Dubravka Bojanic

Quantification of dynamic EMG patterns during gait in children with cerebral palsy.

Our goal was to simplify the representation and interpretation of surface electromyographic (EMG) activity during gait to develop a clinical method for evaluating gait disabilities in children with cerebral palsy (CP). EMG was recorded from four muscles of a lower extremity. Gait cycles were tracked from one force-sensing resistor signal that was recorded synchronously with EMG. The method is based on the comparison of a patients dynamic EMG envelope shapes and the normative gait-related patterns (norms). Developed norms were based on EMG data obtained in 10 healthy children. Due to newly introduced techniques for time and amplitude normalization, norms were developed regardless of differences in subject age, gender, basic gait parameters and the EMG measurement process. The proposed gait metric quantifies the similarity between a patients gait-related patterns and norms by a single global value suitable for gait analysis in general, including a detailed analysis using the 10 partial values. The gait metric was experimentally validated with a control group of healthy children and a group of children with CP with different degrees of motor deficits. Gait metric values obtained in children from the control group are high for all muscles, which means that gait-related patterns are close to norms, whereas in children with CP the higher the degree of motor deficit, the lower the gait metric values. The method could be a very useful clinical tool for the recognition and tracking of motor disorders of the lower extremities in children with CP as well as many other neuromotor pathologies.

Dyadic Wavelets for Real-time Heart Rate Monitoring

We developed the new intelligent virtual ECG device by integrating the dyadic wavelet (DyWT) based algorithm for QRS complex detection into the virtual teleECG. The new virtual instrument (VI) was realized by using LabVIEW software. The system allows real-time detection of the heart rhythm, offline analysis of the previously recorded signals or offline analysis when using the system via Internet. The new system allows the physician to locate and recognize life threatening events in ECG recordings and provides the patient with an ECG alarm system. The tests on data from a MIT-BIH database show that the DyWT based detector detects accurately 99.53% of QRS complexes, and similarly better than 99% for the clinical recordings. The analysis in clinical environment showed that in ECG signals comprising a sharp and large T wave the algorithm must be fine tuned, otherwise it could result with classifying the T wave as the R peak

Assisting Humans with Special Needs: Curriculum for HUman-TOol Interaction Network - HUTON

The arising field of biomedical engineering is becoming more popular worldwide. TEMPUS project ”Assisting humans with special needs: Curriculum for HUman-TOol interaction Network - HUTON” has a goal to provide new curriculum and necessary equipment for three universities in Serbia, as well as to establish a collaboration between nine partners involved in project from four countries (Serbia, Slovenia, Greece and Italy) for future research in the field of mechatronic in medical rehabilitation. This is of great importance for institutions in Serbia to be recognized and introduced to EU scientific community.