Using multiple wearable sensors to track sleep in nightshift workers
A Multi-Sensor Machine Learning Approach to Precision Sleep Tracking for Nightshift Workers
This project will test whether combining wearable sensors and machine learning can more accurately detect daytime sleep in nightshift workers.
Quick facts
| Phase | Not applicable |
|---|---|
| Study type | Interventional |
| Enrollment | 100 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Henry Ford Health System Academic / other |
| Locations | 1 site (Novi, Michigan) |
| Trial ID | NCT06670287 on ClinicalTrials.gov |
What this trial studies
The trial enrolls nightshift workers and compares a legacy actigraphy algorithm (using only wrist accelerometer data) with a multi-sensor approach that feeds accelerometer and additional sensor data into a machine learning model. Participants complete an in-lab validation with gold-standard polysomnography across planned sleep opportunities totaling eight hours to directly compare algorithm performance. A portion of participants then uses the multi-sensor system at home for four weeks to evaluate real-world implementation and to identify facilitators and barriers. The investigators will release an open-source machine learning algorithm and report differences in sleep detection accuracy and continuity metrics between the methods.
Who should consider this trial
Good fit: Ideal candidates are adults (18+) who have worked a fixed nightshift schedule for at least six months, work at least three night shifts per week with shifts starting between 6:00 PM and 2:00 AM and lasting 8–12 hours, and plan to maintain that schedule for the study.
Not a fit: People who do not work regular nightshifts (including rotating schedules), cannot provide an average 8-hour bed opportunity, are unwilling to use the study sensors at home, are pregnant, have certain neurological disorders, or have recent illicit drug use may not benefit from this intervention.
Why it matters
Potential benefit: If successful, the approach could give nightshift workers more accurate at-home sleep tracking that better reflects daytime sleep and supports improved diagnosis and management.
How similar studies have performed: Some prior research has applied actigraphy and machine learning, but multi-sensor machine learning specifically aimed at detecting daytime sleep in nightshift workers is relatively novel and not yet well validated.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Participants must be working a fixed nightshift schedule, operationalized as: a) working at least three night shifts a week, b) shifts must begin between 18:00 and 02:00, and last between 8 to 12 hours, and c) must also plan to maintain the nightshift schedule for the duration of the study * Participants must have worked the nightshift for at least six months * Must plan to maintain the nightshift schedule for the duration of the study * Participants must be at least 18 years old Exclusion Criteria: * Termination of nightshift schedule or planned travel during the study period * Does not have at least an average of 8-hour time bed opportunity per 24-hour period * Unwilling to integrate the study smart sensors in their bedroom environment * Illicit drug use via self-report and urine drug screen * History of neurological disorders * Alcohol use disorder * Pregnancy
Where this trial is running
Novi, Michigan
- Henry Ford Columbus Medical Center — Novi, Michigan, United States (Recruiting)
Study contacts
- Study coordinator: Philip Cheng, PhD
- Email: pcheng1@hfhs.org
- Phone: 248-344-7361
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.