TECH & SAFETY
Tech & Safety
"How do you measure focus without video?" —
the technical answer to our most-asked question.
MEASUREMENT PRINCIPLE
Research-grade optics, desk-top size
Laboratory eye trackers estimate gaze from the relationship between tiny corneal reflections created by infrared illumination and the pupil. We put the principle behind equipment costing thousands of dollars into a palm-sized cube.
The camera and infrared LEDs sit at the bottom of the cube's front face for the same reason — minimizing the angle between the camera and eyes looking down at a book, keeping the optical axis as close to the child's eye line as possible.
A specialized camera that reads motion and pupils
A motion-robust infrared sensor paired with our integrated analysis algorithms — no distortion even when the child moves their head.
Validated optical design
Scientifically validated down to the number and control scheme of its LEDs — our own design that stabilizes pupil detection through glasses reflections, off-angles, and lighting changes. It also automatically detects whether glasses are worn and accounts for it in analysis.
Real-time on-device
Camera frames are analyzed on the spot inside the device and reduced to focus-related numbers only — the video vanishes immediately.
DATA BY DESIGN
The video doesn't exist — by design
What reaches the server is a time series of numbers (about 2–4MB per session). Facial video never leaves the device, and the server was built with no capability to receive video.
| Sessions | 20 / 30 / 40 minutes — 30-second gaze calibration first |
|---|---|
| Recording | Through analysis of eye movement and pupil response, only scientifically grounded 'focus-related data' is recorded |
| Transfer | Encrypted (HTTPS) upload after the session — nothing is transmitted during measurement |
| Personal data | Name (nickname) and birth date only — used for age-appropriate interpretation; bulk deletion available in the app |
| Video | Not stored · not transmitted · server cannot receive it |
READING RHYTHM
Focus is read as the rhythm of the eyes
Focused eyes don't sit still. They travel along lines of text, sweep back at line breaks, alternate between a problem and scratch paper.
concentration-cube analyzes this reading rhythm, not just gaze position — aiming to distinguish even "looking at the book but not reading."
And when the camera loses the eyes, that interval is never counted as distraction — separating measurement failure from focus loss is a core design rule.
EYE SAFETY
Infrared eye safety — layered twice, three times
Dual output limits
A hardware current ceiling plus software limits. If one fails, the other holds.
Proximity cutoff
Output drops or measurement stops automatically when the face gets too close.
International standard
IEC 62471 photobiological safety assessment is a mandatory gate before mass production.
Minimum distance
A 30cm+ use distance is guided in the product and the app.
Note: we validate safety under specified conditions — and never use unconditional absolute-safety claims. That is our labeling principle.
The validation study — an exam with an answer key
Before launch, a validation study will run under hospital IRB approval.
Record the answer key
Participants study naturally, then record when they truly focused, zoned out, or did something else.
Score the device
That record becomes the answer key against which the device is scored. Success criteria are pre-registered and never changed.
Publish honestly
Results are published with their limitations — saying what works and what's still hard is what trust means to us.
WHAT IT DOES NOT DO
What this device does NOT do
It doesn't diagnose
Not a medical device; it judges no condition. Only observable study behavior.
It doesn't rate ability
Not intelligence or potential — a record of focus behavior shown in today's session.
It doesn't compare kids
The score exists for comparison with your own past sessions, not with others.