Step 01 of 04 · Patient Input
Log — AI-Assisted Flexible Entry
Live App — Log Entry Screen
Step 01
Three input paths — one patient reality
Patients log in the way that fits their energy level. A full sentence on a good day. A voice note when brain fog makes typing hard. A photo when words aren't enough. The modality adapts to the patient, not the other way around.
AI Free Text Entry
The patient describes — the AI categorizes. Free text input accepts anything: "Very tired," "Had a migraine today," "neck pain started." The AI analyzes the entry against the patient's prior history and clinical patterns, then offers 2–3 category suggestions. The patient chooses — or creates their own.
Input Design
Voice Entry and Attach Photo are first-class options. Not buried in a menu. For a population whose primary symptoms include fatigue and cognitive difficulty, reducing the friction of logging is a clinical intervention in itself — patients who can log easily log more often, producing better longitudinal signal.
Governance
AI analyzes but never assigns silently. The category suggestion flow surfaces after submission — the patient always sees and approves how their entry is categorized before it persists in their timeline.
Try the live app
The full AI logging flow — including category suggestions and the health timeline — is available in the pre-beta deployment.
Open App →
Step 02 of 04 · Unified View
Health Timeline — Patient + Provider Data in One View
Live App — Health Timeline
Step 02
The timeline is the product
Every entry — patient-logged symptoms, provider-added medication changes, visit summaries, lab orders — appears in a single chronological view. This is the core design decision: longitudinal signal over point-in-time snapshots.
Unified Data Model
Patient and provider data in the same view. Today: a medication change to Levothyroxine 50mcg (provider-entered) and a Visit Summary with consultation notes and lab orders. Yesterday: patient-logged symptoms. The relationship between these events — what changed, and what followed — becomes visible in a way it never was before.
Clinical Value
This is what the patient brings to their appointment. Instead of trying to recall three months of symptoms in a 15-minute window, the patient arrives with a scrollable, legible record their provider can actually use. The timeline bridges the gap between self-tracking and clinical care.
Architecture
Built on a flexible per-patient schema in Supabase. No two patients track the same categories — the data model accommodates completely individual symptom vocabularies while keeping provider-entered data structured and consistent for clinical use.
Step 03 of 04 · Pattern Recognition
Health Insights — Symptoms Correlated with Lab Values
Live App — Health Insights
Step 03
Patterns invisible entry-by-entry become visible over time
The Insights view correlates symptom severity with TSH lab values across customizable time windows. A patient can see whether their energy tracks with their thyroid levels — the kind of signal that only emerges from longitudinal data, not individual log entries.
Correlation View
Symptom trends alongside lab values on one chart. Energy, Mood, Sleep plotted against TSH (mIU/L) — dual axis, same time range. The patient can see, for the first time, whether a TSH spike preceded a fatigue week, or whether a dosage change correlated with improved mood. This is the clinical insight most patients never have access to.
Time Windows
7 days, 30 days, 60 days, 90 days, 6 months. The patient chooses the resolution that matches what they're trying to understand. Short windows for acute patterns, longer windows for medication response tracking.
Patient Authority
Customize button lets patients choose which symptoms to display. The chart is built from their personal category library — not a fixed set of symptoms. Energy appears because they log energy. The insight layer is as individual as the tracking layer.
Step 04 of 04 · Closing the Loop
Care Coordination — Provider Handoff
Live App — Care Coordination Dashboard
Step 04
The loop closes — patient data reaches the provider
The Care Coordination Dashboard gives the patient structured tools to escalate signals to their provider. Dosage, lab values, and symptom trends are correlated in a clinician-ready view — built for the 15 minutes at the appointment when everything needs to be legible immediately.
Dosage-Lab-Symptom Correlation
Three data streams on one chart. Dosage (mcg) as bars, Energy and Mood as lines, TSH as a secondary axis. The provider can see — at a glance — whether the patient's subjective experience tracks with their objective lab values, and whether a dosage change produced the expected response.
Export Data
One tap to export the full timeline for clinical use. The patient controls what goes to their provider and when. The export is structured for clinical review — not a raw data dump. This is the bridge from self-tracking to clinical care that most health apps never complete.
Alerts Layer
2 pending alerts requiring action. The system surfaces signals that may warrant provider review — not autonomously escalating, but making the escalation decision legible. The patient reviews and decides whether to escalate. The same governance principle as Billing Intelligence: AI surfaces, human decides.
Full care loop available in the live app
Log symptoms, view your timeline, explore insights, and access the Care Coordination Dashboard — all live on Vercel.
Open App →