The increasing performance and availability of embedded systems increases their attractiveness for biomedical applications. With advances in sensor processing and classification algorithms, real-time decision support in patient monitoring becomes feasible. However, the gap between algorithm design and their embedded realization is growing. This paper overviews an approach for development of biomedical devices at an abstract algorithm level with automatic generation of an embedded implementation. Based on a case study of a Brain Computer Interface (BCI), this paper demonstrates capturing, modeling and synthesis of such applications.

Schirner_HLDVT11.pdf