Smartwatches for non‐invasive hypoglycaemia detection during cognitive and psychomotor stress

BACKGROUNDHypoglycaemia is one of the most relevant complications of diabetes 1 and induces alterations in physiological parameters 2,3 that can be measured with smartwatches and detected using machine learning (ML).4 The performance of these algorithms when applied to different hypoglycaemic ranges or in situations involving cognitive and psychomotor stress remains unclear.Demanding tasks can significantly affect the physiological responses on which the wearable-based hypoglycaemia detection relies.5 The present analysis aimed to investigate MLbased hypoglycaemia detection using wearable data at different levels of hypoglycaemia during a complex task involving cognitive and psychomotor challenges (driving). | METHODSThis was a pre-specified analysis of two consecutive studies conducted at the University Hospital Bern (Study A: 10/2019-07/2020; Study B: 11/2021-03/2022).The studies followed the Declaration of Helsinki, guidelines of good clinical practice, and legally applicable requirements.We included individuals with type 1 diabetes, aged 21-50 years (up to 60 years for Study B).Key exclusion criteria included pregnancy, seizure disorders, severe organ dysfunction, or alcohol/drug abuse.Following approval by the local Ethics Committee