Respiration, arousals, and position
Breathing rate, sleep position, activity and arousals through the night
The sensor on NAPPA records more than sleep states. The same signals give three measures of their own: how fast and how regularly the infant breathes, which position they sleep in, and how much they move — from which night-time arousals are detected. Each is computed in the sensor itself and stored every few seconds through the night (Figure 1A).
Arousals are where direct measurement matters most. Parents usually report them in a diary, and what a parent notices is not the same as what the infant does; NAPPA counts them by a fixed rule, at any duration a study needs (Figure 1B).
AWhat the sensor computes
BHow an arousal is detected
What you get
- Breathing
- Breathing rate and regularity every 30 seconds, for the stretches where the signal passes the automatic quality check — mostly the calmer stretches of sleep.
- Body position
- Supine, prone, left side, right side, head up or head down every 5 seconds, and the time spent in each.
- Activity
- Movement every 5 seconds, comparable to actigraphy but from the trunk rather than a limb.
- Arousals
- The number and timing of arousals at durations from 30 seconds to 10 minutes, each classed as calm or restless by how often the infant changed position during it.
What you need
- Garment
- NAPPA, tightened at the waist so the sensor moves with the belly.
- Sensor
- One Movesense sensor: Flash, MD or HR2. Everything is computed in the sensor, so no streaming is needed.
- Recording
- A night's sleep; for arousals, the published study recorded three consecutive nights.
Breathing rate
Breathing rate is read from the gyroscope, which in this position follows the belly's breathing movement more faithfully than the accelerometer does. Every 30 seconds the sensor finds the dominant rhythm of the filtered signal by autocorrelation; the height of that autocorrelation peak doubles as a quality measure, because movement or genuinely irregular breathing flattens it. Windows below the threshold are left out rather than reported with a wrong number.
In 24 infants aged 1 to 9 months, recorded during clinical polysomnography, NAPPA's breathing rate was compared with the rate from capnography over 88 hours. The median error was 1.5 breaths per minute (interquartile range 1.1–2.6), with no bias and 95% limits of agreement of ±5.7 breaths per minute. The quality check accepted 39% of the 30-second windows; a stricter threshold gives closer agreement but covers less of the night (Acta Paediatrica 2021).
Body position
Position comes from the direction of gravity at the sensor. Because the garment holds the sensor in the same place and orientation every night, the angle of the gravity vector in two planes maps directly to six positions: supine, prone, left side and right side from the roll of the trunk, head up and head down from its pitch. That is enough to measure, for instance, how much of the night an infant spends on their front.
Arousals, and what parents notice
In a study of 104 families, infants wore NAPPA at home for three consecutive nights at around 7 and 12 months of age while the parents kept a sleep diary of arousals and rated each night's sleep quality from 0 to 10. NAPPA data came from 372 nights of 97 infants, and 165 nights of 73 infants had both a diary and a matching recording (Pediatric Research 2026).
Parents' ratings of sleep quality followed both the arousals they reported (ρ = −0.47) and those NAPPA measured (ρ = −0.39 to −0.40 for 3- and 5-minute arousals). But the two counts were different things. Parents' reports matched measured arousals best at a duration of about 3 minutes (ρ = 0.40); they missed many of the shorter arousals and reported more long ones than NAPPA measured. When the infant slept in the parents' room, parents reported more arousals — 2.5 against 1.9 a night — although the measured arousals did not differ by sleep setting.
The difference is not a failing of either source. A diary describes the parents' night; the recording describes the infant's. Studies of infant sleep need both, and a wearable makes the second practical at home and at scale.
How do measured arousals relate to waking in the clinical sense? In a separate group of 40 infants recorded with NAPPA during polysomnography, about half of the shortest arousals (30 seconds) coincided with wake in the polysomnogram, rising to 70–80% for arousals of 3 minutes or more. The shorter ones are still behaviourally real — brief movements that may wake a parent without waking the infant.
Limits
A single sensor cannot separate the infant's own movement from movement imposed on them, such as a parent moving in a shared bed or a pram being pushed. For nights in mixed sleeping arrangements, a short sleep diary helps interpret the activity and arousal measures (Pediatric Research 2026, review).
Available on
Publications
No publication here covers all of the selected products.