Evaluation of Fidgety Movements of Infants Based on Gestalt Perception Reflects Differences in Limb Movement Trajectory Curvature

Intelligence Sparse Sensor Network for Automatic Early Evaluation of General Movements in Infants

General movements (GMs) have been widely used for the early clinical evaluation of infant brain development, allowing immediate evaluation of potential development disorders and timely rehabilitation. The infants' general movements can be captured digitally, but the lack of quantitative assessment and well-trained clinical pediatricians presents an obstacle for many years to achieve wider deployment, especially in low-resource settings. There is a high potential to explore wearable sensors for movement analysis due to outstanding privacy, low cost, and easy-to-use features. This work presents a sparse sensor network with soft wireless IMU devices (SWDs) for automatic early evaluation of general movements in infants. The sparse network consisting of only five sensor nodes (SWDs) with robust mechanical properties and excellent biocompatibility continuously and stably captures full-body motion data. The proof-of-the-concept clinical testing with 23 infants showcases outstanding performance in recognizing neonatal activities, confirming the reliability of the system. Taken together with a tiny machine learning algorithm, the system can automatically identify risky infants based on the GMs, with an accuracy of up to 100% (99.9%). The wearable sparse sensor network with an artificial intelligence-based algorithm facilitates intelligent evaluation of infant brain development and early diagnosis of development disorders.

Relationship between fidgety movement and frequency of movement toward midline: An observational study

BACKGROUND

Infants show other movements and posture patterns during the fidgety movement period, including movement toward midline (MTM). Few studies have quantified MTM occurring during the fidgety movement period.

AIMS

This study aimed to examine the relationship between fidgety movements (FMs) and MTM frequency and occurrence rate per minute, from two video data sets (video attached to Prechtl video manual and accuracy data from Japan).

STUDY DESIGN

Observational study.

SUBJECTS

It encompassed 47 videos. Of these, 32 were deemed normal FMs. The study amalgamated FMs that were sporadic, abnormal, or absent into a category of aberrant (n = 15).

OUTCOME MEASURES

Infant video data were observed. MTM item occurrences were recorded and calculated for occurrence percentage and MTM rate of occurrence per minute. The differences between groups for the upper limbs, lower limbs, and total MTM were statistically analysed.

RESULTS

Twenty-three infant videos of normal FMs and seven infant videos of aberrant FMs showed MTM. Eight infant videos of aberrant FMs showed no MTM, and only four with absent FMs were included. There was a significant difference in the total MTM rate of occurrence per minute between normal FMs versus aberrant FMs (p = 0.008).

CONCLUSIONS

This study presented MTM frequency and rate of occurrence per minute in infants who showed FMs during the fidgety movement period. Those who showed absent FMs also demonstrated no MTM. Further study may need a larger sample size of absent FMs and information on later development.

Towards human-level performance on automatic pose estimation of infant spontaneous movements

Assessment of spontaneous movements can predict the long-term developmental disorders in high-risk infants. In order to develop algorithms for automated prediction of later disorders, highly precise localization of segments and joints by infant pose estimation is required. Four types of convolutional neural networks were trained and evaluated on a novel infant pose dataset, covering the large variation in 1424 videos from a clinical international community. The localization performance of the networks was evaluated as the deviation between the estimated keypoint positions and human expert annotations. The computational efficiency was also assessed to determine the feasibility of the neural networks in clinical practice. The best performing neural network had a similar localization error to the inter-rater spread of human expert annotations, while still operating efficiently. Overall, the results of our study show that pose estimation of infant spontaneous movements has a great potential to support research initiatives on early detection of developmental disorders in children with perinatal brain injuries by quantifying infant movements from video recordings with human-level performance.

Infant's Behaviour Checklist for low birth weight infants and later neurodevelopmental outcome

Assessment of the characteristics of spontaneous movements and behaviour in early infancy helps in estimating developmental outcomes. We introduced the Infant Behaviour Checklist (IBC) and examined the relationship between the behavioural characteristics of low-birth-weight infants and neurodevelopmental outcomes at 6 years of age. The behavioural characteristics during the neonatal (36-43 weeks, adjusted) and early infancy periods (49-60 weeks, adjusted) were assessed in very-low-birth-weight infants. The IBC includes 44 common behaviours. We assessed the appearance of individual behavioural characteristics at each period according to the neurodevelopmental outcome. Of the 143 infants assessed during the neonatal period, 89 had typical development (TD), 30 had intellectual disability (ID), and 24 had autism spectrum disorder (ASD). In 78 infants assessed during early infancy, 40, 21, and 17 had TD, ID, and ASD, respectively. The frequency of appearance of three behaviour-related items was significantly lower in the ID group than in the TD group. The frequency of appearance of three posture- and behaviour-related items was significantly lower, while that of two posture-related items was significantly higher, in the ASD group than in the TD group. Behavioural assessment using the IBC may provide promising clues when considering early intervention for low-birth-weight infants.

Atypical general movements in the general population: Prevalence over the last 15 years and associated factors

AIM

To determine the prevalence of atypical general movements (GMs) in the general population, to examine its time trend and associated factors.

