Pergamon PIh S0021-9290(96)00074-7 J. Biomechanics, Vol. 29, No. 12, pp. 1643-1647, 1996 Copyright© 1996 Elsevier Science Ltd.All rightsreserved Printedin GreatBritain 0021-9290/96 $15.00+ .00 TECHNICAL NOTE A TECHNIQUE FOR THE MEASUREMENT OF CADENCE USING WALKWAY VIBRATIONS D. G. Lloyd and N. L. Svensson Department of Safety Science, University of New South Wales, Sydney, N.S.W., 2033, Australia Abstraet--A new method has been developed to measure cadence while walking. The method uses instrumentation to detect the vibrations transmitted in the walkway surface. These vibrations are produced when a subject's foot makes initial contact with the walking surface. The method was tested against heel switches and was found to measure stride duration with RMS errors of around 1%. The vibration method requires no instrumentation to be attached to the subject, can be used with bare feet or any other footwear, is simple to use and is very robust and reliable. Copyright © 1996 Elsevier Science Ltd. Keywords: Gait analysis; Cadence; Measurement; Vibrations; Heel strike. INTRODUCTION METHODS The temporal-spatial parameters of cadence, walking speed, and stride length are recognized as good indicators of overall gait function and are used regularly to characterize and assess gait (e.g. Andriacchi et al., 1977; Todd et al., 1989). Cadence, or stride duration, can be determined if the time between successive foot contacts is known. Previously, instrumentation attached to the foot or shoe has been used for this purpose, e.g. shoes with switches or sensors (Nilsson et al., 1985; Winter et al., 1972), foot switches or sensors (Hausdorff et al., 1995; Ross and Ashman, 1987; Todd et al., 1989) and conductive walkways (Chao et al., 1983). However, these methods may alter gait since they employ a tether electrical cord or telemetry unit attached to the subject, or switches or sensors attached to the barefoot. Additionally, the analysis of barefoot gait is not possible with instrumented shoes and switches are prone to failure (Nilsson et al., 1985). Resistive walkways (Crawford et al., 1985), multiple switch walkways (Hirokawa and Matsumara, 1987) and multiple or large force plates (Brodland and Thorton-Trump, 1987) do not rely on instrumentation attached to the subject. The resistive and multiple switch walkways usually have a compliant walking surface which may not be desirable in some circumstances. Multiple or large force plates can be used to measure stride duration with no interference with normal gait; however, these are not always available and are expensive. We have developed a new system to measure cadence, which overcomes some of the shortcomings of the previously men- tioned systems. The new system was developed to (a) not encum- ber the subject, (b) be simple to setup and use, (c) be robust and reliable, (d) be inexpensive to implement, and (e) enable immedi- ate feed back of results. This new system was evaluated by comparison with heel switches. Received in final form 19 April 1996. Address correspondence to: Dr David Lloyd, Department of Human Movement, University of Western Australia, Nedlands, W.A., 6907, Australia. System description Initial foot contact timing is detected from the vibrations transmitted in the walkway (Fig. 1) when a subject's foot made initial contact with the walking surface (usually heel strike). The walkway vibrations are measured by two custom made sensitive accelerometers (Fig. 2) secured on the side of the walkway frame. The signals from the accelerometers are electronically pro- cessed by (a) amplification, (b) band pass filtering, (c) full wave rectification, (d) envelope filtering, (e) differentiation and (f) thresholding to form square wave pulses (Fig. 3). The final processing includes some debouncing to remove any false trig- gering from secondary vibrations. This debouncing is a 350 m s period of delay in which no triggering of the output can occur. This delay corresponds to a maximum cadence of around 170 stepsmin -l. Since the walkway and force platform are es- sentially isolated from each other, a heel strike on the force platform is registered by the accelerometers as an attenuated and delayed vibration. It is then necessary to use the vertical force signal from the force platform at foot contact to override the vibration detection circuit. The output of the electronics unit also drives a buzzer which gives an audible click at every heel strike. This provides the operator with a qualitative check that the system is operating correctly. The dominant response of the accelerometers is from the initial impact of the foot, which is a damped oscillation of the walkway at its natural frequency, approximately 200 Hz. The accelerometers also detect low amplitude and low-frequency movements of the walkway, which correspond to the loading of the walkway during the stance phase of gait. There is also a small, delayed and unreliable secondary vibration that may occur due to the foot flat event or reflections from the initial vibration. The band pass filtering and debouncing of the output ensure the initial impact component of the vibration generates the system's output signal. In the present gait system, the vibration method is used to measure cadence by measuring the time between foot contacts that occur at the beginning and end of two strides, i.e. four steps, one stride either side of the force plate. This means that cadence is measured from either left foot contacts or right foot contacts which occurred within 0.5 m of an accelerometer. 1643