Assessing Neonatal Intensive Care Unit (NICU) Graduates Across Varied Settings: A Study on the Feasibility of the Baby Moves App.

We assessed the feasibility of obtaining parent-collected General Movement Assessment videos using the Baby Moves app. Among 261 participants from 4 Chicago NICUs, 70% submitted videos. Families living in higher areas of childhood opportunity used the app more than those from areas of lower opportunity.

Cerebral oxygenation during immediate fetal-to-neonatal transition and fidgety movements between six to 20 weeks of corrected age: An ancillary study to the COSGOD III trial.

Fidgety movements provide early information about a potential development of cerebral palsy in preterm neonates. The aim was to assess differences in the combined outcome of mortality and fidgety movements defined as normal or pathological in very preterm neonates according to the group allocation in the randomised-controlled multicentre COSGOD III trial. Preterm neonates of two centres participating in the COSGOD III trial, whose fidgety movements were assessed as normal or pathological at six to 20 weeks of corrected age, were analysed. In the COSGOD III trial cerebral oxygen saturation (crSO2) was measured by near-infrared spectroscopy (NIRS) during postnatal transition and guided resuscitation in preterm neonates randomised to the NIRS-group, whereby medical support was according routine, as it was also in the control group. Fidgety movements were classified in normal or abnormal/absent at six to 20 weeks of corrected age. Mortality and fidgety movements of preterm neonates allocated to the NIRS-group were compared to the control-group. Normal outcome was defined as survival with normal fidgety movements. One-hundred-seventy-one preterm neonates were included (NIRS-group n = 82; control-group n = 89) with a median gestational age of 29.4 (27.4-30.4) and 28.7 (26.7-31.0) weeks in the NIRS-group and the control-group, respectively. There were no differences in the combined outcome between the two groups: 90.2% of the neonates in the NIRS-group and 89.9% in the control-group survived with normal outcome (relative risk [95% CI]; 0.96 [0.31-2.62]).Conclusions: In the present cohort of preterm neonates, monitoring of crSO2 and dedicated interventions in addition to routine care during transition period after birth did not show an impact on mortality and fidgety movements defined as normal or pathological at six to 20 weeks corrected age. What is Known • Fidgety movements display early spontaneous motoric pattern and may provide early information about a potential development of cerebral palsy in preterm neonates.  What is New  â€¢ This retrospective observational study of the randomised-controlled multicentre COSGOD III trial is the first study investigating the potential influence of cerebral oxygenation guided resuscitation during postnatal transition period on combined outcome of mortality and fidgety movements up to 20 weeks of corrected age in very preterm neonates. • This study adds to the growing interest of assessing cerebral oxygenation, that monitoring of cerebral oxygen saturation and dedicated interventions during postnatal transition period according to the COSGOD III trial has no significant influence on mortality and fidgety movements defined as normal or pathological in very preterm neonates.

Implementation of remote general movement assessment using the in-motion instructions in a high-risk norwegian cohort.

BACKGROUND: General Movement Assessment (GMA) is recommended for early detection of risk for cerebral palsy but requires trained clinical experts. We aimed to implement home- and hospital-based filming for remote GMA in a Norwegian high-risk infant cohort, as well as evaluating parents' experiences in filming their infant at home. METHODS: This knowledge translational study used a prospective cohort design including participants referred to neurodevelopmental follow-up across three sites in the Central Norway Regional Health Authority. Two home films of the fidgety type of general movements were collected between 12+1-14+6 and 15+1-17+6 weeks after term by parents. An additional film was collected at the hospital between 12+1 and 17+6 weeks after term. The instructional guide for all filming was the In-Motion App standards. Videos were transferred to a remote GMA team and classified as either "GMA scorable" or "GMA not scorable" based on Prechtl's GMA standards. Parents responded to an online survey using a 5-point Likert scale to collect information about their perspectives, experiences, and possible worries by filming their infant at home. RESULTS: One-hundred-and-two infants from 95 families participated. Ninety-two (96.8%) families transferred 177 home-based videos. Eighty-four (92%) of these had 95 videos taken in their local hospital. All 177 home-videos were "GMA scorable" and three (3,1%) out of 95 hospital-based videos were classified as "GMA not scorable". Eight families did not respond to the survey and two families did not receive the survey due to a technical error. Seventy-eight (91.7%) respondents agreed or strongly agreed that it was easy to perform home filming and five (5.9%) agreed that they were more worried about their child`s development after filming at home. Almost 80% of respondents agreed that a video for GMA can be taken at home instead of in hospital. CONCLUSIONS: This study strengthens the clinical implementation of home filming by parents and remote GMA for early detection of CP in high-risk follow-up programs. The implementation of remote GMA has the potential to facilitate early intervention to improve function in children with CP in line with international recommendations. TRIAL REGISTRATION: ClinicalTrials.gov ID: NCT04287166 Date of registration: 27/02/2020.

