Choosing Data Clustering Tools For GIS-Based Visualization Of Disease Incidence In The Population

Abstract

Objective — To substantiate the choice of optimal tools for clustering spatially referenced data on disease incidence for GIS-based analysis of their spatial distribution. Material and Methods — We used primary data on the incidence of malignant neoplasms, chronic alcoholism, and asthma in the population of eight administrative areas in Arkhangelsk Oblast as a constituent entity of the Arctic Zone of the Russian Federation. Disease incidence was averaged over a 5-year period from 2016 to 2020. We assessed the methods for visualizing the distribution of spatially referenced indicators using the ArcMap geoinformation system tools. Results — The study yielded differences in the outcomes of automated clustering of spatially referenced data in ArcMap, depending on the normality of the distribution in individual samples and the spread of indicator values, which was visually reflected on the resulting map. The parameter values in the samples directly affected the features of data clustering. Hence, this issue is important to consider for ensuring the correct choice of the appropriate analytical tool. Conclusion — Our study demonstrated that when using tools for automated clustering of spatially referenced incidence data in terms of their visualization in ArcGIS, it is necessary to consider the factors that directly affect the accuracy of their presentation. We consider it most appropriate to use a clustering tool based on the geometric interval method.