Informatique

—The processing of the electrocardiogram (ECG) signal consists essentially in the detection of the characteristic points of signal which are an important tool in the diagnosis of heart diseases. The most suitable are the detection of R waves. In this paper, we present various mathematical tools used for filtering ECG using digital filtering and Discreet Wavelet Transform (DWT) filtering. In addition, this paper will include two main R peak detection methods by applying a windowing process: The first method is based on calculations derived, the second is a time-frequency method based on Dyadic Wavelet Transform DyWT.

  RESUME  Ce travail se situe dans le cadre de calcul du réseau de neurones dans des domaines  d'applications de très haute performance. L'article propose une architecture qui sera  adaptée aux contraintes de conception détaillées. L'implémentation matérielle des réseaux  neuronaux optimisés sur circuits reconfigurables est notamment pour accélérer au  maximum le processus d'apprentissage sur puce. Ce papier va étendre la méthodologie  de prototypage rapide Adéquation Algorithme Architecture (AAA) associée au logiciel  SynDEx-IC pour optimiser et implémenter un algorithme de réseau de neurones MLP

Aim:  to  investigate  vascular  reactivity  on  forearm  skin  blood  flow  (FSBF)  in  response  to  iontophoresis  of acetylcholine (ACh) after an acute exercise. Materials  and  method: 49 healthy male subjects 29 trained and 20 sedentary (recruited according to their oxygen consumption) participated in this case control study (mean age: 15 ± 1 year). Incremental exercise testing was performed by means of an individualized and personalized exercise test protocol. FSBF was measured before and after exercise using a laser Doppler flowmeter in response to local iontophoresis of a cumulative dose of ACh. Results: Although groups were not different in age, height or body mass index, weight, resting heart rate, maximum oxygen consumption was significantly different (sedentary 40.05 ± 4.53 and trained 51.65 ± 5.76 ml.min-1.kg-1) (p<0.05; p<0.001; p<0.001, respectively). The mean response of FSBF to ACh was significantly increased after incremental exercise in the two groups (349%). Maximal FSBF response to ACh was significantly greater in trained (987% ± 78) than in sedentary (638% ± 42) (p=0.001) after a single acute exercise. Conclusion: Response of FSBF to ACh-induced endothelium dependent relaxation was affected by the level of training whether before or after an acute exercise.