Reliability of the Motor Optimality Score-Revised: A study of infants at elevated likelihood for adverse neurological outcomes.

AIM: To assess the inter-assessor reliability of the Motor Optimality Score-Revised (MOS-R) when used in infants at elevated likelihood for adverse neurological outcome. METHODS: MOS-R were assessed in three groups of infants by two assessors/cohort. Infants were recruited from longitudinal projects in Sweden (infants born extremely preterm), India (infants born in low-resource communities) and the USA (infants prenatally exposed to SARS-CoV-2). Intraclass correlation coefficients (ICC) and kappa (κw) were applied. ICC of MOS-R subcategories and total scores were presented for cohorts together and separately and for age-spans: 9-12, 13-16 and 17-25-weeks post-term age. RESULTS: 252 infants were included (born extremely preterm n = 97, born in low-resource communities n = 97, prenatally SARS-CoV-2 exposed n = 58). Reliability of the total MOS-R was almost perfect (ICC: 0.98-0.99) for all cohorts, together and separately. Similar result was found for age-spans (ICC: 0.98-0.99). Substantial to perfect reliability was shown for the MOS-R subcategories (κw: 0.67-1.00), with postural patterns showing the lowest value 0.67. CONCLUSION: The MOS-R can be used in high-risk populations with substantial to perfect reliability, both in regards of total/subcategory scores as well as in different age groups. However, the subcategory postural patterns as well as the clinical applicability of the MOS-R needs further study.

Concurrent and predictive validity of the infant motor profile in infants at risk of neurodevelopmental disorders.

BACKGROUND: Preterm infants and infants with perinatal brain injury show a higher incidence of neurodevelopmental disorders (NDD). The Infant Motor Profile (IMP) is a clinical assessment which evaluates the complexity of early motor behaviour. More data are needed to confirm its predictive ability and concurrent validity with other common and valid assessments such as the Alberta Infant Motor Scale (AIMS) and Prechtl's General Movement Assessment (GMA). The present study aims to evaluate the concurrent validity of the IMP with the AIMS, to assess its association with the GMA, to evaluate how the IMP reflects the severity of the brain injury and to compare the ability of the IMP and the AIMS to predict an abnormal outcome in 5-month-old infants at risk of NDD. METHODS: 86 infants at risk of NDD were retrospectively recruited among the participants of two clinical trials. Preterm infants with or without perinatal brain injury and term infants with brain injury were assessed at 3 months corrected age (CA) using the GMA and at 5 months CA using the IMP and the AIMS. The neurodevelopmental outcome was established at 18 months. RESULTS: Results confirm a solid concurrent validity between the IMP Total Score and the AIMS (Spearman's ρ 0.76; p < .001) and a significant association between IMP Total Score and the GMA. Unlike the AIMS, the IMP Total score accurately reflects the severity of neonatal brain injury (p < .001) and proves to be the strongest predictor of NDD (p < .001). The comparison of areas under receiver operating characteristic curves (AUC) confirms that the IMP Total score has the highest diagnostic accuracy at 5 months (AUC 0.92). For an optimal IMP Total Score cut-off value of 70, the assessment shows high sensitivity (93%) and specificity (81%) (PPV 84%; NPV 90%). CONCLUSIONS: Early motor behaviour assessed with the IMP is strongly associated with middle-term neurodevelopmental outcome. The present study confirms the concurrent validity of the IMP with the AIMS, its association with the GMA and its ability to reflect brain lesion load, hence contributing to the construct validity of the assessment. TRIAL REGISTRATION: NCT01990183 and NCT03234959 (clinicaltrials.gov).

