SuggiCare, Baselines Explained

How every resident, child or in-home wearer gets their own personal "normal" after ten days.

Document type, Knowledge Sheet Audience, Centre Owners, Care Admins, Floor Staff, Families Version 1.0

What is a baseline?

A baseline is the wearer's own personal "normal" pattern. Things like resting heart rate, breathing rate, activity, sleep, hydration and skin temperature are different for every person. A 92 year old in residential aged care, a four year old in childcare and a 70 year old living independently at home each have their own healthy range. SuggiCare learns that healthy range for each individual instead of comparing everyone to a textbook average.

The simple idea. SuggiCare watches the wearer for ten days, learns what is normal for that person, then only raises an alert when something drifts meaningfully away from their normal, not someone else's.

Why ten days?

Ten days is long enough to see two weekends, a few different sleep patterns, a typical week of meals, activity and rest. It is short enough to get the wearer onto personalised alerts quickly. Research and our own pilots showed that fewer than seven days produces noisy averages, and more than fourteen days delays useful alerts. Ten days is the sweet spot.

The ten day learning window, step by step

Day 0 Setup and pairing

The wearer is paired to a resident, child or in-home recipient using the facility bridge tablet or the in-home setup screen. A starting "sector baseline" is applied straight away (Aged Care, Childcare, In-Home Care, Disability) so the wearer is never unsupervised. This is a safety net only.

Days 1 to 10 Quiet learning

The system collects readings every few minutes and builds a personal profile. During this window:

Critical safety triggers are still active (falls, very low oxygen, very high or very low heart rate, device offline).
Soft alerts (small drifts in hydration, sleep or activity) are suppressed so families and staff are not flooded by harmless variation.
A "Learning baseline" badge appears on the wearer's card.

Day 10 Personal baseline goes live

SuggiCare calculates the wearer's personal range for each metric using a robust statistical method (we trim the top and bottom 5% to ignore one off readings, then take the typical range across day, evening and night). The personal baseline replaces the sector default. The card flips from "Learning" to "Personalised".

Day 11 onwards Continuous tuning

The baseline is not frozen. It updates gently every day using a rolling 30 day window, so it stays current as the wearer's life changes (new medication, seasonal patterns, recovery after illness). Big sudden changes do not move the baseline, they trigger an alert instead.

What the algorithm actually does, in plain words

Collect. Pull readings from the wearer every few minutes and store them with a timestamp.
Group by time of day. Morning, afternoon, evening and overnight are treated separately, because heart rate at 3am is not the same as at 3pm.
Trim the outliers. Drop the most extreme 5% at each end so a single bad reading or a moment of exercise does not skew the picture.
Find the personal range. Calculate the typical low, typical high and median for each metric in each time of day band.
Compare new readings. Each new reading is checked against the wearer's own range for that time of day. Drift outside their range, sustained over several readings, becomes a soft alert. Big drift becomes a critical alert.
Update gently. Each day, yesterday's data is folded into a rolling 30 day window. Old data falls off, recent life is reflected.

What a wearer needs for the baseline to work

Requirement	Why it matters
Wear the device most of each day	The system needs enough readings across morning, afternoon, evening and night.
Charge the device daily	Gaps in data slow the learning window down. The system extends the ten days if a wearer has too many quiet hours.
One wearer, one device	Sharing a wearable contaminates the baseline. Each device is bound to one recipient via QR pairing.
Confirm life events	Staff or family can flag "in hospital" or "on new medication" so the system pauses or resets the baseline.

What happens during the ten days if something goes wrong?

The wearer is never unprotected. Three layers of cover are always running, even on Day 1.

Layer 1, sector baseline. A safe default range based on the wearer's age band and care setting.
Layer 2, hard safety triggers. Falls, no movement for an extended period, very high or very low heart rate, very low oxygen, device offline. These fire from minute one.
Layer 3, escalation. Unacknowledged alerts walk up the configured chain, on shift carer, then care admin, then nominated emergency contact.

Sector differences that the algorithm respects

Setting	Sector default while learning	Special handling
Aged Care, residential	Older adult ranges	Sleep and inactivity weighted heavily.
Childcare	Paediatric ranges	Activity peaks expected, hydration and rest emphasised.
In-Home Care	Independent adult ranges	Routine breaks (missed meals, no movement at usual times) emphasised.
Disability and NDIS	Plan specific defaults	Baseline accepts higher day to day variability.

How families, staff and admins see it

Family view. A simple "Learning" badge until Day 10, then a "Personalised" badge with their loved one's typical ranges shown in plain numbers.
Staff view. The handover card shows which residents are still learning so the team knows soft alerts are intentionally quiet.
Admin view. A dashboard tile shows the percentage of wearers on personalised baselines vs still learning, plus any wearers whose learning was extended.

Frequently asked questions

Can I shorten the ten days? No. Shorter windows produce false alerts and miss real drift. The ten days is a clinical and engineering decision.

What if a wearer was already wearing a device elsewhere? SuggiCare still learns from scratch. Devices and life context vary, so baselines are not portable between systems.

What if a wearer is unwell during the learning period? Staff or family flag the event, SuggiCare extends the learning window so the baseline reflects the recovered state, not the unwell state.

Does the baseline ever get reset? Yes, after a hospital stay, a major medication change, a new diagnosis, or on staff request. A reset starts a fresh ten day window.

Care Alerts is not a medical device. It does not provide medical advice, triage or emergency dispatch. In an emergency, contact your local emergency service. Baselines are designed to surface meaningful drift for human review, not to diagnose.