The aim of this work was to develop a proprietary device for real-time cough detection in audio recordings. Using models such as CNN, ResNet-50, and MobileNet, the system classifies cough sounds, enabling the early detection of potential infection cases, such as COVID-19. The device is intended for use in both private and public spaces, including medical facilities, nursing homes, and doctor's offices.

The aim of the research was to develop a new device for monitoring and diagnosing functional disorders of the lower urinary tract, which will enable measurement of muscle tension (EMG) and electrical impedance tomography (EIT) with the possibility of electrostimulation of bladder reconstruction.

Minimal hardware requirements for electrical impedance tomography can enhance the method's applicability for medical tracking and diagnostic tasks in e-health. The principle of electrical tomography is that the response to electrical excitation of biological tissues provides information about the material structure. In order to reduce the required hardware, we study pseudo random binary sequence excitation patterns, instead of the standard sinusoidal excitation. We implement the measurement system in reconfigurable hardware with a mixed signal SoC. The measurements are validated using system identification in the resulting data to estimate the discrete transfer function of the system under measurement.