Machine Learning–Based Prediction of Attention-Deficit/Hyperactivity Disorder and Sleep Problems With Wearable Data in Children

Abstract

ImportanceEarly detection of attention-deficit/hyperactivity disorder (ADHD) and sleep problems is paramount for children’s mental health. Interview-based diagnostic approaches have drawbacks, necessitating the development of an evaluation method that uses digital phenotypes in daily life.ObjectiveTo evaluate the predictive performance of machine learning (ML) models by setting the data obtained from personal digital devices comprising training features (ie, wearable data) and diagnostic results of ADHD and sleep problems by the Kiddie Schedule for Affective Disorders and Schizophrenia Present and Lifetime Version for Diagnostic and Statistical Manual of Mental Disorders, 5th edition (K-SADS) as a prediction class from the Adolescent Brain Cognitive Development (ABCD) study.Design, Setting, and ParticipantsIn this diagnostic study, wearable data and K-SADS data were collected at 21 sites in the US in the ABCD study (release 3.0, November 2, 2020, analyzed October 11, 2021). Screening data from 6571 patients and 21 days of wearable data from 5725 patients collected at the 2-year follow-up were used, and circadian rhythm–based features were generated for each participant. A total of 12 348 wearable data for ADHD and 39 160 for sleep problems were merged for developing ML models.Main Outcomes and MeasuresThe average performance of the ML models was measured using an area under the receiver operating characteristics curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). In addition, the Shapley Additive Explanations value was used to calculate the importance of features.ResultsThe final population consisted of 79 children with ADHD problems (mean [SD] age, 144.5 [8.1] months; 55 [69.6%] males) vs 1011 controls and 68 with sleep problems (mean [SD] age, 143.5 [7.5] months; 38 [55.9%] males) vs 3346 controls. The ML models showed reasonable predictive performance for ADHD (AUC, 0.798; sensitivity, 0.756; specificity, 0.716; PPV, 0.159; and NPV, 0.976) and sleep problems (AUC, 0.737; sensitivity, 0.743; specificity, 0.632; PPV, 0.036; and NPV, 0.992).Conclusions and RelevanceIn this diagnostic study, an ML method for early detection or screening using digital phenotypes in children’s daily lives was developed. The results support facilitating early detection in children; however, additional follow-up studies can improve its performance.

Machine learning algorithms to the early diagnosis of fetal alcohol spectrum disorders

“…ML has been effectively applied in the medical field to diagnose neurological disorders, including ASD (Vakadkar et al, 2021 ; Bahathiq et al, 2022 ; Briguglio et al, 2023 ) and ADHD (Slobodin et al, 2020 ; Mikolas et al, 2022 ; Briguglio et al, 2023 ; Kim et al, 2023 ). These studies have demonstrated the potential of ML to increase diagnostic accuracy, reduce time to diagnosis and improve reproducibility.…”

“…Similarly, ML classifiers for ADHD have been developed based on clinical and psychological data (i.e. attention, impulsiveness, sleep, and emotional disorders) (Slobodin et al, 2020 ; Mikolas et al, 2022 ; Kim et al, 2023 ).…”

Prediction of Postnatal Growth Failure in Very Low Birth Weight Infants Using a Machine Learning Model

“…Adding certain informative features can improve the performance of the model. A previous study, which investigated the use of a machine learning model for predicting the risk of ADHD or pediatric sepsis, has shown the results of adding demographic features or laboratory features compared to using only clinical features [22,23]. In a study about a machine learning model used to predict the risk of pediatric asthma exacerbation, adding environmental features-such as air quality and temperature-improved the performance of the model compared to using only clinical features [24].…”

Generalized sleep decoding with basal ganglia signals in multiple movement disorders

“…We evaluated the performance of eight commonly used machine learning classifiers in our PD-STN dataset. It is shown that the LightGBM classifier, a lightweight gradient boosting framework based on decision tree algorithm 23,24 that has been used in multiple sleep-related studies 21,25 , was constantly associated with the highest accuracies as well as a sensible model convergent speed in both the three-class (wake/NREM/REM) and five-class (wake/N1/N2/N3/REM) decoding contexts (Supplementary Fig. 1).…”