| Title |
Development of a Characteristic Point Detection Algorithm for the Calculation of Pulse Wave Velocity |
| Authors |
이낙범(Lee, Lark-Beom) ; 임재중(Im, Jae-Joong) |
| Keywords |
PWV ; Pulse wave velocity ; Intersecting tangent method ; Least square method |
| Abstract |
Shape of the pulse waveform is affected by the visco-elasticity characteristics of the arterial wall and the reflection waves generated at the bifurcations of arterial branches. This study was designed to improve the accuracy for the extraction of pulse wave features, then proved the superiority of the developed algorithm by clinical evaluation. Upstroke point of the pulse wave was used as an extraction feature since it is minimally affected by the waveform variation. R-peak of the ECG was used as a reference to decide the minimum level, then intersection of the least squares of regression line was used as an upstroke point. Developed algorithm was compared with the existing minimum value detection algorithm and tangent-intersection algorithm using data obtained from 102 subjects. Developed algorithm showed the least standard deviation of 0.29~0.44 m/s compared with that of the existing algorithms, 0.91~3.66 m/s. Moreover, the rate of standard deviation of more than 1.00m/s for the PWV values reduced with the range of 29.0~42.4%, which proved the superiority of the newly developed algorithm. |