The present invention relates to a method and a program to detect a pole point. More specifically, the present invention relates to a method and a program to stably and accurately detect a pole point of a quasi-periodic signal like a plethysmogram (PG) in real time. The method to detect the pole point comprises: a step of initializing a decay coefficient, an adaptive threshold curve value, and a tracking state value a step of inputting the quasi-periodic signal a step of removing noises of the quasi-periodic signal a step of updating the adaptive threshold curve value with a current value of the quasi-periodic signal to provide an adaptive threshold curve value of a next time, and to set the tracking state value as a &primetracking state&prime when the adaptive threshold curve value is smaller than or equal to the current value of the quasi-periodic signal by comparing the adaptive threshold curve value and the current value of the quasi-periodic signal a step of repeating a process of comparing a current adaptive threshold curve value with a current quasi-periodic signal value inputted by a time flow by returning to the step of inputting the quasi-periodic signal, and calculating the adaptive threshold curve value of the next time by adding the decay coefficient to the current adaptive threshold curve value when the current adaptive threshold curve value exceeds the current quasi-periodic signal value a step of designating a generation point of the current adaptive threshold curve value as a pole point candidate when the tracking state value is the &primetracking state&prime, and returning to the step of inputting the quasi-periodic signal when the tracking state value is a &primenon-tracking state&prime and a step of returning to the step of inputting the quasi-periodic signal by setting the tracking state value as the &primenon-tracking state&prime during a refractory period after designating the pole point candidate, and designating the pole point candidate as a