The study, led by Dr. Ann M. Murray and Dr. Julie M. Baughn along with colleagues at the Mayo Clinic Center for Sleep Medicine, set out to answer a deceptively simple question. Could a photoplethysmography-based home sleep apnea test, worn as a ring, reliably identify obstructive sleep apnea in young children? Photoplethysmography, or PPG, is an old technology finding new life in consumer wearables. It works by shining light into the skin, usually at the fingertip, and measuring how much light is reflected back. Because blood absorbs light differently depending on how much oxygen it carries and how much of it is pulsing through the vessels, a PPG sensor can track pulse rate, blood oxygen saturation, and subtle changes in the pulse waveform. Software algorithms can then infer breathing disturbances from these signals, since each apnea or hypopnea event leaves a characteristic fingerprint in the pulse and oxygen data.

In adults, this approach has proven accurate enough that home testing is now standard practice for many patients with suspected obstructive sleep apnea. Young children, however, are a different physiological and behavioral challenge. Their breathing patterns are faster, their apnea events are often shorter and more subtle, and their small fingers and restless sleep make sensor placement difficult. The American Academy of Sleep Medicine has issued a position paper emphasizing that home sleep apnea tests in children require rigorous validation before widespread adoption, and previous studies have mostly focused on adolescents rather than preschoolers, the age group in which obstructive sleep apnea peaks due to enlarged tonsils and adenoids.

To test the device under the most favorable possible conditions, the researchers enrolled fifty children aged two to six who had already been referred for polysomnography because of suspected obstructive sleep apnea. Rather than sending the device home, the team asked each child to wear the ring-shaped SleepImage PPG sensor concurrently during their standard in-laboratory sleep study. This design is important: it means the new technology was evaluated side by side with the full polysomnographic apparatus, attended by trained technologists, with no risk of the setup errors or lost nights that plague true home testing. If the device struggled even in this ideal scenario, its performance in an unattended home environment would likely be no better.

The results revealed problems at two distinct levels: feasibility and accuracy. Of the fifty children enrolled, thirty-nine completed testing, and even among these the device did not always deliver usable data. Device-related issues such as sensor detachment, intolerance by the child, and user setup errors precluded successful data collection in eleven percent of studies. An additional twenty-five percent of the recordings captured less than two hours of total sleep time as measured by the PPG device, a duration generally considered too short to characterize a child’s sleep-disordered breathing reliably. In other words, in more than a third of cases the device either failed outright or produced too little data to be clinically meaningful, and this occurred under supervised laboratory conditions rather than in a child’s bedroom.

The technical shortfall showed up starkly in the comparison of sleep duration measurements. Using the automated SleepImage scoring software, the mean difference in total sleep time between the PPG device and full polysomnography was minus 220 minutes, a gap of nearly four hours, with a 95 percent confidence interval of minus 278 to minus 62 minutes and a p value below 0.001. This enormous discrepancy reflects the fact that the PPG-based algorithm could not distinguish sleep from wakefulness in young children the way electroencephalography-based polysomnography can. When the researchers created a limited polysomnography dataset, trimming the laboratory recording to include only the periods when PPG data were available, the mean difference in sleep time shrank to just minus 11 minutes, which was not statistically significant. That comparison confirmed the device was capturing data during the same windows of time; it simply could not tell how much of that time the child was actually asleep.

Diagnostic accuracy proved equally troubling. Using automated scoring, the device showed a sensitivity of 94.1 percent for diagnosing obstructive sleep apnea, meaning it correctly flagged most children who truly had the condition. But the specificity was only 22.7 percent, indicating the device frequently produced positive results in children who did not have obstructive sleep apnea on polysomnography. The positive predictive value was 0.48, meaning that fewer than half of the children flagged by the device actually had the diagnosis confirmed by the gold standard, while the negative predictive value of 0.83 was somewhat stronger. The obstructive apnea-hypopnea index, the key severity metric counting breathing pauses and reductions per hour of sleep, was significantly higher when measured by the PPG device than by either full or limited polysomnography, with p values below 0.001. The device also detected no central apnea events at all, a notable omission because central sleep apnea, in which the brain temporarily stops sending breathing signals, is clinically relevant in young children and cannot be assessed by pulse-based inference alone.

There was one genuine bright spot in the findings. Caregivers reported good ease of use and tolerance of the ring-shaped device, suggesting that the fundamental form factor is acceptable even to toddlers and preschoolers who might resist more cumbersome equipment. This matters because the barriers to pediatric sleep diagnosis are not purely technical; they are also human. Polysomnography requires an overnight stay in a sleep laboratory with dozens of electrodes attached to the scalp, chest, abdomen, and face, an experience that can be distressing for young children and demanding for families. Previous work by some of the same Mayo Clinic investigators has focused on improving the emotional experience of pediatric sleep studies, including the use of certified child life specialists to ease setup. A comfortable, minimally invasive alternative remains an attractive goal, which is precisely why rigorous validation studies like this one are needed before such devices are trusted.

The stakes of getting pediatric sleep apnea diagnosis right are considerable. Obstructive sleep apnea affects a meaningful share of preschool-aged children, with systematic reviews suggesting a notable prevalence in the general population, and it has been linked to behavioral problems, impaired learning, cardiovascular strain, and metabolic consequences. The standard treatment, adenotonsillectomy, has been shown in randomized trials to improve sleep and related outcomes even in children with mild disease. Yet the decision to proceed with surgery in a young child is not one to make on the basis of an inaccurate test. A device that overcalls the diagnosis could send children to unnecessary surgery, while one that misses severity could delay needed treatment. The authors conclude that this particular PPG-based home sleep apnea test was technically inadequate and diagnostically inaccurate in children two to six years of age, and they emphasize that future research must determine the best technology and the appropriate ages of use before any home sleep apnea test can be recommended for widespread use in children. For now, the message for parents and clinicians is clear: when it comes to diagnosing sleep apnea in preschoolers, the overnight laboratory study remains the standard that home gadgets have yet to meet.

Subject of Research: Validation of a photoplethysmography-based home sleep apnea test against polysomnography for diagnosing obstructive sleep apnea in preschool children

Article Title: A photoplethysmography-based home sleep apnea test compared with polysomnography in children 2 to 6 years of age

Article References: Murray, A. M., Matarese, C. A., Lloyd, R. M., Sorensen, C., LeMahieu, A., Reckward, D., Tullis, R., Bellinger, M., Brown, V., Herold, D. L., Morgenthaler, T. I., & Baughn, J. M. (2026). A photoplethysmography-based home sleep apnea test compared with polysomnography in children 2 to 6 years of age. Journal of Clinical Sleep Medicine, 22(1), Article 140. https://doi.org/10.1007/s44470-026-00156-x

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00156-x

Keywords: pediatric sleep apnea, home sleep apnea test, photoplethysmography, polysomnography, preschool children, obstructive sleep apnea, SleepImage, apnea-hypopnea index, sleep medicine, diagnostic accuracy, Mayo Clinic, wearable sensors