Smart stethoscopes, smart inhalers, and ChatGPT: latest developments of digital health in childhood asthma

ABSTRACT
Asthma is the most common chronic disease in children, yet outcomes remain suboptimal, with high rates of exacerbations, emergency visits, and preventable deaths. Despite well-established treatment guidelines, care delivery continues to
be limited by three major structural barriers: incomplete or unreliable symptom assessment, infrequent and reactive follow-up, and insufficient day-to-day support for families. Digital health technologies offer new opportunities to address these challenges by enabling objective data collection, continuous monitoring, and accessible guidance outside clinical settings. This mini-review explores three key innovations that are transforming paediatric asthma care: smart stethoscopes, smart inhalers, and large language models (LLMs). Smart stethoscopes use artificial intelligence to detect wheezing and other abnormal breath sounds at home with high sensitivity. Although evidence in routine clinical practice is still limited, they offer promise in improving symptom recognition, especially in preschool-aged children. Smart inhalers and spacers provide objective data on medication use and inhaler technique, helping clinicians distinguish poor asthma control from non-adherence. Real-time monitoring systems have also been shown to improve asthma control scores, although they may increase healthcare utilisation due to heightened clinical vigilance. Predictive modelling based on inhaler sensor data has demonstrated good accuracy in forecasting exacerbations several days in advance, offering a shift from reactive to preventative care. LLMs such as ChatGPT, Claude, and Gemini provide immediate, comprehensible responses to asthma-related questions from families, filling a critical gap in support between clinic visits. Recent studies show that their responses are generally accurate, clear, and appropriate for parents, particularly when using paid versions. They may also assist healthcare professionals by generating educational materials and synthesising clinical guidance, though concerns around hallucinations, data privacy, and safety in acute settings currently limit their clinical use. Together, these digital tools offer promising avenues to personalise and modernise asthma care for children. However, further validation, integration into care pathways, and attention to safety and equity will be essential to translate this potential into improved outcomes.

IMPACT STATEMENT
Digital health tools can provide objective monitoring, timely interventions, and accessible support, offering new opportunities to improve outcomes in childhood asthma.

Received: Aug 27, 2025
Accepted: Nov 25, 2025
Published: Nov 25, 2025