Wearable accelerometers for measuring and monitoring the motor behaviour of infants with brain damage during CareToy-Revised training.
Mattia Franchi De' Cavalieri, Silvia Filogna, Giada Martini and 6 others
PMID 37149595WHAT IT FOUND
Wearable arm and trunk sensors recorded movement in 17 infants with brain damage during a home training programme.
Sensor readings tracked AIMS and IMP total scores, but not IMP fluency. The sample was too small to validate any clinical tool.
Key findings
01Accelerometer readings during play were significantly related to baseline AIMS total scores and to IMP symmetry, performance and total scores.
02The IMP fluency domain showed no relationship to the sensor data in either the correlation or the prediction models.
03The authors state the models are not fully validated because of the small number of infants, the variability between them and the absence of a control group.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTsWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Seventeen infants, with only 190 time points kept for modelling. The authors say this is too few to claim full validation. There was no control group of typically developing infants, so nothing can be said about what normal movement looks like on these sensors. Every infant did a different, personalised set of activities at any given minute, so the comparison between infants is not like for like. The clinical scores varied widely between infants. The models predicted preterm infants to have higher AIMS scores than term infants, which the authors admit is the opposite of clinical experience. This is a first-week feasibility report of a measurement method, not a trial of an intervention.
Declared interests
The paper states funding from the Italian ministry of health, ministero della salute. No author conflict-of-interest declaration is given in the text.
The easy way to misread this
Do not read the adjusted R squared of 0.99 for AIMS as evidence that accelerometers can replace the AIMS or IMP. The authors state there is too little data to claim a full validation, there was no control group, and the leave-one-out error for IMP symmetry was 21% to 30%. A model fitted on 17 infants will look excellent on those 17 infants and nothing more.