MULTIRESOLUTION SEGMENTATION OF RESPIRATORY ELECTROMYOGRAPHIC SIGNALS

Analysis of respiratory electromyographic (EMG) signals in the study o f respiratory control requires the detection of burst activity from ba ckground(signal segmentation), and focuses upon the determination of o nset and cessation points of the burst activity (boundary estimation). This paper describes a new automated multiresolution technique for si gnal segmentation and boundary estimation; During signal segmentation; a new transitional segment is defined which contains the boundary bet ween background and burst activity. Boundary, estimation is then perfo rmed within this transitional segment. Boundary candidates are selecte d and a probability is attributed to each candidate, using an artifici al neural network. The final boundary for a given transitional segment is the boundary estimate with the maximum a posteriori probability. T his new method has proved accurate when compared to boundaries chosen by two investigatiors.