The Accuracy of Wireless Accelerometers in Detecting the Leg Movements of Young Infants: A Pilot Study
1Chapman D, 2Lopez C, 2Zhao K
1St. Catherine University, Minneapolis, MN, United states; 2Mayo Clinic, Rochester, MN, 55905
OBJECTIVE: Infant leg movements (LMs) are important behaviors to study because they are related to when babies learn to walk. Video-based behavior coding is the ‘gold-standard’ for identifying infant LMs. Little research has been conducted to determine if wearable accelerometers can accurately detect LMs produced by young infants. The purposes of this pilot study were to examine if wearing the accelerometers affected how often the babies moved their legs and kicked over developmental time, and compare the accuracy of the sensors to behavior coded video-taped data to identify the LMs produced by pre-walking infants. Our long term goal is to assess early LMs in infants with spina bifida post-operatively.
DESIGN/METHOD: Following IRB approval, four typically developing (TD) infants, 2 boys and 2 girls, were recruited to participate in this longitudinal pilot study. They entered the study at 4 to 5 weeks of age and completed their participation when they were between 16 and 17 weeks of age. None of the infants presented with a known visual or auditory impairment and all passed their newborn hip screen. The infants’ spontaneous LMs were video-taped when they were supine for 1.0 to 1.5 minutes at 30 frames per second once a month for 4 consecutive months with & without the wearable sensors attached to the anterior surface of their thighs and lower legs. The video-taped data was behavior coded through frame-by-frame analysis to identify the frequency of LMs and kicks in each condition. Matla-based (Mathworks, Inc.) algorithms are being developed to automatically identify infant LMs based on the mean peak resultant acceleration of each infant’s LMs.
RESULTS: Wearing the accelerometers did not result in a significant change in the number of LMs or kicks generated by this small group of infants at each age (p = .619, partial eta = .263, observed power = .294).The mean peak resultant acceleration from each infant ranged between 1.40 to 10.17 m/s2. These acceleration data are currently being used to determine the accuracy of algorithms for automatic detection of kicks.
CONCLUSIONS: Young TD infants between 1 and 4 months of age can wear wireless accelerometers on their legs without decreasing how often they move their legs and kick. The current results are similar to previous research. Additional research is warranted to help validate the use of wearable accelerometers for automatic kick detection.
SUPPORT: This research is funded by the Mayo Clinic Children’s Research Center Pediatric Team Science Award.
You Must Be Logged In
This Reminders tab will display for your scheduled lectures 15 minutes before start time and for new messages waiting.