Cloud analytics

Upload a recording and get back validated measures instead of raw signal — the same measures the published studies used.

19 peer-reviewed publications behind the methods

ρ = 0.97 agreement of the motor score with the Alberta Infant Motor Scale

220 motor metrics on normative growth charts

A wearable recording is hours of acceleration and rotation data. Turning it into something a study can report — a motor score, a centile, a night's sleep — takes a validated chain of classifiers and reference data. The cloud analytics is that chain, run for you. For MAIJU it is the same fully automated pipeline that produced the results in the published studies, so a result from your recording means what it means in the papers.

From recording to results

Record. The garment is worn at home and the recording runs from the Tessalogger app — during free play for MAIJU, during sleep for NAPPA.
Upload. Tessalogger sends the recording to the cloud when the session ends. Nothing is computed on the phone. Data upload can be performed manually or automatically.
Check and select. Every recording first goes through the quality steps below, which keep only the stretches the analyses were validated on.
Analyse. Classifiers label the recording moment by moment, and the labels are summarised into the measures in the table below and, where reference data exist, positioned on normative charts. For MAIJU, typical analysis time ranges from ~10 seconds (for an hour-long recording) to ~10 minutes (for a multi-day recording).
Collect. Results per recording and per participant, ready to take into a statistical package. Export formats: .pdf, .csv, .sbem (binary). Accessable via website or API.

You keep your raw data. Garments, sensors and Tessalogger give you the raw recordings with or without a subscription; analytics is what turns them into the measures below.

Run on every MAIJU recording

Home recordings are not laboratory recordings: the suit comes off for a nappy change, a sensor loses contact, a parent picks the infant up. Three detectors run before any analysis, so that these moments are left out rather than mistaken for the infant's own behaviour.

Wear and quality detection

Finds the stretches with usable data. Stretches where more than one sensor has dropped out are excluded; the classifiers are trained to tolerate a single missing sensor with little loss of accuracy.

Playtime detection

Finds the free-playtime segments inside a long recording, which is what the motor analyses are computed on. Trained on 4021 hours of home recordings against the times parents reported as free play.

Carrying and holding detection

Separates being carried or held by an adult from the infant's own movement, at 96% accuracy against video annotation (Scientific Reports 2024).

A MAIJU recording counts as successful when it holds at least 30 minutes of detected free playtime with data from at least three sensors. An hour or more is better: the motor score settles once a recording passes about an hour, and the published recordings averaged about two hours of playtime.

Analyses

The analyses you can order depend on the garment, because the garment decides where the sensors sit. Each analysis has its own page with what it returns, how it is derived and how it has been validated.

Analyses available for each garment
Analysis	Ages	Publications	MAIJU	NAPPA
BABA Infant Motor Score (BIMS) Extensively validated score for quantifying infants' gross motor development	5-19 months	7	Available	Not available
Posture and movement profile Second-by-second classification based on multi-sensor data.	4-24 months	7	Available	Not available
Gross motor growth charts Motor development plotted on age-normative reference curves, read like a physical growth chart	5-19 months	4	Available	Not available
Sleep detection Wake, light and deep sleep through the night, from breathing and movement	0-18 months	1	Not available	Available
Respiration, arousals, and position Breathing rate, sleep position, activity and arousals through the night	0-24 months	3	Not available	Available

Publications counts the peer-reviewed papers on each analysis in the research list.

What you get back

Per recording

For MAIJU: the BABA Infant Motor Score, a centile and z-score for each of the 220 growth-chart metrics, and the posture and movement profile. For NAPPA: sleep and wake states, respiration rate, body position and activity across the night. For both: how much usable data the recording held.

Per participant

Repeated recordings of the same child line up as a trajectory against the normative charts, so change over months is read the way a paediatrician reads a growth curve.

Per study

A spreadsheet (.csv) containing all the relevant metrics per recording.

What it costs

Garments and sensors are bought once; the analytics is a subscription. There is no price list. Every quote is put together for the study in question — which garments, how many participants, which analyses, and over how long — so tell us the protocol in the quote request and we will send pricing.

Getting started

Tell us the protocol. Garment, number of participants, ages and the measures you need — the quote request asks for exactly this.
Set up the account.You will receive your username and password after purchase.
Run a pilot recording.Pilot analyses can be run free of charge.

Data protection

Recordings of infants at home are health-related personal data. The short answers a data protection officer will ask for:

Where data is stored: Hosted by Microsoft Azure servers, physically located in Sweden. Tessacloud does not see, process, or store any sensitive patient data. All analytics process only raw IMU sensor data (legally seen as anonymous) and unidentifiable metadata (e.g., age in months).
Controller and processor: The customer is the data controller and Momega Ltd is the data processor.
Encryption: Data is transferred securely over the https protocol. The data is stored in encrypted databases in the cloud.
Retention and deletion: Customer data is stored for 1 year after a subscription ends. Users can delete their data at any time.
Access: Only users with the highest security levels at Momega Ltd can see customer data. All access instances leave an audit log trace that is visible to the customer.

Shorter answers are on the FAQ; how this website handles your own details is in the privacy notice.

Request a quote See the evidence