Jane P. Valentine

IEEE journal of biomedical and health informatics • October 03, 2024

Romero Morais, Truyen Tran, Caroline Alexander, Natasha Amery, Catherine Morgan, Alicia Spittle, Vuong Le, Nadia Badawi, Alison Salt, Jane Valentine, Catherine Elliott, Elizabeth Hurrion, Paul Dawson, Svetha Venkatesh

Typically developing infants, between the corrected age of 9-20 weeks, produce fidgety movements. These movements can be identified with the General Movement Assessment, but their identification requires trained professionals to conduct the assessment from video recordings. Since trained professionals are expensive and their demand may be higher than their availability, computer vision-based solutions have been developed to assist practitioners. However, most solutions to date treat the problem as a direct mapping from video to infant status, without modeling fidgety movements throughout the video. To address that, we propose to directly model infants' short movements and classify them as fidgety or non-fidgety. In this way, we model the explanatory factor behind the infant's status and improve model interpretability. The issue with our proposal is that labels for an infant's short movements are not available, which precludes us to train such a model. We overcome this issue with active learning. Active learning is a framework that minimizes the amount of labeled data required to train a model, by only labeling examples that are considered "informative" to the model. The assumption is that a model trained on informative examples reaches a higher performance level than a model trained with randomly selected examples. We validate our framework by modeling the movements of infants' hips on two representative cohorts: typically developing and at-risk infants. Our results show that active learning is suitable to our problem and that it works adequately even when the models are trained with labels provided by a novice annotator.

Child: Care, Health And Development • November 03, 2024

Sue-anne Davidson, Ashleigh Thornton, Deborah Hersh, Courtenay Harris, Catherine Elliott, Jane Valentine

Background: Early detection of cerebral palsy (CP) risk is possible from 12 weeks corrected gestational age (CGA) using standardised assessments; however, up to half of children at risk are not referred early, missing out on early intervention. We investigated the barriers and facilitators to accessing early intervention from the perspective of parents of children who did not receive services by 6 months CGA. Methods: Parents of children with CP were invited to participate in qualitative semistructured interviews. Reflexive thematic analysis was used to analyse the data and develop themes. Results: Eight mothers of children who did not receive standardised screening participated in interviews, from which three themes, 'responding to delays', 'systemic barriers' and 'complexities of diagnosis', were developed from the data. Conclusions: Parents require more support to access and engage in early detection services; health system processes are difficult to navigate, and health professionals require education and training to recognise risk factors for CP in all health settings and refer promptly. Improving system processes, education and training and partnering early with parents to improve their experience when interacting with the health system may increase early engagement and optimise long-term outcomes for children at risk of CP and their families.

The Journal Of Pediatrics • July 01, 2024

Roslyn Boyd, Susan Greaves, Jenny Ziviani, Iona Novak, Nadia Badawi, Kerstin Pannek, Catherine Elliott, Margaret Wallen, Catherine Morgan, Jane Valentine, Lisa Findlay, Andrea Guzzetta, Koa Whittingham, Robert Ware, Simona Fiori, Nathalie Maitre, Jill Heathcock, Kimberley Scott, Ann-christin Eliasson, Leanne Sakzewski

