Attention-Deficit Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental condition affecting both children and adults alike, characterized primarily by problematic inattention and hyperactivity/impulsivity that causes functional impairment in daily life. The current ADHD diagnosis relies on the physician-administered approach and requires significant manual intervention, which in turn could be more prone to human error and may also suffer from recall bias. The ubiquitous nature of smartphones and their rich set of embedded sensors make them an ideal solution for behavior tracking and diagnosis. In this study, we introduce SmartADHDMonitor, a novel approach for predicting weekly ADHD symptom levels and diagnostic status. This approach utilizes passively collected smartphone app usage data from 12 college-age students on the Android platform. We calculated a comprehensive set of features using the smartphone app usage behavioral data and constructed a family of machine-learning models for predicting weekly levels of ADHD symptoms and ADHD diagnostic status. Our results demonstrate that the app usage data could be used for predicting ADHD diagnostic status fairly accurately with F1scores as high as 0.88. Our preliminary findings offer a promising research direction into machine learning applications for ADHD diagnosis and monitoring through smartphone sensing data. As one of the first studies in this domain, SmartADHDMonitor offers a novel, technology-driven perspective on addressing the growing need for accessible mental health care solutions, further advancing the smart connected health in the field of ADHD monitoring.

Recent studies have demonstrated that geographic location features collected using smartphones can be a powerful predictor for depression. While location information can be conveniently gathered by GPS, typical datasets suffer from significant periods of missing data due to various factors (e.g., phone power dynamics, limitations of GPS). A common approach is to remove the time periods with significant missing data before data analysis. In this paper, we develop an approach that fuses location data collected from two sources: GPS and WiFi association records, on smartphones, and evaluate its performance using a dataset collected from 79 college students. Our evaluation demonstrates that our data fusion approach leads to significantly more complete data. In addition, the features extracted from the more complete data present stronger correlation with self-report depression scores, and lead to depression prediction with much higher$F_1$scores (up to 0.76 compared to 0.5 before data fusion). We further investigate the scenario when including an additional data source, i.e., the data collected from a WiFi network infrastructure. Our results show that, while this additional data source leads to even more complete data, the resultant$F_1$scores are similar to those when only using the location data (i.e., GPS and WiFi association records) from the phones.