Infant predictors of neurodevelopmental outcomes in early-onset epilepsy: integrating video-based electronic health records

Epilepsy, the most common brain condition in childhood, is associated with conditions affecting children's development, which can impact on learning, social and everyday life skills, leading to poor life outcomes and long-term mental health problems with high health service utilization. Epilepsy and neurodevelopmental conditions commonly co-occur and this comorbidity is associated with reduced quality of life and shortened life expectancy (Tye et al. 2019).

Identification of markers that predict neurodevelopmental outcomes will further our understanding of the impact of and effective interventions for developmental difficulties in early-onset epilepsy, a research priority identified by patients, carers, policymakers and healthcare professionals (Baulac et al. 2015).

While studies have suggested the role of maternal and perinatal risk factors on epilepsy, for example, maternal weight, preeclampsia, and recurrent pregnancy loss (e.g. Paz Levy et al. 2019), most have been from primary care databases with limited data quality or have required active recruitment and thus not included a representative population group.

Early age of onset, seizure type and aetiology of epilepsy have been identified as contributing factors in autism diagnosis (Strasser et al. 2018), but it is critical that prospective studies are performed to assess maternal and early-life modifiable risk factors and predictors prior to the emergence of developmental difficulties, an approach which has been established in infant siblings of children diagnosed with neurodevelopmental conditions (Tye et al., 2022) and tuberous sclerosis complex (Lindsay et al. 2023). The linkage of routinely collected data is a particularly valuable technique for prospectively studying the risk factors involved in co-occurring early-onset epilepsy and developmental conditions.

Embedding digital technology will help better power early intervention studies with objective markers which are currently lacking and to monitor changes over time or with treatment (Bolte et al., 2016). Aim of the investigation (up to 150 words) State primary research question and where appropriate the primary hypotheses being tested

This research will take an interdisciplinary approach combining AI-automated video coding with routine medical records and prospective measurement to measure and predict neurodevelopmental difficulties in epilepsy, to address the following aims:

To provide proof-of-concept for the application of machine learning to large-scale video dataset to assess development and behaviour in early-onset epilepsy

To test the feasibility of combining videos and routinely collected medical data to determine associations with maternal and perinatal factors To identify potential predictors of neurodevelopmental outcomes for early-onset epilepsy