Software as a Medical Device · Early Detection

For bipolar patients and their doctors — episode prediction from your wearable, in real time.

PolarGuard monitors sleep, heart rate variability, and activity continuously — alerting your care team before a mood episode fully develops.

Early access. No spam. Unsubscribe anytime.
You're on the list. We'll be in touch.
PG POLARGUARD time is clinical

The gap between appointments is where patients get lost.

Bipolar disorder is episodic — but psychiatric care is appointment-based. Clinicians have no visibility into what happens between visits. By the time a patient arrives in crisis, the window to intervene has already closed.

The physiological warning signs are there days before an episode develops. Sleep shortens. HRV drops. Activity shifts. No one is watching.

5–10
Years average diagnostic delay for bipolar disorder
Merikangas et al., 2011
7
Days before mania, activity begins diverging from baseline
Reinertsen et al., 2017
46M
People worldwide living with bipolar disorder
WHO Global Burden of Disease
3hrs
Average sleep reduction before a manic episode begins
Harvey et al., 2009

Wearable data, turned into clinical early warning.

PolarGuard reads from Apple Watch via HealthKit, builds a personal baseline over 30 days, then monitors deviations continuously — no manual logging required.

01
Wear
Patients wear Apple Watch continuously. PolarGuard passively reads sleep duration, HRV, resting heart rate, and step count via Apple HealthKit. No manual input required.
02
Monitor
An AI risk engine builds a personalized baseline over 30 days, then monitors deviations across four biometric streams. A risk score from 0 to 100 updates continuously.
03
Alert
When the risk score crosses a clinician-set threshold, the care team receives an automated alert. Earlier intervention. Fewer crises. Better outcomes.

Built by someone who lives it.

"The hardest part of bipolar disorder isn't the episodes themselves — it's that no one sees them coming until it's too late."

I built PolarGuard because I have bipolar disorder and I know what it feels like to lose weeks of your life to an episode that could have been caught earlier. The physiological signals were there — in my sleep, my HRV, my activity. No one was watching.

PolarGuard is my attempt to give patients and clinicians the early warning system that doesn't exist yet. It's a graduate research project at Fordham University's Gabelli School of Business, developed with clinical research guidance and built on published evidence in bipolar prodromal signal detection.

Alanah Richardson Founder · Fordham University Gabelli School of Business
NSF I-Corps Mid-Atlantic Alumna, 2026
Stage
Functional prototype — iOS app and clinical dashboard live
Technology
Apple HealthKit · AI risk engine · FastAPI · Real-time alerts
Regulatory
Software as a Medical Device (SaMD) — FDA De Novo pathway
Research
Phase I feasibility study in development with NIH-affiliated clinical advisors
Company
PolarGuard Health LLC · Registered 2025 · Order #INC797147
Trademark
USPTO — Class 009, Intent to Use

Interested in PolarGuard?

Whether you're a clinician, researcher, patient, or potential partner — we'd love to hear from you.