A Novel Application for the Efficient and Accessible Diagnosis of ADHD Using Machine Learning (Extended Abstract)

Attention-deficit/hyperactivity disorder is the most pervasive neurodevelopmental disorder among children and adolescents. Current clinical diagnosis, however, is inaccurate, inefficient, and inaccessible in developing nations, hindering the administration of proper treatment regimens. Clinical assessments are based on qualitative observations of perceived behavior. They are time-consuming and costly, preventing minorities and socioeconomically disadvantaged groups from gaining the support they need to succeed academically, socially, and occupationally. A more accurate and accessible method of detection is necessary to ensure that all children are able to be diagnosed and given proper treatment regimens. This research proposes a novel machine learning-based method to analyze pupil-dynamics data as an objective biomarker to characterize ADHD. After visualizing and engineering pupillometric features, a voting ensemble classification algorithm and meta learner were developed and yielded the most optimal leave-one-out-cross-validation metrics on a declassified dataset. The ensemble model, in particular, classified ADHD with. 821 sensitivity, 0.727 specificity, and 0.856 AUROC. This model was implemented in a web application that administers a memory task and captures pupil biometrics in realtime. This application is the first to use pupil-size dynamics as a biomarker, and offers a time-efficient, accurate, and accessible approach to diagnose ADHD in developing nations.

I. Introduction

Equitable access to medical diagnosis is integral to achieve sustainable development goals in developing nations, where poverty and a lack of medical institutions create stark inequalities in the accessibility of diagnosis and treatment. Attention-deficit/hyperactivity disorder (ADHD) is the most pervasive neurodevelopmental disorder in the world, particularly in sub-Saharan Africa and Latin America: the prevalence rate, according to a 2007 sample study, is 8.5% and 11.8%, respectively, compared to a 9.2% prevalence rate in the U.S. [1]. Yet, despite the similar prevalence rates in the U.S. and developing nations, ADHD and similar neurobehavioral disorders remain widely underdiagnosed in the developing world as a result of both societal stigmas on youth neurological disorders [2] and deficient medical practices [3]. Chinawa et al. note the egregious neglection of ADHD in developing countries, notably Nigeria, where the reported prevalence rate is uncharacteristically low despite the high prevalence rate in neighboring developing nations [3]. Lower prevalence rates in rural areas of certain countries can be attributed to the inac-cesibility of diagnostic services, precluding the administration of proper treatment regimens [4].