A review of methods and challenges involved in biomanufacturing & evaluating the validity of wrist worn pedometers

Date of Award


Degree Name

M.S. in Chemical Engineering


Department of Chemical and Materials Engineering


Advisor: Donald A. Comfort


Over the last couple decades, there has been much research invested in developing biocompatible sensors and integrating these into monitoring devices, but few have achieved commercial implementation. There are many hurdles that have limited the technology transfer of microelectromechanical systems (MEMs), microfluidics, and bio-micro-sensors from bench-scale production to commercialization of these devices including improvements to manufacturing techniques, assembly challenges, and sterility concerns. Development of new manufacturing technologies such as LIGA and miniaturization of existing technologies such as lithography, hot embossing, injection molding, and plasma etching have created manufacturing options for these bio-micro-sensors. These techniques, as well as an assessment of material of construction and sterility and packaging were part of the manufacturing process analysis. Once produced, the reliability and accuracy of the devices must be confirmed. One of the means of assessing devices produced through biomanufacturing processes is to perform validity testing. As a means of assessing future manufactured devices, commercially available pedometers were used to validate an approach to compare their performance against highly reliable data. Pedometers and physical activity tracking devices are built with MEMs and micro-sensors to give step count data and other physiological information. In the last few years, step-counting technology has been incorporated into activity tracking devices worn around the wrist as opposed to having a pedometer attached to the waist of one's shorts. The accuracy of these devices is relatively unexplored, and therefore the accuracy of the step counting feature in three different brands of activity trackers worn around the wrist (Nike+ Fuelband, Fitbit Flex, and Sportline SyncBurn) was assessed. Sixteen participants (8 males and 8 females) ages 19 to 25 walked 500 steps around an indoor track with one activity tracker around their wrist and steps were manually tracked using hand clickers. Using ANOVA analysis (p < 0.05), it was determined the Nike+ Fuelband counted 82.3 ± 15.8% of the total steps walked compared to the hand-tally clickers. The Fitbit Flex (97.8 ± 17.4%) and Sportline SyncBurn (99.0 ± 14.7%) activity trackers did not produce values statistically different from one another or the hand-tally clickers. The experiment demonstrated the utility of a small sample size in assessing device performance when using a statistically balanced experimental design.


Pedometers Reliability, Biosensors Reliability, Pedometers Production control, Exercise Measurement, Chemical Engineering, Biomedical Engineering, Biomanufacturing, Micro-devices, Sterility, Accelerometers, Activity Tracking, Wrist Pedometers

Rights Statement

Copyright © 2014, author