Summary: A medical wrist device detected disrupted sleep patterns 24 hours before participants started shedding the flu virus. While the initial study was focused on influenza, researchers speculate the device could be sufficient at spotting the beginning of other infections, including COVID-19. While it should not be considered a full diagnostic device, the wristband data will be useful to guide when people should self isolate as a precaution.
Source: University of Michigan
A pre-print study using data from medical wristbands suggests that evidence of disrupted sleep could predict when study participants were coming down with the flu—24 hours before they became contagious.
The study is posted on the arXiv and has been submitted to IEEE Transactions on Biomedical Engineering. It is in its second round of review at the journal.
A machine learning algorithm used data from a medical wristband to deduce the sleep patterns of 25 study participants who had been deliberately exposed to a strain of the flu. For seven of the eight participants who came down with the flu and also had usable data, evidence of disrupted sleep appeared 24 hours before the participants began shedding viruses.
While the initial study was focused on the flu, this method may be general enough to spot the beginning of other infections—possibly including COVID-19.
“As we get more data from the populations wearing smart watches through this pandemic, our predictive models will be refined. We imagine that these refined models could be used to generate an early warning signal and even potentially enable the prediction of asymptomatic spreading without tests,” said corresponding author Alfred Hero, the John H. Holland Distinguished University Professor of Electrical Engineering and Computer Science and the R. Jamison and Betty Williams Professor of Engineering.
While this approach would in no way be able to diagnose COVID-19, it may be able to give useful guidance about when people should self-isolate out of precaution. In the current pandemic, those experiencing symptoms of cold and flu are asked to self-isolate as many who have tested positive for the new coronavirus have not experienced the most common symptoms of fever, dry cough and shortness of breath. If the algorithm can work with the less complete data of a smartwatch or fitness tracker, critical workers could get an early warning that they are falling ill and self-isolate.
Hero is also a professor of biomedical engineering in the College of Engineering and professor of statistics in the College of Literature, Science, and the Arts.
About this neurotech research article
Source: University of Michigan Media Contacts: Kate McAlpine – University of Michigan Image Source: The image is in the public domain.
An unsupervised transfer learning algorithm for sleep monitoring
Objective: To develop multisensor-wearable-device sleep monitoring algorithms that are robust to health disruptions affecting sleep patterns.
Methods: We develop an unsupervised transfer learning algorithm based on a multivariate hidden Markov model and Fisher’s linear discriminant analysis, adaptively adjusting to sleep pattern shift by training on dynamics of sleep/wake states. The proposed algorithm operates, without requiring a priori information about true sleep/wake states, by establishing an initial training set with hidden Markov model and leveraging a taper window mechanism to learn the sleep pattern in an incremental fashion. Our domain-adaptation algorithm is applied to a dataset collected in a human viral challenge study to identify sleep/wake periods of both uninfected and infected participants.
Results: The algorithm successfully detects sleep/wake sessions in subjects whose sleep patterns are disrupted by respiratory infection (H3N2 flu virus). Pre-symptomatic features based on the detected periods are found to be strongly predictive of both infection status (AUC = 0.844) and infection onset time (AUC = 0.885), indicating the effectiveness and usefulness of the algorithm.
Conclusion: Our method can effectively detect sleep/wake states in the presence of sleep pattern shift.
Significance: Utilizing integrated multisensor signal processing and adaptive training schemes, our algorithm is able to capture key sleep patterns in ambulatory monitoring, leading to better automated sleep assessment and prediction.