Background: Epilepsy is a common neurological condition in children, and accurate detection of seizures and their frequency is essential for diagnosis and treatment. Standard monitoring using electroencephalography alongside clinical observation is often burdensome in pediatric settings, as electrodes can cause discomfort and restrict mobility. Contactless sensor technologies may offer a promising supplement by enabling monitoring without physical contact. Objective: This study aims to explore challenges in standard pediatric epilepsy monitoring from the perspective of health care professionals and examines the potential benefits and requirements of supplementary contactless sensor technologies in this setting. Methods: Participant observation of routine processes in standard pediatric epilepsy monitoring was conducted at a German university hospital. Field notes from 40 observed procedures were analyzed using structuring content analysis. Building on these findings, a focus group with pediatric neurologists, nurses, and medical technical assistants (n=6) explored the potential benefits and implementation requirements of contactless sensor technologies. Focus group data were analyzed using focus group illustration maps. Results: A reference workflow of standard pediatric epilepsy monitoring was derived, revealing psychosocial, medical, and organizational challenges faced by health care professionals. Electroencephalography recordings and clinical observation required considerable reassurance of patients and parents or carers, were vulnerable to movement artifacts and incomplete seizure documentation, and were labor- and resource-intensive. Focus group participants viewed contactless sensor technologies as a potentially valuable supplement by enabling continuous long-term monitoring with minimal additional burden. Conclusions: By identifying challenges associated with standard pediatric epilepsy monitoring, this study provides a foundation for the needs-based development and implementation of supplementary contactless sensor technologies. Such technologies should be tailored to the clinical setting and designed to address existing burdens, with the potential to complement standard monitoring. Trial Registration: Deutsches Register Klinischer Studien (DRKS) DRKS00027017; https://drks.de/search/de/trial/DRKS00027017
Depression subtype classification from social media posts: few-shot prompting vs. fine-tuning of large language models
BackgroundSocial media provides timely proxy signals of mental health, but reliable tweet-level classification of depression subtypes remains challenging due to short, noisy text, overlapping symptomatology,




