INTERGROWTH-21st Project international INTER-NDA standards for child development at 2 years of age: an international prospective population-based study.

OBJECTIVES: To describe the construction of the international INTERGROWTH-21st Neurodevelopment Assessment (INTER-NDA) standards for child development at 2 years by reporting the cognitive, language, motor and behaviour outcomes in optimally healthy and nourished children in the INTERGROWTH-21st Project. DESIGN: Population-based cohort study, the INTERGROWTH-21st Project. SETTING: Brazil, India, Italy, Kenya and the UK. PARTICIPANTS: 1181 children prospectively recruited from early fetal life according to the prescriptive WHO approach, and confirmed to be at low risk of adverse perinatal and postnatal outcomes. PRIMARY MEASURES: Scaled INTER-NDA domain scores for cognition, language, fine and gross motor skills and behaviour; vision outcomes measured on the Cardiff tests; attentional problems and emotional reactivity measured on the respective subscales of the preschool Child Behaviour Checklist; and the age of acquisition of the WHO gross motor milestones. RESULTS: Scaled INTER-NDA domain scores are presented as centiles, which were constructed according to the prescriptive WHO approach and excluded children born preterm and those with significant postnatal/neurological morbidity. For all domains, except negative behaviour, higher scores reflect better outcomes and the threshold for normality was defined as ≥10th centile. For the INTER-NDA's cognitive, fine motor, gross motor, language and positive behaviour domains these are ≥38.5, ≥25.7, ≥51.7, ≥17.8 and ≥51.4, respectively. The threshold for normality for the INTER-NDA's negative behaviour domain is ≤50.0, that is, ≤90th centile. At 22-30 months of age, the cohort overlapped with the WHO motor milestone centiles, showed low postnatal morbidity (<10%), and vision outcomes, attentional problems and emotional reactivity scores within the respective normative ranges. CONCLUSIONS: From this large, healthy and well-nourished, international cohort, we have constructed, using the WHO prescriptive methodology, international INTER-NDA standards for child development at 2 years of age. Standards, rather than references, are recommended for population-level screening and the identification of children at risk of adverse outcomes.

Hierarchical Class Incremental Learning of Anatomical Structures in Fetal Echocardiography Videos.

This paper proposes an ultrasound video interpretation algorithm that enables novel classes or instances to be added over time, without significantly compromising prediction abilities on prior representations. The motivating application is diagnostic fetal echocardiography analysis. Currently in clinical practice, recording full diagnostic fetal echocardiography is not common. Diagnostic videos are typically available in varying length and summarize a number of diagnostic sub-tasks of varying difficulty. Although large clinical datasets may be available at onset to build ultrasound image-based models for automatic image analysis, data may also become available over extended time to assist in algorithm refinement. To address this scenario, we propose to use an incremental learning approach to build a hierarchical network model that allows for a parallel inclusion of previously unseen anatomical classes without requiring prior data distributions. Super classes are obtained by coarse classification followed by fine classification to allow the model to self-organize anatomical structures in a sequence of categories through a modular architecture. We show that this approach can be adapted with new variable data distributions without significantly affecting previously learned representations. Two extreme situations of new data addition are considered; (1) when new class data is available over time with volume and distribution similar to prior available classes, and (2) when imbalanced datasets arrive over future time to be learned in a few-shot setting. In either case, availability of data from prior classes is not assumed. Evolution of the learning process is validated using incremental accuracies of fine classification over novel classes and compared to results from an end-to-end transfer learning-derived model fine-tuned on a clinical dataset annotated by experienced sonographers. The modularization of subsequent learning reduces the depreciation in future accuracies over old tasks from 6.75% to 1.10% using balanced increments. The depreciation is reduced from 6.95% to 1.89% with imbalanced data distributions in future increments, while retaining competitive classification accuracies in new additions of fine classes with parameter operations in the same order of magnitude in all stages in both cases.