The ChAMP App: A Scalable mHealth Technology for Detecting Digital Phenotypes of Early Childhood Mental Health

Abstract

Childhood mental health problems are common, impairing, and can become chronic if left untreated. Children are not reliable reporters of their emotional and behavioral health, and caregivers often unintentionally under-or over-report child symptoms, making assessment challenging. Objective physiological and behavioral measures of emotional and behavioral health are emerging. However, these methods typically require specialized equipment and expertise in data and sensor engineering to administer and analyze. To address this challenge, we have developed the ChAMP (Childhood Assessment and Management of digital Phenotypes) System, which includes a mobile application for collecting movement and audio data during a battery of mood induction tasks and an open-source platform for extracting digital biomarkers. As proof of principle, we present ChAMP System data from 101 children 4-8 years old, with and without diagnosed mental health disorders. Machine learning models trained on these data detect the presence of specific disorders with 70-73% balanced accuracy, with similar results to clinical thresholds on established parent-report measures (63-82% balanced accuracy). Features favored in model architectures are described using Shapley Additive Explanations (SHAP). Canonical Correlation Analysis reveals moderate to strong associations between predictors of each disorder and associated symptom severity (r = .51-.83). The opensource ChAMP System provides clinically-relevant digital biomarkers that may later complement parent-report measures of emotional and behavioral health for detecting kids with underlying mental health conditions and lowers the barrier to entry for researchers interested in exploring digital phenotyping of childhood mental health.

Using Wearable Digital Devices to Screen Children for Mental Health Conditions: Ethical Promises and Challenges

“…Digital phenotype-based screening tools (DPSTs) may help surmount these obstacles by leveraging sensor data passively collected by smartphones or other wearable devices to screen for mental health disorders (Figure 1). Research studies to date have developed DPSTs (Figure 2) and are in evaluative stages for Attention Deficit Hyperactive Disorder (ADHD) [16,17], bipolar disorder [18], and internalizing disorders [19][20][21][22][23] in children. Our team has found that digital phenotyping can take as little as one to three minutes conducted in a medical office space [21], and others have found that screening could occur remotely between clinical visits [17] which may help overcome time barriers to pediatrician screening for mental illness.…”

“…Research studies to date have developed DPSTs (Figure 2) and are in evaluative stages for Attention Deficit Hyperactive Disorder (ADHD) [16,17], bipolar disorder [18], and internalizing disorders [19][20][21][22][23] in children. Our team has found that digital phenotyping can take as little as one to three minutes conducted in a medical office space [21], and others have found that screening could occur remotely between clinical visits [17] which may help overcome time barriers to pediatrician screening for mental illness. In practice, DPSTs could assess a child's movement, voice, heart rate variability, respirations, eye movements, and galvanic skin response [24][25][26][27][28][29] while they react to a brief potentially threatening situation like walking into a dimly lit room or giving a speech while wearing a small monitor that resembles an electrocardiogram lead or watch-like device [21].…”

“…Our team has found that digital phenotyping can take as little as one to three minutes conducted in a medical office space [21], and others have found that screening could occur remotely between clinical visits [17] which may help overcome time barriers to pediatrician screening for mental illness. In practice, DPSTs could assess a child's movement, voice, heart rate variability, respirations, eye movements, and galvanic skin response [24][25][26][27][28][29] while they react to a brief potentially threatening situation like walking into a dimly lit room or giving a speech while wearing a small monitor that resembles an electrocardiogram lead or watch-like device [21]. After data collection, devices incorporate artificial intelligence (AI), typically using machine learning techniques to generate diagnostic likelihood [30], which informs the pediatrician delivering feedback to the patient's family.…”

“…described above, are now starting to move from adults to children [20][21][22][24][25][26][28][29][30][44][45][46]. Notably, a small number of companies are now pursuing digital phenotype diagnostics or diagnostic aids targeted specifically at pediatric mental health.…”

“…More recent efforts are beginning to expand to consider additional data sources including biomarkers derived from wearable movement sensors [16,17,22], videos of body and facial movements [42,43], and a variety of physiological measurements such as heart rate, heart rate variability, respirations, and galvanic skin response [24][25][26]28,29]. These techniques, as described above, are now starting to move from adults to children [20][21][22][24][25][26][28][29][30][44][45][46]. Notably, a small number of companies are now pursuing digital phenotype diagnostics or diagnostic aids targeted specifically at pediatric mental health.…”