Verification of metabolic syndrome checkup data with a self-organizing map (SOM): Towards a simple judging tool

Abstract

Minor cases of the metabolic syndrome (MS) for which the lipid, blood pressure, and blood glucose levels are at the border between "normal" and "abnormal" require careful monitoring. We devised a method that addresses this issue by first introducing a "non-ill" condition between the former two. Based on observations of the MS indicator distribution of all examinees, the checkup data was then labeled as "normal", "non-ill" and "abnormal" and applied to a Self-Organizing Map (SOM) whose aim was to visualize the MS indicator distribution in relation to the 3 patient conditions mentioned. Our method was then validated by comparing the MS judgment results with those obtained using the conventional method. The ability to visualize with our method the positional relations between the MS indicators and the 3 conditions further adds to its usability as a health guidance tool. © 2013 Springer-Verlag.