Writhing Movement Detection in Newborns on the Second and Third Day of Life Using Pose-Based Feature Machine Learning Classification.

Observation of neuromotor development at an early stage of an infant's life allows for early diagnosis of deficits and the beginning of the therapeutic process. General movement assessment is a method of spontaneous movement observation, which is the foundation for contemporary attempts at objectification and computer-aided diagnosis based on video recordings' analysis. The present study attempts to automatically detect writhing movements, one of the normal general movement categories presented by newborns in the first weeks of life. A set of 31 recordings of newborns on the second and third day of life was divided by five experts into videos containing writhing movements (with occurrence time) and poor repertoire, characterized by a lower quality of movement in relation to the norm. Novel, objective pose-based features describing the scope, nature, and location of each limb's movement are proposed. Three machine learning algorithms are evaluated in writhing movements' detection in leave-one-out cross-validation for different feature extraction time windows and overlapping time. The experimental results make it possible to indicate the optimal parameters for which 80% accuracy was achieved. Based on automatically detected writhing movement percent in the video, infant movements are classified as writhing movements or poor repertoire with an area under the ROC (receiver operating characteristics) curve of 0.83.

AI Approaches Towards Prechtl's Assessment of General Movements: A Systematic Literature Review.

General movements (GMs) are spontaneous movements of infants up to five months post-term involving the whole body varying in sequence, speed, and amplitude. The assessment of GMs has shown its importance for identifying infants at risk for neuromotor deficits, especially for the detection of cerebral palsy. As the assessment is based on videos of the infant that are rated by trained professionals, the method is time-consuming and expensive. Therefore, approaches based on Artificial Intelligence have gained significantly increased attention in the last years. In this article, we systematically analyze and discuss the main design features of all existing technological approaches seeking to transfer the Prechtl's assessment of general movements from an individual visual perception to computer-based analysis. After identifying their shared shortcomings, we explain the methodological reasons for their limited practical performance and classification rates. As a conclusion of our literature study, we conceptually propose a methodological solution to the defined problem based on the groundbreaking innovation in the area of Deep Learning.

Formal training in general movement assessment is required to effectively evaluate infants with perinatal asphyxia in outpatient settings.

AIM: General movement assessment (GMA) can help to identify children with a high risk of developing neurological dysfunction, such as cerebral palsy, and certified training is provided in this specialism. The aim of this study was to investigate the feasibility and reliability of using video recordings to assess GMA, in a busy Danish outpatient clinic. METHODS: The study comprised 30-term infants born with perinatal asphyxia, who were video recorded at three months. They were assessed by two certified GMA observers and re-assessed two weeks later. Interobserver and intra-observer agreements were analysed using proportional agreement, and nominal kappa statistics were used to calculate 95% confidence intervals (95% CI). RESULTS: We found substantial and almost perfect interobserver and intra-observer reliability. Intra-observer agreement was 0.85 (95% CI: 0.65-1.00; p < 0.0001) and 0.85 (95% CI: 0.62-1.00; p < 0.0001), and interobserver agreement was 0.71 (95% CI: 0.45-0.96; p < 0.0001) at time point one and 0.85 (95% CI: 0.63-1.00; p < 0.0001) two weeks later. All video recordings were completed within our multidisciplinary outpatient clinic without delay. CONCLUSION: This study demonstrated the reliability of the GMA method in a busy multidisciplinary Danish paediatric outpatient setting, when assessors had been formally trained in the method and used it regularly.

A self-supervised spatio-temporal attention network for video-based 3D infant pose estimation.

