Assessing motor development with wearables in low-resource settings: feasibility in rural Malawi.

Abstract

Background: Tracking of early motor development is essential for all neurodevelopmental assessments. A multisensor wearable system, MAIJU (Motor Assessment of Infants with a JUmpsuit), was recently developed for an objective and scalable measurement of developing motor skills in out-of-hospital settings. Here, we assessed its feasibility in remote low-resource settings.

Methods: We recruited 44 infants for repeated at-home measurements (total N = 121) with the MAIJU wearable in rural Malawi. We assessed (i) technical quality of the measured wearable data, (ii) reliability of the cloud-based analysis outputs, and (iii) maternal user experience. A dataset from 47 infants (total N = 111 measurements) in Finland served as a reference from a high-resource environment.

Results: Altogether 94% of the measurements were technically successful. The analysis outputs from the automated cloud pipeline were all comparable to the reference cohort in Finland. The method was rapidly learned by the local study personnel, and it was well received by the mothers.

Conclusion: Our results suggest that advanced multisensor wearables and cloud-based analytics can be readily used in remote and low-resource settings. Uptake of such objective methods holds promise for harmonizing and increasing equality in developmental assessments, as well as facilitating a wide range of global health studies on early life.

Impact: Motor development is an effective measure of infants' overall neurodevelopment.^1-4 A multisensor wearable system was recently developed for an objective and scalable tracking of infants' developing gross motor skills.^5-7 Here, we assessed feasibility of using such wearable systems in low-resource settings in rural Malawi. Our findings show that the measurements are technically reliable, the outputs from the cloud-based analysis pipeline are comparable to those from our reference study in Finland, and the wearable recordings are well-received by the parents. The findings support the use of multisensor wearables in remote settings and highlight their potential for benchmarking early-life global health studies.