Technical note: A device for long term ambulatory monitoring in trans-tibial amputees
H. J. Stam *
F. Eijskoot *
J. B. J. Bussmann *
Long term monitoring of walking in trans-tibial amputees (TTA) is considered important for prosthetic prescription and therapy evaluation.
The purpose of this study was to develop a device with the following design criteria: lightweight, easy attachment to the prosthesis, energy and memory capacity for five days and practical in clinical use.
The prototype (CAMP: Continuous Ambulatory Monitoring of Prosthetic walking) consists of a cylindrical unit containing an accelerometer, a miniature computer and six batteries. Specifications, data acquisition and processing, instructions for users and first results are presented and discussed.
The CAMP prototype proved to be a useful device for measuring relevant aspects of prosthetic use for up to five days.
In physical medicine and rehabilitation, clinicians and researchers are especially interested in measuring patient performance in relation to functional or disability level. For trans-tibial amputation (TTA) patients walking time and number of walking periods is a relevant parameter in prosthetic prescription and therapy evaluation.
Instruments for measuring walking time/frequency, such as observational techniques and diaries are subjective, time-consuming or difficult to interpret (Montoye and Taylor, 1984; Geurts et al., 1991; Washburn et al., 1989), Measurement within a laboratory or hospital setting is artificial and tends to be of limited duration, thus limiting the value of these data (Geurts et al., 1991; Leo and Woltring, 1990; Rose et al., 1991; Rozendal, 1989; Rozendal, 1990).
Ambulatory monitoring of walking is more objective and enables measurement in a natural setting during routine daily activity. Methods to assess physical activity have been described, but are based on outdated electronic or mechanical techniques which count steps and/or register intensity of movement (Montoye and Taylor, 1984; Meijer et al., 1991).
A potential shortcoming of ambulatory monitoring is that the instrument influences spontaneous behaviour (Cook and Campbell, 1976) so that results from one day may not be representative of the performance on other days.
For that reason the authors developed the CAMP (Continuous Ambulatory Monitoring of Prosthetic walking) device.
The following were the design criteria:
- Lightweight in order to avoid interference with the biomechanics of the prosthesis.
- Attachable to the tube of the prosthesis under the foam cover. This allows measurement whilst the patient is unaware of being monitored.
- Sufficient energy and memory capacity for 5 days.
- Easy data processing and presentation for clinicians.
The CAMF consists of a cylindrical delrin unit, which can be attached around the metal tube of the prosthesis (Fig. 1). The unit contains batteries, a sensor and hard-computer and software. The six batteries (AAA Duracell MN2400) have a capacity of 7 days. A miniature piezoresistive accelerometer registers the acceleration peak at heel strike. A specially designed detection circuit (Fig. 2) detects "walking" from the accelerometer signal.
The software in the unit directs the datalog function, which results in activation of the logger every second. Depending on the detector indicating walking or no walking, a "zero" or "FF" is loaded in a HEX file. The memory capacity of 490 K allows a total registration time of 136 hours (5.6 days). The weight of the CAMP, batteries included, is 0.3 kg.
An IBM compatible personal computer (486 or 386 coprocessor) is used to start and stop the unit. Computer and unit are attached by a RS232 plug and the instructions are presented by a menu.
Results are displayed as walking time and number of walking periods per day and histograms indicate the number of walking periods, with a preset time interval. Walking is only registered if more than x seconds are detected, x being preset between 0 and 20 (walking threshold).
Fig. 3 and Fig. 4 show the results of ambulatory monitoring in a 62 year old TTA patient. The amputation was performed 8 weeks before the CAMP test and the patient was currently involved in an outpatient rehabilitation programme 3 times a week. The walking threshold was set at 10. In Fig. 3 the walking periods during the respective days are presented as interrupted lines. On Wednesday, Friday and Monday the walking activity is most prominent in the early afternoon (days 1, 3 and 6). On Saturday (day 4) some walking is registered in the late afternoon until after midnight; Sunday (day 5) is a restful day.
For the same week histograms represent the number of walking periods per day with 20 intervals (Fig. 4). Most walking periods last no longer than 1 minute. Walking longer than 2 minutes is exceptional.
The CAMP is a lightweight instrument for measuring walking time of amputees for up to 5 days. If required, the patient can be kept unaware of the device, which is attached to the prosthesis under the foam cover. Further research will be directed at improving the size, weight and memory capacity.
- Cook TD, Campbell DT (1976). The design and conduct of quasi-experiments and true experiments in field setting. In: Handbook of industrial and organizational psychology ./edited by MD Dunette. -Chicago: Rand McNally. p223-323.
- Geurts ACH, Mulder T, Rijken RAJ, Nienhuis B (1991). From the analysis of movements to the analysis of skills: bridging the gap between laboratory and clinic. J Rehabil Sci 4, 9-12.
- Leo T, Woltring HJ (1990). The CAMARC project: an attempt to bridge the gap. J Rehabil Sci 3, 110-112.
- Meijer GAL, Westerterp KR, Verhoeven FMH, Koper HBM, ten Hoor F (1991). Methods to assess physical activity with special reference to motion sensors and accelerometers. IEEE Trans Biomed Eng 38, 221-229.
- Montoye HJ, Taylor HL (1984). Measurements of physical activity in population studies: a review. Hum Biol 56, 195-216.
- Rose SA, Ounpuu SA, DeLuca PA (1991). Strategies for the assessment of pediatric gait in the clinical setting. Phys Ther 71,961-980.
- Rozendal RH (1990). Gait analysis and ICIDH. J Rehabil Sci 2, 89-93.
- Rozendal RH (1989). Gait analysis, a clinical tool? J Rehabil Sci 3,96-97.
- Washburn RA, Cook TC, Laporte RE (1989). The objective assessment of physical activity in an occupationally active group. J Sports Med Phys Fitness 29, 279-284.