Medical signal processing methods based on
compressed sensing; applications and their implementation

PN II -

RU - TE

Romanian Academy

Institute of Computer Science Iasi

The objectives of the project aim at continuing and developing the current preoccupations of the team members and consist of analyzing the compressed sensing possibilities and processing of the biomedical signals and implementation of the methods that prove to be the best after computer simulations. The applications shall be interdisciplinary, combining the field of engineering (compressed sensing, signal processing, etc) with the medical field, aiming to find the best solutions for medical signals processing, especially the ECG and EEG signals.

Specific objectives are:

·           Design and implementation of optimization algorithms for the projection matrix depending on the signal characteristics or the used dictionaries in reconstruction

·           Build specific dictionaries depending on pathologies or ECG / EEG tracked characteristics

·           Implement compression and classification in compressed space on embedded systems devices

The final aim of the project is the implementation of compression and classification in the compressed space for ECG and EEG on hardware devices. Specific dictionaries will be built, depending on the groups of pathologies or the particularities of the ECG/EEG signals.