The clinical layer
Mendi has been waiting for
Mendi's headband is exceptional consumer hardware. But licensed neurofeedback clinicians need more than a progress graph — they need clinical oversight, session documentation, and outcome reporting. That's EEGBase.
Book a meeting
30 minutes · live demo · your team
We’ll walk through the demo, answer the 3 things you’d most like answered, and end with a concrete next step. No slide deck unless you ask.
Email us with three windows that work for you. We reply same day.
Self-serve calendar coming soon. Email gets you a slot in under a day.
The gap
What clinicians need that the Mendi app doesn't provide
The Mendi app is designed for consumers. Clinicians who prescribe or supervise Mendi-based neurofeedback sessions are flying blind — no caseload view, no documentation, no clinical accountability.
- 🗂️One dashboard showing every client and every session
- 📊See the full brain data — not just a single score
- 📋Track changes in mood, focus, anxiety, and energy over time
- 📝Write clinical notes tied to each session
- 🤖AI-generated clinical summaries for documentation efficiency
- 📤Shareable progress reports for clients, families, and referrers
- 🏥Billing codes, scheduling, and outcome measure tracking
- 💾EDF+ / CSV / PDF export for research and insurance
- 🔒HIPAA-friendly U.S.-hosted architecture — clinic-owned data, full export anytime
Partnership models
Three ways to partner
From a lightweight referral arrangement to a fully white-labeled B2B product — we can structure this however creates the most value for Mendi.
Referral program
Mendi recommends EEGBase to clinicians who ask. EEGBase recommends Mendi to ours. Both sides get a referral cut. No code changes needed.
- Affiliate link in Mendi Pro onboarding
- EEGBase referral link to Mendi hardware
- Revenue share: ~10–15%
- Zero engineering required
Deep integration
Mendi sends session data to EEGBase automatically. Clinicians see every client's sessions in one dashboard — even when the client trains alone at home.
- Mendi BLE SDK streams session data into EEGBase
- Clinician dashboard pulls live session data
- Per-clinician explicit consent in the Mendi app — revocable at any time (GDPR Art. 9 / HIPAA authorization)
- Joint Controller Agreement signed with Mendi before any EU at-home data lands in a clinical view
- Engineering effort split between Mendi (SDK build) and EEGBase (adapter wiring)
White-label portal
EEGBase powers a "Mendi for Clinicians" branded dashboard. Mendi gets a B2B SaaS product and recurring revenue stream. EEGBase gets distribution to Mendi's clinician user base. Full co-brand or full white-label.
- Mendi-branded clinician portal
- B2B SaaS revenue for Mendi
- EEGBase provides ongoing development
- Revenue split: negotiable
Signal architecture
Built on Mendi signals
EEGBase already understands every signal Mendi broadcasts. Here's what each one tells clinicians.
Hardware: 660 + 805 nm wavelengths · ~31 Hz sampling · Brodmann area 10 (Fp1 / Fp2) · 2 long channels + 1 short-separation channel for skin-blood-flow regression.
Blood flow on the left side of the forehead. Goes up when someone is focused. ADHD clinicians track this during sessions.
Blood flow on the right side of the forehead. Tells you about emotional control and how alert someone is. Left-right differences matter clinically.
Tracks when oxygen is being used up by neurons. Helps confirm whether the brain is actually working harder, not just getting more blood.
Same signal on the right side. Used to compare left and right brain activity — useful for anxiety, stress, and trauma work.
Composite metric computed from OxyHb bilateral balance. Used for real-time neurofeedback. EEGBase preserves raw samples for clinical post-processing.
Integration requirements
One afternoon of
engineering time
Deep integration requires only three values from Mendi's firmware team. Once we have them, EEGBase can connect directly to the Mendi headband via Web Bluetooth — no Mendi app required for clinical sessions.
We've already reverse-engineered the BLE advertisement and connection flow. The values below are the only missing pieces. This is one afternoon of work for a firmware engineer.
The 3 values we need
BLE Service UUID
The primary GATT service UUID for the Mendi data stream
e.g. "6e400001-b5a3-f393-..."
Characteristic UUID
The notify characteristic that broadcasts fNIRS packet data
e.g. "6e400003-b5a3-f393-..."
Packet format
Byte layout of each packet: offsets for OxyHb L/R, DeoxyHb L/R, reward score, timestamp
e.g. [0-3]=OxyHb_L float32, [4-7]=OxyHb_R...
About EEGBase
Built by clinicians,
for clinicians
EEGBase is a clinician-built neurofeedback practice-management platform. Free for licensed clinicians. Hosted by us on HIPAA-friendly U.S. infrastructure (Vercel + Neon, AWS us-east-1).
Patient data is clinic-owned. Standard exports (BIDS-fNIRS, SNIRF, EDF+, CSV, PDF) anytime. No vendor lock-in, no third-party cloud access by default. HIPAA-friendly architecture designed with clinical privacy as the primary constraint — not an afterthought.
Pricing
Free for licensed clinicians — no card, no per-seat fee
Stack
Next.js 16, PostgreSQL (Neon), TypeScript
Hosting
Vercel + Neon · U.S. East · HIPAA-friendly architecture
Data residency
U.S. by default · clinic-owned · full export anytime
Devices
Mendi fNIRS, Muse EEG, Polar HRV, Apple Watch, Oura, extensible adapter API
AI features
Claude Haiku for clinical summaries — opt-in only
Try it now
Explore the full clinical dashboard
The /demo route opens straight into a working clinical dashboard with 5 sample clients across ADHD, PTSD, burnout, sleep, and performance archetypes — including 20-session arcs, SOAP notes, outcome measures, and a live Mendi fNIRS simulator. No install, no signup.
Open the demo
No login or signup required. Click below and you land directly inside the clinical dashboard with the Mendi simulator running.
- 11 tabs · Live Session, Brain Map, AI Insights, Reports, Compare
- 5 sample clients · Mendi-only, hybrid, EEG-cap archetypes
- 20-session learning arcs with PHQ-9 / GAD-7 outcomes
- Live Mendi fNIRS simulator · 660 + 805 nm · BA10 · 2 long + 1 short