سال انتشار: ۱۳۹۱

محل انتشار: بیستمین کنفرانس سالانه مهندسی مکانیک

تعداد صفحات: ۴

نویسنده(ها):

Mohammad R. Homaeinezhad – Department of Mechanical Engineering, K. N. Toosi University of Technology
Ali Golmirzaei –
Hossein Naseri –

چکیده:

This paper describes how one can design a computer program and implement it on an ordinary PC or personal digital assistant (PDA) for detecting of very long duration electrocardiogram (ECG) PQRST events with processing capability (time complexity) less than 1300 clock/samples along with the accuracy more than 99.86%. Although up to now, very efficient ECGdetection algorithms have been proposed by authors; however, an everlasting question is that whether it is possible to apply those algorithms to very long duration holter ECG using a PDA, an ordinary PC or a mobile phone central processor with outstanding accuracy. The answer is NO because in the proposed algorithms a strong direct proportionality can be found between the numerical complexity of the methods and total accuracy. The constitutive methods applied in this studyare mainly based on the human thinking models. For detection and delineation of QRS complexes, a noiserobust instantaneous concavity analysis was applied. For detection and delineation of P- and T- waves, the analysis of local extremum of the QRS-eliminated signal was used. The proposed method was applied to several databases (more than 5,000,000 beats which were normal or abnormal) with different samplingfrequencies and bit-rates. Also, the method was applied to 100000 number of beats obtained from some male rats during ischaemia and reperfusion tests. The accuracy of method was evaluated as 99.86% with very slight dependency of the algorithm parameters to acquisition conditions.