ABSTRACT
E-health system applications are becoming more widely used with improvements in many fields of healthcare such as remote patient monitoring, interactive patient care and prediction and detection of specific health conditions. This chapter discusses decisions, challenges and solutions relating to implementation of an E-health data acquisition system. A single-board computer system for the acquisition of biomedical signals from equipment including blood pressure sensors, pulse oximeter, airflow sensor, galvanic skin response sensor, temperature sensor, EEG helmet, etc. is also presented. The data acquired can be transmitted remotely, stored locally or in the cloud, processed and analyzed using shallow and deep neural networks. The chapter describes an automatic recognition system for motor imagery movement activities based on EEG signals using convolutional neural networks, discussing factors that affect the recognition rate of some activities and comparing the efficiency of the architecture of four different convolutional neural networks.
