Spatiotemporal Hotspots of Study Areas in Research of Gastric Cancer in China Based on Web-Crawled Literature.

Zhen Wang,Hongyan Ren,Dafang Zhuang,An Zhang

doi:10.3390/ijerph18083997

Abstract

Gastric cancer is a common malignancy worldwide and poses a serious threat to human public health. The difficulty in obtaining epidemiological data limits the development of cross-disciplinary related research. In this study, 99,364 publications on gastric cancer from 1991 to 2019 were obtained using web-crawler technology, and a technical framework for extracting toponyms from these publications was constructed to analyze spatiotemporal hotspots of study areas in gastric cancer research in China. The results showed the following: (1) The accuracy of toponym extraction was greatly improved after eliminating the systematic exclusion words and adding historical toponyms, with a precision of 95.31% and a recall of 94.86%. (2) Gastric cancer research (GCR) and gastric cancer research with toponyms (GCRWT) are attracting increasing amounts of attention. The amount of GCR results published in Chinese and English is gradually leveling off, and the imbalance between those of GCRWT is gradually widening. (3) The spatial distribution of gastric cancer research in China is uneven, and the hotspots are mainly located in the eastern coastal areas. There were huge advances in gastric cancer research at the province/city/county scale in Eastern China, while the central region has only increased research at the county scale. We suggest that gastric cancer research should pay more attention to the central region, which has the highest gastric cancer incidence/mortality. This study provides important clues for research on and investigations of gastric cancer.

Highlights

Gastric cancer (GC) is a common malignancy worldwide, and it caused about 0.8 million deaths in 2018 [1]
In order to test the accuracy of the toponym extraction, in this study, the recall and precision rate of the toponym extraction were compared with the geographical named entity recognition (GNER) and GNERPP
In this study, the extraction accuracy of the toponyms was relatively high after the elimination of the systematically excluded words and the addition of historical toponyms, and it met the requirements of the subsequent spatiotemporal analysis

Summary

Introduction

Gastric cancer (GC) is a common malignancy worldwide, and it caused about 0.8 million deaths in 2018 [1]. China is one of the countries with the highest GC incidence and mortality, accounting for about 50% of new cases and deaths in the world, making it a major threat to the life and health of the Chinese people [2,3]. Risk factors, and prevention and control strategies for gastric cancer based on surveillance data or statistical data is the basis and an important means of preventing and controlling gastric cancer [4,5]. China has had a sound national tumor registration system and a Disease Surveillance Points system (DSP) for decades, and the country has accumulated long-term tumor incidence and cause-of-death registration data [6,7]. The strategy for obtaining epidemiological data from the literature is a compromise for solving this problem, in which a meta-analysis based on published For non-health research institutions in China, it is still difficult to obtain these epidemiological data, which to some extent affects the development of cross-disciplinary research (e.g., health and environment) and limits the possibility of disease prevention and control work being supported by theories and methods from more disciplines.

Objectives

Methods

Results

Discussion

Conclusion