Patel Priya, Shi Yan, Hajiaghajani Faezeh, Biswas Subir, Lee Mei-Hua
Department of Kinesiology, Michigan State University, East Lansing, MI, USA.
Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI, USA.
Infant Behav Dev. 2019 Nov;57:101383. doi: 10.1016/j.infbeh.2019.101383. Epub 2019 Oct 16.
Spontaneous movements, which refer to repetitive limb movements in the absence of any external stimulus, have been found to be reflective of neurodevelopmental status during infancy. These movements are modulated by both individual and environmental factors, including physical contact (holding) with the caregiver. However, it is a challenge to measure spontaneous movements during physical contact because infant-generated movements become coupled with caregiver-generated movements in such contexts. Here, we propose the use of a novel two-body sensor system to distinguish infant-generated movements in the presence of physical contact with the caregiver. Data from seven typically developing infants and their caregivers were recorded during different simulated home activities, which involved different combinations of physical interaction, caregiver's movement and infant positions. The two-body sensor system consisted of two wearable accelerometers - one placed on the infant's arm and one on the caregiver's arm, and we developed a Kalman-filter based algorithm to isolate the infant-generated movements. In addition, video was recorded for qualitative analysis. Results indicated that spontaneous movement activity was higher when there was no physical contact with caregiver. When there was physical contact, spontaneous movements were increased when the caregiver was still and when the infant was held horizontally. These results show that the novel two-body sensor system and the associated algorithms were able to isolate infant-generated movements during physical contact with the caregiver. This approach holds promise for the automated long-term tracking of spontaneous movements in infants, which may provide critical insight into developmental disorders.
自发运动是指在没有任何外部刺激的情况下肢体的重复性运动，已被发现可反映婴儿期的神经发育状况。这些运动受个体和环境因素的调节，包括与照顾者的身体接触（怀抱）。然而，在身体接触期间测量自发运动具有挑战性，因为在这种情况下婴儿产生的运动与照顾者产生的运动相互耦合。在此，我们提出使用一种新型的双体传感器系统，以区分在与照顾者身体接触时婴儿产生的运动。在不同的模拟家庭活动中记录了七名发育正常的婴儿及其照顾者的数据，这些活动涉及身体互动、照顾者运动和婴儿姿势的不同组合。双体传感器系统由两个可穿戴加速度计组成——一个放在婴儿手臂上，一个放在照顾者手臂上，并且我们开发了一种基于卡尔曼滤波器的算法来分离婴儿产生的运动。此外，还录制了视频用于定性分析。结果表明，在没有与照顾者身体接触时，自发运动活动更高。当有身体接触时，照顾者静止且婴儿被水平怀抱时，自发运动会增加。这些结果表明，新型双体传感器系统及相关算法能够在与照顾者身体接触期间分离出婴儿产生的运动。这种方法有望用于对婴儿自发运动的自动化长期跟踪，这可能为发育障碍提供关键见解。
