Proposal for a fuzzy logic-based system to determine cardiovascular risk

Abstract

One of the key variables to determine the level of cardiovascular risk is the heart rate variability, which associates different metrics such as average of the RR intervals (average RR), standard deviation of the RR intervals (SDRR) and percentage of differences greater than 50 ms in RR intervals (pRR50). Given that these metrics make use of different measurement units, scales, and ranges, it is necessary to determine an output risk level in intelligible terms, taking as input the values of each one of them. Thus, this article proposes the development of a system based on fuzzy logic to determine the percentage or cardiovascular risk level. The fuzzy system is connected to an Arduino board with a heart rate sensor where the heart rate and heart rate variability values are obtained, so they are used to calculate the risk level metrics. Using the input values of each metric, as well as the 3 membership functions of the inputs, the output membership function, and a total of 18 inference rules defined from the inputs and outputs, the system obtains the output cardiovascular risk level. The fuzzy system was implemented using free hardware and software tools, making it available in medical campaigns for the early identification of heart conditions.