Elsevier

Gait & Posture

Volume 23, Issue 1, January 2006, Pages 118-125
Gait & Posture

Accuracy and repeatability of the Pedar Mobile system in long-term vertical force measurements

https://doi.org/10.1016/j.gaitpost.2005.05.008Get rights and content

Abstract

Portable insole pressure systems can be used to measure the vertical force during long-term (hours) measurements to determine the patient's amount of weight bearing during daily activities in the hospital and at home. Especially for long-term measurements, the amount and duration of loading pressure insoles can have a large influence on the accuracy, as previous studies found a time-dependent behavior after a relatively short period (minutes) of constant loading. Therefore, this study assessed the accuracy and repeatability of a portable capacitive insole system (Pedar, NovelGmbH) to measure vertical force during long-term loading. Static loading experiments were performed during which the Pedar insoles were loaded with 5 and 10 N/cm2 for 7 h. Dynamic loading experiments were performed with one Pedar insole which was cyclically loaded with 300, 500 and 1000 N during two sessions of 1200 load cycles. The static and dynamic experiments were repeated 3 days later. Accuracy, due to offset drift, decreased in time during the start of the static experiments (percent error: −1.9% to 0.3% at hour 0; 26.3% to 34% at hour 7). The percent error for the dynamic experiments ranged from −16% to −19%, from −3% to −7% and from −8% to ∼0% when the insole was loaded with 300, 500 and 1000 N, respectively. The amount of drift ranged from 12 to 62 N for the 500 and 1000 N loads, respectively. The mean day-to-day percentage difference for the static and dynamic experiments ranged from −2.3% to 0.5%, and from −2.9% to 3.0%, respectively. The results indicate that drift correction is necessary for accurate assessment of vertical force by the Pedar Mobile system to determine the amount of weight bearing during long-term measurements.

Introduction

Partial weight bearing is commonly instructed for 6–12 weeks for patients with a lower limb fracture or arthroplasty because weight bearing higher than the treating surgeon has allowed can lead to non-union of bone fragments or instable implants [1], [2], [3]. Long-term measurement of weight bearing in the hospital and at the patient's home during daily activities can provide more relevant information about the patient's loading pattern (e.g. average or extreme peak loads during a day) than the assessment of weight bearing in a laboratory, where only a few steps can be measured and evaluated.

Portable insole pressure systems can be used to measure the vertical force during weight bearing [4], [5], [6], [7]. However, only a few studies described portable insole devices that can record in-shoe pressures or ground reaction forces over a long-term period [8], [9]. To our knowledge, no studies have investigated the validity and reliability of these systems to measure the vertical force over a long-term period. The Pedar Mobile system (NovelGmbH, Munich, Germany), a portable in-shoe pressure system using 99 capacitive sensors per insole, can record vertical force data for a maximum of one hour when continuously measuring with two insoles at a sample rate of 50 Hz. The restrictions of the Pedar system for mobile measurements during several hours are that the recording period is limited due to the maximum data storage capacity of the largest PCMCIA card (40 Mb: 60 min data collection at 10.000 samples/s) used by the Pedar system, and because the largest Pedar battery (2100 mAh NiMH) provides power for a maximum of one hour. To perform long-term evaluation of weight bearing we adapted the Pedar Mobile system by connecting an electronic device to the Pedar box that works as an automatic start-stop trigger so that data are only recorded during standing and walking, and by using a custom-made battery-unit (consists of two Sony NP750 Li-ion batteries) which supplies the Pedar system power for 3.5 h.

During long-term insole measurements, the accuracy and repeatability may, besides factors such as temperature and humidity, particularly be influenced by the amount and duration of loading [10], [11], [12], [13], [14]. Previous bench test studies reported time-dependent behavior of capacitive insoles in which sensor values increased by 3.5–4.5% after 10–15 min of continuous loading of the insoles [10], [11]. Especially when the insoles are continuously loaded for a much longer period (hours), this drift may be even more significant. Besides the amount of drift during long-term measurements, we were also interested in the type of sensor drift. When offset drift occurs the discrepancy between the measured value (Pedar) and the actual value (testing device) is independent from the actual value, thus this difference will be constant. Therefore, offset drift is easy to correct for. This is in contrast with gain drift in which the discrepancy between the measured value (Pedar) and the actual value (testing device) is dependant from the actual value and, therefore, this difference will not be constant. No studies were found regarding the effect of long-term static and/or dynamic loading on the amount and type of sensor drift of capacitive pressure insoles.

The purpose of this study was to assess the accuracy and repeatability of the Pedar Mobile system to measure vertical force during long-term static and dynamic loading by conducting a series of bench tests. In addition, the type of drift was investigated during a long-term static loading experiment.

Section snippets

Materials

The Pedar Mobile system is a matrix insole system in which vertical (columns) and horizontal (rows) metal strips are fixed to either side of a dielectric material. At each intersection point of these rows and columns, there is a capacitance sensor. The sensors are placed between two layers of polyethylene, which provides insensitivity to humidity within the shoe, and are covered on both sides with artificial leather. Each insole consists of 99 sensors, which are equally spread over the insole

Static long-term loading experiment

The percent error of measurement of the insoles during the static experiments ranged from −2.2% and 0.3% at hour 0 (Table 1). From hour 1–7, the percent error of measurement varied from 10% to 43% depending on the duration of loading and applied load. The relative drift after 10 and 20 min was 9–9.5% when loading the Pedar insoles with a static load of 5 N/cm2, after which a more steady increase in drift was found of 12.3% and 12.6% (left and right W type insole) at hour 1 to a drift of 26.3% and

Discussion

The present study investigated the accuracy and repeatability of the Pedar system to measure vertical force during long-term static and dynamic loading. In addition, we studied the type of drift because if the drift was an offset drift then correction of this drift would be relatively straightforward.

The accuracy of the Pedar system at the start (hour 0) of the static experiments was good with relatively small differences compared to the used loads, but from hour 1 to 7, a substantial amount of

Acknowledgements

The authors thank J. van de Breevaart Bravenboer, Department of Orthopaedic Surgery, Erasmus MC, University Medical Center, Rotterdam, The Netherlands, for writing the loading program for the Lloyd instruments loading device, and H.L.D. Horemans, Department of Rehabilitation Medicine, Erasmus MC, University Medical Center, Rotterdam, The Netherlands, for writing the Matlab programs for the data analysis.

References (17)

There are more references available in the full text version of this article.

Cited by (79)

  • Evaluation of novel plantar pressure-based 3-dimensional printed accommodative insoles - A feasibility study

    2022, Clinical Biomechanics
    Citation Excerpt :

    Therefore, the results of the two pairs of insoles with accommodative top layer reflects the initial use condition and may not generalize to the condition after the break-in period. Although the pressure measurement tool used in this study has been shown to have −19% and 3% error when the insole was loaded with different levels of loading (Hurkmans et al., 2006) the accuracy and consistency outperforms other plantar pressure systems (Quesada et al., 1997). This preliminary work validates our ability to manufacture the 3D printed insoles and demonstrates their ability to reduce plantar pressure more than the standard of care insole in our hospital while not significantly increasing pressure in the adjacent regions.

View all citing articles on Scopus
View full text