Learning to Trust AI Models in Cardiology

Lead Supervisor
Professor Jonna Kuntsi
Professor of Developmental Disorders and Neuropsychiatry
SGDP Centre, IoPPN
jonna.kuntsi@kcl.ac.uk

Co-supervisor
Professor Richard Dobson
King’s College London

Project Details

Remote technology is changing the way in which many conditions are assessed and managed, with King’s investigators leading international developments in relation to disorders such as depression, epilepsy and multiple sclerosis (RADAR-CNS.org; RADAR-Base.org). Benefiting from the RADAR-base platform developed at King’s, we have recently developed novel remote assessment and monitoring technology for adults and adolescents with attention-deficit/hyperactivity disorder (ADHD), which has the potential to transform longitudinal research and clinical long-term monitoring of individuals with the condition.

ADHD is a common disorder, affecting 5% of children and 3% of adults. It is associated with a high risk of adverse outcomes, including the development of comorbid conditions, and represents an extensive burden on the healthcare system. Yet outcomes are highly variable. To transform early intervention, we need to improve our understanding of what predicts and underlies longer-term improvement in ADHD and related impairments in some individuals, but persistence or escalation of the condition in many others. Longitudinal, real-world data is particularly lacking on the association of modifiable health behaviours, such as physical activity, with outcomes. There is also a pressing need to move from the often exclusive reliance on self-reports of ADHD symptoms towards a more objective, detailed characterisation of the symptomatology.

Our project, ADHD Remote Technology (ART), directly addresses these challenges. The ART system incorporates active (questionnaires, cognitive tasks) and passive (physical activity, sleep, smartphone sensor data) monitoring using mobile and web technologies. We are currently in the pilot study phase that involves remote monitoring of participants with and without ADHD for 10 weeks. Our subsequent plan is to investigate the longer-term clinical course in adults and adolescents with ADHD. We seek to identify predictors of improvement of ADHD and co-occurring symptoms and impairments, targets for intervention and novel digital biomarkers. The ongoing multimodal measurement will enable us to build models of the clinical course at a level of detail that was previously impossible.

Datasets

ART (ADHD Remote Technology) project ethics approval number: 20/NE/0034, North East – Tyne & Wear South Research Ethics Committee. Information Compliance approval received from Data Protection Officer, Albert Chan, on 06/11/2019.

Keywords

remote measurement technology; RMT; attention deficit hyperactivity disorder; ADHD