Introduction
In the HBCD study, Axivity Ax6 sensors were used to record infant leg movements across 72 continuous hours. Research assistants of the study placed one sensor on the distal right ankle and another on the distal left ankle, using legwarmers with a pocket to hold the sensor.
Sensors were set to start recording at 10 a.m. eastern / 9 a.m. central / 8 a.m. mountain / 7 a.m. pacific. Caregivers were instructred to go about their typical activities in the natural environment but to remove the sensors if the baby went into water (e.g., bathtub or pool) and replace them afterward.
Leg movement data recording occurred at V02 (0-1 months of age) and at V03 (3-8 months of age).
Before the 72 hours of data were collected, a calibration file was collected for each sensor. Instructions for collection of the calibration data were: “There are 6 flat surfaces of the sensor and we want to record data with the sensor sitting still on each of its flat surfaces. To do this: place the sensor on a level, flat surface (e.g., the surface of a desk or table). Wait 10 seconds. Rotate it so that it is resting on its next flat surface. Wait 10 seconds. You should put the sensor in 6 different positions and collect 10 seconds of data in each position, so just over a minute of data in total (including the time to rotate it). It does not matter what order you do this in.”
Data files included in the data release are raw sensor data in BIDS format for the calibration and 72-hour files for the right and the left leg (Data inputs to the container: raw BIDS files), as well as files containing processed data Processed data outputs.
Key references
Describing the protocol decision-making process: Pini, N., Fifer, W. P., Oh, J., Nebeker, C., Croff, J. M., Smith, B. A., & Novel Technology/Wearable Sensors Working Group (2024). Remote data collection of infant activity and sleep patterns via wearable sensors in the HEALthy Brain and Child Development Study (HBCD). Developmental Cognitive Neuroscience, 69, 101446. https://doi.org/10.1016/j.dcn.2024.101446
BIDS format for raw data files: Jeung, S., Cockx, H., Appelhoff, S., Berg, T., Gramann, K., Grothkopp, S., Warmerdam, E., Hansen, C., Oostenveld, R., BIDS Maintainers, & Welzel, J. (2024). Motion-BIDS: an extension to the brain imaging data structure to organize motion data for reproducible research. Scientific Data, 11 (1), 716. https://doi.org/10.1038/s41597-024-03559-8
Calibration process to prepare data for calculation of infant leg movement characteristics: Oh, J., Loeb, G. E., & Smith, B. A. (2024). The Utility of Calibrating Wearable Sensors before Quantifying Infant Leg Movements. Sensors, 24 (17), 5736. https://doi.org/10.3390/s24175736
Algorithms to identify infant leg movement characteristics: Smith, B. A., Trujillo-Priego, I. A., Lane, C. J., Finley, J. M., & Horak, F. B. (2015). Daily Quantity of infant leg movement: Wearable sensor algorithm and relationship to walking onset. Sensors, 15 (8), 19006-19020. https://doi.org/10.3390/s150819006
Trujillo-Priego, I. A., & Smith, B. A. (2017). Kinematic characteristics of infant leg movements produced across a full day. Journal of Rehabilitation and Assistive Technologies Engineering, 4, 2055668317717461. https://doi.org/10.1177/2055668317717461
Trujillo-Priego, I. A., Zhou, J., Werner, I. F., Deng, W., & Smith, B. A. (2020). Infant Leg Activity Intensity Before and After Naps. Journal for the Measurement of Physical Behaviour, 3 (2), 157-163. https://doi.org/10.1123/jmpb.2019-0011
Oh, J., Ordoñez, E. L. T., Velasquez, E., Mejía, M., Del Pilar Grazioso, M., Rohloff, P., & Smith, B. A. (2024). Associating neuromotor outcomes at 12 months with wearable sensor measures collected during early infancy in rural Guatemala. Gait & Posture, 114, 477-489. https://doi.org/10.1016/j.gaitpost.2024.08.005
Algorithm to estimate intensity of infant physical activity: Ghazi, M. A., Zhou, J., Havens, K. L., & Smith, B. A. (2024). Accelerometer Thresholds for Estimating Physical Activity Intensity Levels in Infants: A Preliminary Study. Sensors, 24 (14), 4436. https://doi.org/10.3390/s24144436
Quality Control (QC) Processes
QC Procedures: Some raw data files were checked for quality. Only a small percentage of data files were randomly checked each week as the process was manual and visual. When checked, calibration files were checked for presence of adequate data for each of 6 axes and 72-hour files were checked for the presence of data, labeling of right and left leg, and sampling rate used.
Common Issues Identified: Common issues identified during the QC proceeses included inadequate data for each of the six axes in calibration files (human error), missing data for calibration files (due to human error or technical difficulties), missing data for 72 hours (due to human error, technical difficulties, or parent/ legal guardian declining to participate in this aspect of the study), sensors being removed from prolonged periods during the 72 hours, or the use of incorrect sampling rate during the 72 hour collection. If possible, errors were corrected (but this was not often possible). All issues occurred rarely overall and the majority of the data were judged to be present and correctly collected. If data from a particular visit fell under any of the aforementioned scenarios, the preprocessing pipeline would fail to process the data and generate an error log (LOG.txt).
Potential Issues Flagged by Subject Matter Experts
No issues were found.