Predictive Modeling of Blood Pressure Progression

There is ample evidence that Blood Pressure (BP) management can benefit from adequate lifestyle changes. To implement personalized strategies, we need solutions to predict how a person's BP will evolve. Regression-based BP prediction models tend to be not sufficient to accurately forecast future values due to the intrinsic complexity and fluctuation of BP data, particularly when no context information is available. This study investigates how categorization approaches can be used to forecast trends and levels of blood pressure over a given time window. Our results highlight the potential of classification-based algorithms in the forecasting of BP levels, opening the door to more proactive and individualized healthcare interventions in BP management.