About Baseline
Your results matter less than where they're headed
Baseline turns the camera and screen you already own into short tests of reaction time, hand coordination, and fine motor control. There is no special equipment and no clinic visit. Every result is saved, so you can watch your own numbers change over weeks instead of staring at a single score.
The idea
A baseline, then the arc
One reaction time of 310 milliseconds doesn't tell you much. Ten of them over two months tell you a lot: whether you're getting faster, whether one hand is lagging, whether bad nights of sleep show up in your numbers.
That's why the first test you take is your baseline, and every test after it is measured against you, not against a population. The results screen is built around that. Each test has its own history, graphs that show change over time, and a detail view for any single run.
What you can take
Three tests, one set of sensors
Each test uses your device's camera and runs in the browser. Three more are shown on the home page as Coming Soon, so you can see where the platform is headed.
Reaction Time
Gesture prompts appear on screen and you copy them with your hands as fast as you can. The camera tracks your hands.
WHAT IT TRACKS
Average, fastest, and slowest reaction, median, and how consistent you were.
Fine Motor Precision
Trace a path with your fingertip while the camera follows your hand. There is no timer, because time pressure can get in the way of fine control.
WHAT IT TRACKS
Accuracy, movement control, how quickly you recover from drifting off the path, and consistency.
Bilateral Coordination
Capture dots with both hands at once. The screen is split down the middle, so each hand works its own side.
WHAT IT TRACKS
Left and right hand scores, symmetry between hands, and how well you alternate.
How it works
Every test follows the same four steps
Welcome
A short explanation of what the test measures and what you'll do.
Practice
A practice round to get comfortable. Nothing from it is scored.
Test
The real round, using the camera to follow your hands.
Results
Your score and a breakdown, saved to your history with the graphs.
Because every test uses the same flow, you only have to learn it once.
Design choices
Small decisions that keep the numbers honest
A test that produces a clean number but measures the wrong thing is worse than a bug, because nothing looks broken. These are some of the calls made to avoid that.
Scores per hand, not one total
A single combined score can hide the fact that one hand is doing most of the work. Bilateral results show left, right, and the symmetry between them.
A split screen
Left-hand targets stay on the left and right-hand targets on the right. Reaching across your body blocks your view, which would turn the test into a measure of eye-hand coordination too.
No timer on the precision test
Time pressure causes stress, and stress affects fine motor control. A precision test shouldn't quietly be a speed test as well.
Framed around what you can do
The same data reads differently as "your accuracy dropped 15%" than as "Movement Control: 78%." Baseline only works if you come back to retest, so results are framed around capability.
A camera view that moves the way you do
Move your hand right and the dot on screen moves right. It took some extra engineering, but it matches how people expect a mirror to behave.
Professional, not gamified
The look is deliberately calm: dark blue, white, and light blue. Research suggests health tools that feel too much like games can be harder to trust.
What Baseline is not
Baseline is not a diagnostic tool and doesn't replace a medical evaluation. It measures how you perform on a given day, on your own device, and shows how that changes. If you have a health concern, talk to a healthcare professional.
What gets saved
The platform saves numbers: scores, timings, and accuracy for each run. On the full version, results are tied to your account, and the database rules only let you see your own. This demo has no accounts at all.
About this demo
What's different here
No sign-in
The full platform asks you to make an account. Here you can go straight to any test.
A sample history
The results screen opens with made-up data for a sample user, so you can see what weeks of progress looks like without taking weeks.
Stays on your device
Tests you take here are added to the sample history and saved only in this browser. The "Reset sample history" button on the results screen clears them.
Who made it
Designed and built by Lily Anderson
Baseline is my AI 201 capstone at SCAD. I'm a designer, not an engineer, so I used AI to write most of the code and spent my time on the decisions: what each test should measure, how results should feel, and when the AI's easy answer was the wrong one.
It took 10 weeks and 20 sessions, from the first prototype to a deployed platform with a real backend and a security review.
Built with