最新糖心Vlog

A dearth of paediatric data collected using artificial intelligence (AI) tools could compromise the understanding of early sleep patterns.

鈥淪leep stages are primarily determined by analysing the brain waves during sleep, which traditionally is done manually (or semi-automated),鈥� said the 最新糖心Vlog of Adelaide鈥檚 Associate Professor Mathias Baumert, School of Electrical and Mechanical Engineering.

鈥淭here has been a big push to automate this labour-intensive and expensive process using deep learning 鈥� an AI method that teaches computers to process data in a way that is inspired by the human brain.

鈥淲e have found evidence that errors in sleep classification generated using AI could lead to misdiagnosis of sleep disorders, particularly in children.鈥�

Nerve impulses in the brain cause electrical activity. An electroencephalogram (EEG) records the brain鈥檚 electrical activity on a person鈥檚 scalp. Collected overnight, it provides the input data to deep learning systems.

鈥淒eep learning systems need big EEG databases to learn the brain wave patterns associated with various sleep stages to give reliable results,鈥� said Associate Professor Baumert.

鈥淗owever, most available sleep EEG databases contain adult data only and therefore, the systems developed on those databases seem to have issues when dealing with kids鈥� sleep because they have never 鈥渟een鈥� and 鈥渓earnt鈥� it.

鈥淭his means sleep disorders in children could be misdiagnosed due to using adult-biased data sets.鈥�

鈥淓ffective strategies are needed to avoid bias in the learning process and also to identify unusual behaviour of AI systems and mitigate the risk of misdiagnoses.鈥� Dr Huy Phan

The association between sleep problems, cognitive development and mental health has long been studied and recognised. Early identification of sleep problems, especially in young people, is vital to maximising the effectiveness of appropriate treatment.

Generally, sleep comprises rapid eye movement (REM) sleep and non-rapid eye movement (NREM) sleep. NREM is categorised into three stages (N1, N2, N3). Prior to the introduction of AI methods for classifying sleep patterns, a technician would examine overnight EEG, 30 seconds at a time, and assign one of the sleep stages to it.

鈥淪leep follows a typical cyclical pattern of sleep stages that make up the 鈥榮leep architecture鈥� and can be seen in a hypnogram - a graph that聽represents the stages of sleep over time,鈥� said Associate Professor Baumert.

鈥淐linicians consider several metrics from hypnograms, such as sleep onset latency, sleep efficiency, time spent in each of the sleep stages, and time spent awake after sleep onset, to help them diagnose sleep disorders.鈥�

On adult EEG, AI systems assign sleep stages correctly about nine out of ten times 鈥� on par with human experts. A similar performance can be achieved for children if paediatric data are included in the deep learning process.

Dr Huy Phan, Alan Turing Institute, UK, who was co-author of the study said improvements are needed in AI systems to avoid bias towards population groups and ensure accurate diagnoses for all patients.

鈥淓ven if trained with paediatric data, current AI systems don鈥檛 work equally well in all patients. While sleep stage classification accuracy appears sufficient when looking at EEG databases as a whole, errors in the sleep metrics of individual patients can be substantial, which is unacceptable for any diagnostic purpose.鈥�

鈥淓ffective strategies are needed to avoid bias in the learning process and also to identify unusual behaviour of AI systems and mitigate the risk of misdiagnoses.鈥�

The study was published in the journal .