Biofeedback: e-health prediction based on evolving fuzzy neural network and wearable technologies

Abstract

AbstractRecent advances in wearable microelectronics and new neural networks paradigms, capable to evolve and learn online such as the Evolving Fuzzy Neural Network (EFuNN), enable the deploy of biofeedback-based applications. The missed physiologic response could be recovered by measuring uninvasively the vital signs such as the heart rate, the bio impedance, the body temperature, the motion activity, the blood pressure, the blood oxygenation and the respiration rate. Then, the prediction could be performed applying the evolving ANN paradigms. The simulation of a wearable biofeedback system has been executed applying the Evolving Fuzzy Neural Network (EFuNN) paradigm for prediction. An highly integrated wearable microelectronic device for uninvasively vital signs measurement has been deployed. Simulation results demonstrate that biofeedback control model could be an effective reference design that enables short and long-term e-health prediction. The biofeedback framework was been then defined.