracted the required data from the WOS database for analysis. The extracted database included: title, publication year, author, country/region, institution, citation count, journal, keyword, and reference. We used R (4.3.0) to load the R package (bibliometrix) for bibliometric analysis. The 1,056 articles came from 273 sources (journals, books, etc.), and 3,661 authors conducted relevant research on GC prognosis models. Frontiers in Oncology published the most articles (N=72), and Gastric Cancer Journal had the most citations (N=1,130). The publication time span ranged from 1991 to 2023, with an average annual growth rate of 13.31%. The number of publications increased from 2017, with a sharp increase from 2020 to 2023. The five countries with the most publications were China (n=826), Japan (n=62), Korea (n=47), USA (n=42), Italy (n=19). China had the most citations (N=9,595), and USA had the highest average citation per article (44.9 times). The most common topic was GC survival (n=236), followed by expression (n=209). Multiple GC prediction models in this study describe the science of predicting GC incidence and prognosis. This work provides the most influential references related to GC prediction and serves as a guide for citable papers.