Objective: To compare efficacy of constraint-induced movement therapy (Baby-CIMT) with bimanual therapy (Baby-BIM) in infants at high risk of unilateral cerebral palsy. Methods: This was a single-blind, randomized-comparison-trial that had the following inclusion criteria: (1) asymmetric brain lesion (2) absent fidgety General Movements, (3) Hammersmith Infant Neurological Examination below cerebral palsy cut-points, (4) entry at 3-9 months of corrected age, and (5) >3-point difference between hands on Hand Assessment Infants (HAI). Infants were randomized to Baby-CIMT or Baby-BIM, which comprised 6-9 months of home-based intervention. Daily dose varied from 20 to 40 minutes according to age (total 70-89.2 hours). Primary outcome measure was the HAI after intervention, with secondary outcomes Mini-Assisting Hand Assessment and Bayley III cognition at 24 months of corrected age. Results: In total, 96 infants (51 male, 52 right hemiplegia) born median at 37-weeks of gestation were randomized to Baby-CIMT (n = 46) or Baby-BIM (n = 50) and commenced intervention at a mean 6.5 (SD 1.6) months corrected age. There were no between group differences immediately after intervention on HAI (mean difference [MD] 0.98 HAI units, 95% CI 0.94-2.91; P = .31). Both groups demonstrated significant clinically important improvements from baseline to after intervention (Baby-BIM MD 3.48, 95% CI 2.09-4.87; Baby-CIMT MD 4.42, 95% CI 3.07-5.77). At 24 months, 64 infants were diagnosed with unilateral cerebral palsy (35 Baby-CIMT, 29 Baby-BIM). Infants who entered the study between 3 and 6 months of corrected age had greater change in HAI Both Hands Sum Score compared with those who entered at ≥6 months of corrected age (MD 7.17, 95% CI 2.93-11.41, P = .001). Conclusions: Baby-CIMT was not superior to Baby-BIM, and both interventions improved hand development. Infants commencing intervention at <6 months corrected age had greater improvements in hand function.

Annals Of Behavioral Medicine : A Publication Of The Society Of Behavioral Medicine • June 25, 2024

Hamsini Sivaramakrishnan, Elizabeth Davis, Lerato Obadimeji, Jane Valentine, Fiona Wood, Vinutha Shetty, Amy Finlay Jones

Background: Behavior change techniques (BCTs) have been extensively used in physical activity interventions for children, however, no systematic reviews have synthesized their effects. Objective: The present review aimed to identify the most promising BCTs used in physical activity interventions associated with (i) increased physical activity behavior and (ii) positive psychosocial outcomes in children with chronic conditions. Methods: A systematic search of 6 databases identified 61 articles as eligible for inclusion. Data, including BCTs, were extracted from these studies and analyzed descriptively. Due to the heterogeneity of interventions, chronic conditions, and outcome measures, a meta-analysis was not conducted. Results: Social support (unspecified), graded tasks, generalization of target behavior, and credible source were the most commonly reported and most promising (i.e., present in 2+ studies evidencing significant effects) BCTs across all studies. These BCTs were found to be especially relevant to improving psychosocial outcomes in the short- and long-term and improving physical activity behaviors in the long-term. Meanwhile, to improve short-term physical activity behaviors, in addition to social support (unspecified), action planning, goal setting (behavior), and problem solving were found to be promising BCTs. Conclusions: The BCTs identified in this review may be relevant to incorporate when planning future interventions to support physical activity and psychosocial outcomes for children with chronic conditions.

Developmental Medicine And Child Neurology • March 28, 2024

Karen Twyford, Susan Taylor, Jane Valentine, Jonathan Pool, Annette Baron, Ashleigh Thornton

Objective: To determine the evidence for functional outcomes experienced by a population with paediatric neurodisability (such as acquired brain injury, cerebral palsy, spinal cord injury, and other neurological disorders), who access music therapy through neurorehabilitation services across the rehabilitation spectrum. Methods: Using scoping review methodology of the JBI and the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR), a systematic search was conducted across eight databases and expert knowledge users were consulted. Articles were screened by title and abstract, and data from eligible studies were categorized using the International Classification of Functioning, Disability and Health: Children and Youth version (ICF-CY). Results: From 1726 records identified, 53 eligible primary sources were included in the synthesis. Most literature (n = 30) related to children and adolescents with an acquired or traumatic brain injury. Physical function was the most frequently reported outcome across sources (n = 27), followed by communication (n = 25), social (n = 22), cognitive (n = 17), emotional (n = 13), psychological (n = 13), behavioural (n = 8), and sensory (n = 5). Conclusions: Evidence for functional outcomes experienced by children and adolescents accessing music therapy as part of their neurorehabilitation is limited. More than half of the included sources were clinical descriptions with small samples. High-quality studies involving children, adolescents, families, and interprofessional teams are needed to identify the most effective music therapy methods and techniques for functional outcomes in paediatric neurodisability.