General movement and pose assessment of infants is crucial for the early detection of cerebral palsy (CP). Nevertheless, most human pose estimation methods, in 2D or 3D, focus on adults due to the lack of large datasets and pose annotations on infants. To solve these problems, here we present a model known as YOLO-infantPose, which has been fine-tuned, for infant pose estimation in 2D. We further propose a self-supervised model called STAPose3D for 3D infant pose estimation based on videos. We employ multi-view video data during the training process as a strategy to address the challenge posed by the absence of 3D pose annotations. STAPose3D combines temporal convolution, temporal attention, and graph attention to jointly learn spatio-temporal features of infant pose. Our methods are summarized into two stages: applying YOLO-infantPose on input videos, followed by lifting these 2D poses along with respective confidences for every joint to 3D. The employment of the best-performing 2D detector in the first stage significantly improves the precision of 3D pose estimation. We reveal that fine-tuned YOLO-infantPose outperforms other models tested on our clinical dataset as well as two public datasets MINI-RGBD and YouTube-Infant dataset. Results from our infant movement video dataset demonstrate that STAPose3D effectively comprehends the spatio-temporal features among different views and significantly improves the performance of 3D infant pose estimation in videos. Finally, we explore the clinical application of our method for general movement assessment (GMA) in a clinical dataset annotated as normal writhing movements or abnormal monotonic movements according to the GMA standards. We show that the 3D pose estimation results produced by our STAPose3D model significantly boost the GMA prediction performance than 2D pose estimation. Our code is available at github.com/wwYinYin/STAPose3D.

Development of Movement and Postural Patterns in Full-Term Infants Who Are at Low Risk in Belgium, India, Norway, and the United States.

OBJECTIVE: The objective of this study was to investigate the influences on motor development in infants who are at low risk from Belgium, India, Norway, and the United States (US) using the General Movement Assessment (GMA) at 10-16 weeks post-term age. METHODS: This was a cross-sectional study of prospectively enrolled full-term infants at low risk (n = 186). Certified GMA observers rated the fidgety movements, quality of the movement patterns, age-adequacy of the movement repertoire, postural patterns, movement character, and overall Motor Optimality Score - Revised (MOS-R). Scores were evaluated for associations with sex, birth weight category, gestational age, post-term age at video, and country. RESULTS: The majority of infants had normal fidgety movements (179/186, 96.2%). This did not vary by sex, birth weight, gestational age, post-term age at video, or country. All infants showed normal>atypical movement patterns. Variability was seen for age adequacy (optimal: 137/183, 74.9%), postural patterns (normal>atypical: 164/183, 89.6%), and smooth/ fluent movement character (138/183, 75.4%). Gestational age and post-term age at video were associated with atypical postural patterns but in multivariable regression, only younger post-term age retained significance (OR 2.94, 95% CI: 1.05-8.24). Lack of age adequacy was associated with post-term age (OR 13.15, 95% CI: 4.36-39.72), and country (compared with Norway; Belgium OR 3.38 95% CI:12.4-9.22; India OR 3.16, 95% CI:1.01-9.87: US not significant). Infants from India also showed lower rates of an optimal MOS-R (25-28) than infants from Norway. CONCLUSIONS: The normality and temporal organization of fidgety movements did not differ by sex, birth weight, post-term age, or country, suggesting that the fidgety movements are free of cultural and environmental influences. The majority of full-term infants who were healthy in this cohort showed normal scores for all aspects of motor development tested using the MOS-R. Differences in age adequacy and MOS-R by country warrant investigation with larger cohorts and longitudinal follow up. IMPACT STATEMENT: Understanding variations in typical motor development is essential to interpreting patterns of movement and posture in infants at risk for atypical development. Using the framework of Prechtl's General Movements Assessment, this study showed that the development of movement and posture in healthy infants were affected by age and country of birth, but the development of the fidgety movements appeared to be free of these influences. Local norms may be needed to interpret the Motor Optimality Score-Revised in all populations but further research on this topic is needed.

Stability of the Motor Optimality Score Revised (MOS-R) in medically complex infants.

BACKGROUND: The Motor Optimality Score-Revised (MOS-R) is a clinical test of infant spontaneous movement at 3-5 months of age and has been associated with neurodevelopmental outcomes in infants with medical complexity. However the stability of the MOS-R tested at different developmental ages is not yet known. AIM: To determine if the repeated scoring of the MOS-R remained consistent when tested at two developmental ages in young infants. STUDY DESIGN: In this prospective cohort study, infants were tested twice with the MOS-R between 12 and 13 weeks corrected age (CA) and 14-16 weeks CA. Bland Altman Plots were used to calculate agreement between the scores. Infants were grouped as having higher or lower medical complexity. MOS-R threshold scores were analyzed to assess changes over time within each group using Fisher's exact test. SUBJECTS: 85 infants with history of hospitalization in a neonatal intensive care unit (NICU). RESULTS: The MOS-R scores had a high agreement with negligible bias (0.058) between timepoints (95 % CI -1.10, 1.22). Using a MOS-R cut point of 19, infants with higher medical complexity were more likely to change groups between timepoints than infants with lower medical complexity (p = 0.008), but this was not significantly different using cut points of 20 or 21. CONCLUSION: The MOS-R scores were stable when measured repeatedly in infants who were hospitalized in a NICU. Infants with high medical complexity had less stable MOS-R scores using certain cut points than infants with lower medical complexity.

