The effect of a novel wearable feedback device on cardiopulmonary resuscitation training in medical students: a three-arm cluster randomized controlled trial.

Digital transformation is reshaping every stage of psychiatric care, from diagnosis to long-term monitoring and personalized intervention. Mental health and neurodevelopmental disorders (NDDs) are among the most heterogeneous categories in medicine, resisting one-size-fits-all protocols and motivating a shift toward personalized and precision approaches. Over the past decade, eHealth technologies, including artificial intelligence (AI), machine learning, wearable sensors, mobile applications, virtual reality, and remote monitoring, have moved from proof-of-concept to clinically oriented tools that provide objective, continuous, and individualized data where clinical observation alone is limited by recall bias, low sampling frequency, and rater subjectivity. AI is increasingly becoming the enabling framework that integrates multimodal data from wearables, smartphones, electronic health records, and behavioral monitoring into clinically actionable knowledge. This Research Topic gathered 29 contributions across four lines of work: objective assessment of neurodevelopmental conditions, passive physiological monitoring, digital therapeutics and care-pathway implementation, and screening or predictive tools for risk and outcome. A substantial cluster of contributions addressed ADHD through instrumented, rater-independent measurement. contributed three related studies using point-of-view (POV) cameras and machine learning: two studies focused on objective hyperactivity assessment in ADHD (including a preschool-specific paradigm with MediaPipe), and a third study extended the same approach to multimodal depression detection, demonstrating that passive behavioral capture generalizes across diagnostic targets. presented a complementary machine learning feature selection approach for ADHD prediction, while provided a scoping review of digital health technology for adults with ADHD, a population underserved by tools designed for children. conducted a randomized controlled trial to test whether the format of the Adult ADHD Self-Report Scale affects screen-positive rates, a reminder that digitizing an instrument does not remain neutral to its performance. used AI-assisted text analysis to trace how the ADHD construct has shifted across six DSM editions. Beyond ADHD, presented “Akshar Mitra,” a multimodal framework for early dyslexia detection, and contributed with a mini review on digital health in Tourette syndrome. Collectively, these contributions illustrate a shift from subjective symptom reporting to objective, technology-enabled phenotyping, an approach intended to complement rather than replace clinical judgment. A second cluster of studies moved from clinic-based assessment to continuous, out-of-clinic data capture. reported a case study on the feasibility of one-month, home-based heart rate variability (HRV) monitoring in ASD using smart clothing; presented a case report on the use of wearable-derived HRV and sleep data to predict mood episodes in bipolar disorder. used deep learning to detect stress in RR-interval data across major depressive disorder, panic disorder, and healthy controls, and used simulated virtual reality experiences to predict early treatment response in panic disorder. Together, these studies point toward continuous, ecologically valid observation as a complement to episodic clinical evaluation.