Robust Metabolic Syndrome Risk Index Based on Triangular Areal Similarity

Publication Date


Document Type

Conference Proceeding

Publication Title

Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021



First Page


Last Page



Metabolic syndrome is a cluster of risk factors associated with certain cardiovascular diseases. It is a chronic disease and has become an important public health issue in many countries, as it is steadily increasing owing to urbanized diets and lifestyles. The current medical diagnostic methods focus on the diagnosis of metabolic syndrome itself; however, they are not suitable for measuring its progression. Several methods have been proposed for measuring the risk of metabolic syndrome aiming to solve this problem, including the 'Areal Similarity Degree' method. In this study, a robust risk index based on radar charts is proposed to overcome the weaknesses of the Areal Similarity Degree method. In the proposed method, the examination values are expressed as two types of triangles on 10 types of three-axis radar charts, and each area of the triangle is synthesized and used as a risk index. The performance of the proposed risk index is evaluated based on the data of 31,070 Korean adults obtained from the Korea Genome and Epidemiology Study health examinee (KoGES HEXA) dataset. The proposed risk index shows a high correlation with current medical diagnostic criteria, and a good diagnostic performance. It produces more stable and generalized risk index values than those from the basic Areal Similarity Degree method. In addition, its sample-independent characteristics are confirmed.

Funding Sponsor

Ministry of Science, ICT and Future Planning


diagnostic features, metabolic syndrome, risk index, sample-independent method


Applied Data Science; Electrical Engineering