The role of fidgety movements and early motor repertoire in predicting mobility outcomes in infants with myelomeningocele.

OBJECTIVE: To describe fidgety movements and co-occurring movements and postures in infants with myelomeningocele (MMC) and their association with mobility at preschool ages. METHODS: A retrospective cohort with early assessment via general movement assessment, followed by mobility assessment between 36 and 70 months of age. RESULTS: Twelve infants were included; 12 of 12 had fidgety movements in the upper limbs, with seven exhibiting them also in the hips and three in both the hips and ankles. The presence of fidgety movements in the lower limbs, kicking, a non-flat posture, a non-monotonous movement character, and a non-absent age-adequate movement repertoire were independently associated with mobility using the Hoffer modified classification and functional mobility scale (FMS) at 5 and 50 m. An optimality score was calculated based on leg movements and postures, ranging from 0 to 10 points. Infants who scored at least 4 points achieved household ambulation and FMS (5 m) of at least level 4. Community ambulation and an FMS (50 m) of level 5 were achieved with a score of at least 7.5. CONCLUSIONS: Assessing fidgety movements with other leg movements and postures in infants with MMC provided relevant information that could potentially predict mobility at preschool age and thus could be used for early intervention planning.

Inter-rater reliability and agreement of the General Movement Assessment and Motor Optimality Score-Revised in a large population-based sample.

BACKGROUND: Prechtl's General Movement Assessment (GMA) at fidgety age (3-5 months) is a widely used tool for early detection of cerebral palsy. Further to GMA classification, detailed assessment of movement patterns at fidgety age is conducted with the Motor Optimality Score-Revised (MOS-R). Inter-rater reliability and agreement are properties that inform test application and interpretation in clinical and research settings. This study aims to establish the inter-rater reliability and agreement of the GMA classification and MOS-R in a large population-based sample. METHODS: A cross-sectional study of 773 infants from birth-cohort in Perth, Western Australia. GMA was conducted on home-recorded videos collected between 12 + 0 and 16 + 6 weeks post term age. Videos were independently scored by two masked experienced assessors. Inter-rater reliability and agreement were assessed using intraclass correlation coefficient and limits of agreement respectively for continuous variables, and Cohen's Kappa and Gwet's Agreement Coefficient, and percentage agreement respectively for discrete variables. RESULTS: The classification of GMA showed almost perfect reliability (AC1 = 0.999) and agreement (99.9 %). Total MOS-R scores showed good-excellent reliability (ICC 0.857, 95 % CI 0.838-0.876) and clinically acceptable agreement (95 % limits of agreement of ±2.5 points). Substantial to almost perfect reliability and agreement were found for all MOS-R domain subscores. While MOS-R domains with higher redundancy in their categorisation have higher reliability and agreement, inter-rater reliability and agreement are substantial to almost perfect at the item level and are consistent across domains. CONCLUSION: GMA at fidgety age shows clinically acceptable inter-rater reliability and agreement for GMA classification and MOS-R for population-based cohorts assessed by experienced assessors.

A Study of General Movement Assessment and Its Association With Neurodevelopmental Outcome at Age 12 to 15 Months Among Term Neonates With Hyperbilirubinemia.

