By Jessica Kent
March 02, 2020 – Researchers could use artificial intelligence to boost efficiencies and precision in sleep disorder treatment, leading to improved care and better patient outcomes, according to a new position statement from the American Academy of Sleep Medicine (AASM).
Published in the Journal of Clinical Sleep Medicine, the statement said that the electrophysiological data collected during polysomnography – the most comprehensive sleep study – is well-positioned for enhanced analysis with AI and machine learning.
Sleep centers collect massive amounts of data, the statement authors said, which could enable AI and machine learning algorithms to advance sleep care. These technologies have the potential to generate more accurate diagnoses, prediction of disease and treatment prognosis, classification of disease subtypes, precision in scoring, and sleep treatment optimization and personalization.
“When we typically think of AI in sleep medicine, the obvious use case is for the scoring of sleep and associated events,” said lead author and committee Chair Dr. Cathy Goldstein, associate professor of sleep medicine and neurology at the University of Michigan. “This would streamline the processes of sleep laboratories and free up sleep technologist time for direct patient care.”
Researchers could also use AI to automate sleep scoring while identifying new insights from sleep data, the authors noted.
“AI could allow us to derive more meaningful information from sleep studies, given that our current summary metrics, for example, the apnea-hypopnea index, aren’t predictive of the health and quality of life outcomes that are important to patients,” said Goldstein.
“Additionally, AI might help us understand mechanisms underlying obstructive sleep apnea, so we can select the right treatment for the right patient at the right time, as opposed to one-size-fits-all or trial and error approaches.”
AI tools have proven their ability to advance sleep disorder diagnosis and treatment. In December 2018, a study published in JAMIA showed that a deep learning algorithm was able to replicate diagnostic scores for sleep staging, sleep apnea, and limb movements at a level of accuracy comparable to that of human clinicians.
“Our results suggest that, given sufficiently large datasets, training on real-world data can yield human level performance and generalize to standardized data sets. The capacity to generalizability is a prerequisite for algorithm deployment in real-world settings,” the JAMIA research team said.
“The potential for substantial clinical impact includes broadening the reach of clinical sleep medicine, augmenting clinical decision-making for sleep specialists, and improving the accuracy and reliability of at-home portable systems.”
The American Sleep Association reports that 50 to 70 million US adults have a sleep disorder, including insomnia, sleep apnea, narcolepsy, and circadian rhythm disorders. These conditions are often linked to decreased work productivity, increased mortality, and poorer quality of life. Timely diagnosis is essential for people to receive appropriate treatment, and AI could help accelerate identification of sleep disorders.
However, the AASM team pointed out that before leaders integrate AI with sleep medicine practice, they should consider transparency and disclosure, as well as testing on novel data and laboratory integration. The authors suggested that manufacturers disclose the intended population and goal of any program used in the evaluation of patients.
Additionally, algorithm developers should test programs intended for clinical use on independent data and aid sleep centers in the evaluation of AI-based software performance.
“AI tools hold great promise for medicine in general, but there has also been a great deal of hype, exaggerated claims and misinformation,” said Goldstein. “We want to interface with industry in a way that will foster safe and efficacious use of AI software to benefit our patients. These tools can only benefit patients if used with careful oversight.”