Development of a Self-Constructing Neuro-Fuzzy Inference System for Online Classification of Physical Movements

This work is a part of the project "context- aware cardiac long-term monitoring (CALM)" of the Institute for information processing technology of the University of Karlsruhe. The aim of our research is the development of a system to help on the prevention of cardiovascular illnesses by continuous telemonitoring of patients. Therefore, the system should enable patient-friendly measurements of blood- pressure and electrocardiogram (ECG). Furthermore, it should allow an objective evaluation of the patient\´s physiological parameters taking into consideration context information. When evaluating cardiovascular parameters, the most important context information is physical activity. For obtaining the activity or movements of a patient, the authors propose a system based on acceleration sensors and adaptive neuro- fuzzy inference system (ANFIS) for pattern classification. Using this method, an online self-constructing activity recognition system has been developed and its results are shown.