BACKGROUND: To determine the association between optimality score at term age and age three to five months and neurodevelopmental outcome among neonates with hyperbilirubinemia. METHODS: Fifty infants with and without hyperbilirubinemia were enrolled. The motor repertoires of the infants were evaluated through general movement assessment (GMA) at term age and three to five months post-term. The association between the General Movement Optimality Score (GMOS), Motor Optimality Score (MOS), and Development Assessment Scale for Indian Infants (DASII) at age 12 to 15 months was also assessed. RESULTS: During term age, the median GMOS was significantly lower among infants in the study group when compared with the control group (40 [29 to 42] vs 42 [42 to 42], P < 0.001). However, at age three to five months, there was no significant difference between the groups. Significantly higher number of neonates had abnormal motor repertoire at term age and age three to five months in the study group when compared with the control group (18 [36%] vs 2 [4%], P = 0.001, at term age and 6 [12.2%] vs 1 [2%], P =0.04, at age three to five months). Among neonates with hyperbilirubinemia, the median GMOS and MOS were significantly lower at term age and age three to five months in infants with motor and mental developmental quotient scores <85 when compared with ≥85. CONCLUSIONS: GMA including GMOS and MOS performed in neonates with hyperbilirubinemia during the neonatal period and early infancy is associated with neurodevelopmental outcomes in the first year of life. GMA can help initiate early intervention in such neonates.

General movements in spinal muscular atrophy type 1.

PURPOSE: To investigate the motor repertoire of infants diagnosed with spinal muscular atrophy Type I (SMA Type I) without administration of any disease-modifying agent. METHODS: Motor Optimality Score-Revised (MOS-R) was calculated from videos recorded between post-term weeks 9-17 for 22 infants with SMA Type I. The MOS-R of infants with SMA Type I was compared with those of 22 infants with cerebral palsy (CP) and 22 infants with typical development. RESULTS: Of the infants with SMA Type I, 17 had absent fidgety movements (FMs) and 5 had sporadic FMs. Age adequate movement repertoire was absent, and the variety of movements in infants was very low. Movements were symmetrical but movements of four limbs remained on the surface level. Antigravity movements were very rare. Movement characterization was monotonous, slow speed, and small amplitude. The MOS-R of infants with SMA Type I was lower than those of infants with typical development but similar to those of infants with CP. CONCLUSIONS: Infants with SMA Type I had a motor repertoire similar to infants with CP, while they had a poorer motor repertoire than infants with typical development in the fidgety period as evidenced by MOS-R. Central nervous system involvement in these infants with SMA Type I with absent FMs and reduced MOS-R is unknown. Further studies are needed to determine the role of problems in the afferent and efferent pathways of spinal cord and muscle atrophy in the observation of normal FMs.

Systemic postnatal corticosteroid use for the prevention of bronchopulmonary dysplasia and its relationship to early neurodevelopment in extremely preterm infants.

BACKGROUND: Systemic postnatal corticosteroid use in extremely preterm infants poses a risk of adverse neurodevelopmental outcomes. This study explores their use beyond seven days of age with early neurodevelopmental assessments during the fidgety period (9-20 weeks postterm age). METHODS: This retrospective single-center cohort study included inborn extremely preterm infants from 1 January 2014 to 31 December 2018. Outborn infants, those with congenital or genetic abnormalities, and those who received postnatal corticosteroids for nonrespiratory reasons were excluded. The cohort was dichotomized based on the status of corticosteroid receipt. Early neurodevelopmental outcomes were reported using Prechtl's General Movements Assessment. RESULTS: Of the 282 infants, 67 (23.75%) received corticosteroids. Of these, 34 (50.75%) received them for dependency on invasive ventilation (intermittent positive-pressure ventilation), and the remainder received them for dependency on non-invasive ventilation continuous positive airway pressure (CPAP) or bi-level positive airway pressure (BiPAP). Abnormal or absent fidgety movements were observed in 13% of infants (7/54) who received corticosteroids compared to 2% of infants (3/146) who did not. An increased odds for an abnormal general movements assessment from corticosteroid use after adjusting for gestational age [adjusted odds ratio (aOR) = 5.5, 95% confidence interval (CI) = 1.14-26.56] was observed. The motor optimality scores differed between the two groups [corticosteroid group: 25.5 (23-26) versus no-corticosteroid group: 26 (24-28); z = - 2.02]. A motor optimality score < 20 was observed in 14.8% of infants (8/54) in the corticosteroid group compared to 2% of infants (3/146) in the noncorticosteroid group. This difference was significant after adjustment for gestational age (aOR 5.96, 95% CI 1.28-27.74). CONCLUSIONS: Abnormal early neurodevelopment was observed in infants who received systemic postnatal corticosteroids. The relationship between these findings and other factors influencing early neurodevelopment needs further exploration.