METHODS

Participants consisted of 300 infants born in 2016-2018 (current cohort; gestational age 39.4 weeks (27-42); 162 boys), representative of the Dutch population. GMs were assessed at 2-4 months corrected age in terms of GM-complexity (definitely abnormal (DA) or not) and fidgety movements (present or absent). GM-complexity data from a cohort of 455 Dutch infants born in 2001-2002 were used to investigate the time trend.

RESULTS

In the current cohort, 10 infants (3%) showed DA GM-complexity and 8 (3%) absent fidgety movements. Only one infant had both GM-impairments (0.3%). The prevalence of DA GM-complexity did not differ from that in the 2001-2002 cohort (adjusted odds ratio (OR) = 1.47 [0.53, 4.06]). DA GM-complexity was associated with maternal smoking (adjusted OR = 3.59 [1.56, 8.28]) and marginally with prematurity (adjusted OR = 2.78 [1.00, 7.74]); absence of fidgety movements was curvilinearly associated with assessment age only (OR = 1.06 [1.01, 1.12]).

CONCLUSION

In the general population, the prevalence of DA GM-complexity and absent fidgety movements is 3%. The finding that they rarely co-occur and are associated with different factors indicates that GM-assessment needs to address both aspects.

General Movement Assessment from videos of computed 3D infant body models is equally effective compared to conventional RGB video rating

BACKGROUND

General Movement Assessment (GMA) is a powerful tool to predict Cerebral Palsy (CP). Yet, GMA requires substantial training challenging its broad implementation in clinical routine. This inspired a world-wide quest for automated GMA.

AIMS

To test whether a low-cost, marker-less system for three-dimensional motion capture from RGB depth sequences using a whole body infant model may serve as the basis for automated GMA.

STUDY DESIGN

Clinical case study at an academic neurodevelopmental outpatient clinic.

SUBJECTS

Twenty-nine high risk infants were assessed at their clinical follow-up at 2-4 month corrected age (CA). Their neurodevelopmental outcome was assessed regularly up to 12-31 months CA.

OUTCOME MEASURES

GMA according to Hadders-Algra by a masked GMA-expert of conventional and computed 3D body model ("SMIL motion") videos of the same GMs. Agreement between both GMAs was tested using dichotomous and graded scaling with Kappa and intraclass correlations, respectively. Sensitivity and specificity to predict CP at ≥12 months CA were assessed.

RESULTS

Agreement of the two GMA ratings was moderate-good for GM-complexity (κ = 0.58; ICC = 0.874 [95%CI 0.730; 0.941]) and substantial-good for fidgety movements (FMs; Kappa = 0.78, ICC = 0.926 [95%CI 0.843; 0.965]). Five children were diagnosed with CP (four bilateral, one unilateral CP). The GMs of the child with unilateral CP were twice rated as mildly abnormal with FMs. GM-complexity and somewhat less FMs, of both conventional and SMIL motion videos predicted bilateral CP comparably to published literature.

CONCLUSIONS

Our computed infant 3D full body model is an attractive starting point for automated GMA in infants at risk of CP.

Wearable strain sensor suit for infants to measure limb movements under interaction with caregiver

Development of motion capture technology has enabled the measurement of body movements over long periods of time in daily life. Although accelerometers have been used as primary sensors, problems arise when they are used to measure the movements of infants. Because infants and caregivers interact frequently, accelerometer data from infants may be significantly distorted by a caregiver's movement. To overcome this problem, a strain sensor suit was developed for infants to measure flexion and extension movements of the limbs. A case study was performed to analyze the strain sensor data of an infant in relation to the accelerometer data of the infant's and a caregiver's body under various types of infant-caregiver interaction. The results demonstrated that the strain sensor data had low correlation with the accelerometer data of the infant and caregiver while the accelerometer data between infant and caregiver had higher correlation. This suggests that the strain sensor is suitable to detect limbs' angular displacements mostly independent from the translational body movements exerted by a caregiver.

The shape of disposable diaper affects spontaneous movements of lower limbs in young infants

This study examined the characteristics of young infants' lower limb spontaneous movements based on differences in shape of diapers. Twenty-seven healthy infants (103 ± 16.3 days old) were enrolled in this study. We measured the spontaneous movements of their lower limbs in four conditions (Naked, wearing Normal type diapers, wearing Type A diapers, and wearing Type B diapers). The Normal diaper has a wider waist belt than the Type A diaper, and the Type B diaper has a narrower crotch area than the Type A diaper. We observed them in seven indices (the velocity of lower limb movements, the trajectory area of knee movement in the sagittal plane and the frontal plane, the distance between both knees and between side of abdomen and knee, and correlation of velocities between side of abdomen and knee and between left and right ankles). The results showed that the velocity of the lower limb movements in the Naked condition was higher than when wearing Normal diapers. The value for the trajectory area of knee movement in sagittal plane, which reflects the range of lower leg lifting movements and closeness of such movements to the trunk, for the Type B diaper condition was higher than that for the Normal diaper condition. This result indicates that the shape of the diaper affects the spontaneous movements of the lower limbs of young infants.