Semi-supervised body parsing and pose estimation for enhancing infant general movement assessment.

General movement assessment (GMA) of infant movement videos (IMVs) is an effective method for early detection of cerebral palsy (CP) in infants. We demonstrate in this paper that end-to-end trainable neural networks for image sequence recognition can be applied to achieve good results in GMA, and more importantly, augmenting raw video with infant body parsing and pose estimation information can significantly improve performance. To solve the problem of efficiently utilizing partially labeled IMVs for body parsing, we propose a semi-supervised model, termed SiamParseNet (SPN), which consists of two branches, one for intra-frame body parts segmentation and another for inter-frame label propagation. During training, the two branches are jointly trained by alternating between using input pairs of only labeled frames and input of both labeled and unlabeled frames. We also investigate training data augmentation by proposing a factorized video generative adversarial network (FVGAN) to synthesize novel labeled frames for training. FVGAN decouples foreground and background generation which allows for generating multiple labeled frames from one real labeled frame. When testing, we employ a multi-source inference mechanism, where the final result for a test frame is either obtained via the segmentation branch or via propagation from a nearby key frame. We conduct extensive experiments for body parsing using SPN on two infant movement video datasets; on these partially labeled IMVs, we show that SPN coupled with FVGAN achieves state-of-the-art performance. We further demonstrate that our proposed SPN can be easily adapted to the infant pose estimation task with superior performance. Last but not least, we explore the clinical application of our method for GMA. We collected a new clinical IMV dataset with GMA annotations, and our experiments show that our SPN models for body parsing and pose estimation trained on the first two datasets generalize well to the new clinical dataset and their results can significantly boost the convolutional recurrent neural network (CRNN) based GMA prediction performance when combined with raw video inputs.

The Relationship between General Movements and Risk Factors in Moderate-Late Preterm Infants: A Prospective Cohort Study.

INTRODUCTION: Moderate-late preterm infants constitute the largest segment of preterm births globally. While previously considered to have a low neurological risk, recent research has uncovered an elevated incidence of neurodevelopmental conditions in this group. This study aimed to assess the relationship between the general movement assessment and birth-related risk factor-based tools in moderate-late preterm infants. METHODS: A prospective cohort study of 65 moderate-late preterm infants in a neonatal intensive care unit involved the evaluation of general movements, the Nursery Neurobiologic Risk Score, and the Perinatal Risk Inventory. Associations were analyzed using Fisher's exact test, Spearman's correlation was used for ordinal variables, and backward stepwise logistic regression was used to identify predictor variables for the assessments. RESULTS: The findings indicated a high prevalence of normal (41%) and poor (52%) repertoire patterns during the writhing period. While no significant associations were found between the three assessments, a slight approximation emerged between dysmorphic traits and patterns (p = 0.053). Furthermore, an extended period of ventilation correlated with a higher likelihood of developing a cramped synchronized pattern and there was a correlation between both risk factor-based tools (p < 0.001). CONCLUSIONS: This research enhances our understanding of the early impact on general movement assessments in moderate-late preterm infants. While no clear relationship emerged between general movement assessment and risk factor-based tools, there was a subtle connection noted with dysmorphic traits. A longer ventilation duration was linked to a higher risk of developing cramped synchronized patterns.

Feeding dysfunction in NICU patients with cramped synchronized movements.

Patients admitted to the neonatal intensive care unit (NICU) have higher association for neurodevelopment deficits, specifically cerebral palsy (CP). We identified patients with risk for CP using abnormal Pretchl's General Movement Assessment (GMA) and sub-category of cramped synchronized movements (CSM) and reported their feeding outcomes at discharge. Over 75 % of these patients required either nasogastric (NGT) or gastrostomy tube (GT) at discharge. Of these, 57 % weaned off their NGT or GT at home and 43 % of patients still needed a GT one year after discharge. Of those that could not wean off their NGT or GT, these patients had longer hospital stay, took lower percentage by mouth, and an older post-menstrual age at discharge. We did not find a difference in NGT or GT use between patients with IVH, ELBW, nor between their birthweight or gestation age at birth. This study provides further clinical characteristics in NICU patients who have higher risk of CP, and supports the need for skilled feeding therapy and resources both during and after NICU admission.