What is measured
| Group | Measurements | Where it comes from |
|---|---|---|
| Basics | Height, body volume, leg length | Floor to the top of the head, and the integral of horizontal slices |
| Girths | Neck, chest, waist (narrowest point), abdomen (navel level), iliac crest, hips; and thigh, calf and ankle — left and right separately | Slices one centimetre apart, from feet to head |
| Widths and ratios | Shoulder and hip width; waist-to-height, waist-to-hip and shoulder-to-hip ratios | Straight from the same slices, no equation |
| Body composition | Body-fat percentage, fat mass and fat-free mass, and a second estimate for comparison | The Hodgdon–Beckett equation and the RFM equation, plus the weight you enter |
| Symmetry | Trunk shift over the pelvis, back-surface rotation, and left–right volume of thigh, shank and arm | The pelvic midline, found from the direction of the feet |
| Mass distribution | Centre-of-mass height, share of load on each foot, each region's share of the weight, and mass up the height of the body | Each segment's volume times the density of that tissue |
Any number the scan cannot honestly give is left out
This is what sets the scan apart. A number on a page gets read and the caveat beside it does not, so instead of a doubtful number with a footnote, the number is removed and the reason is printed at the top of the report, before any other figure. Cases that have actually happened:
- Arms held against the trunk: on the scanned surface arm and trunk become one, and no arithmetic separates them. The trunk girths, the body-fat figure and every ratio are then withheld.
- A waist lost in the surface noise: if the dip at the waist is shallower than the scan's own noise, the "narrowest point" is really the worst slice, not the waist. The waist is then not reported; the abdomen and iliac crest, taken at fixed heights, are unaffected.
- Shorts: they fill the gap between the thighs. The scan's own colour finds the hem, the thigh girth is taken below it on skin — and the report says the clothing set the limit.
- A scan that does not match the weight: the scan's volume is divided into the weight. If the density that comes out is impossible for a body — usually loose clothing, or furniture close to the person caught in the scan — nothing but height is reported. Height survives because it is taken from no girth, and a loose coat adds not a centimetre to it.
Body fat: an estimate, with its error band
Body fat is computed from girths and height with the Hodgdon–Beckett equation and is always reported with the equation's own error band — about 3.5 percentage points either way; the band is what should be read, not its midpoint. Beside it sits a second estimate by the RFM method, which needs only height, waist and sex. Two methods that agree are more trustworthy than either alone, and two that disagree tell you something worth knowing.
Two optional inputs make a real difference: neck and abdomen measured with a tape. The scan counts hair, beard and the jawline as neck, and the equation takes the neck as its lean reference, so a scanned neck makes body fat read low. When a tape value is entered it is used, and the figure shows both numbers side by side.
RFM was derived by Woolcott and Bergman in 2018 against DXA in US national survey data (Scientific Reports). ⚠️ The paper itself says it was tested in three US ethnic groups only and cannot be extended to others; it has not been validated in an Iranian population. That is why it is the second estimate, not the main one.
How to take the scan
- A device with LiDARAn iPhone or iPad Pro, with a 3D scanning app that exports OBJ, PLY, STL, GLB, glTF or FBX. The metric scale comes from the sensor itself; a model built from ordinary photographs may not have true dimensions.
- Close-fitting, minimal clothingThe scanner measures whatever surface it sees and cannot tell a shirt from a ribcage.
- Arms about 30 degrees outAn arm held against the trunk merges with it and takes the trunk girths with it.
- Feet hip-width apart, weight on bothStand straight and still and breathe normally; chest volume changes with every breath. Tie long hair back.
- Clear space, floor in the frameNothing within half a metre of you, and the floor in the scan — every height is measured from it.
- Someone walks slowly around youMuch of a scan's error comes from moving during it, not from the sensor.
The report: the numbers drawn on the body
Your 3D model is shown with a ring for each girth drawn where it was measured, so you can see exactly where each number runs from and to. Rings taken at a fixed height — chest, navel level and iliac crest — have their own colour and a "fixed height" label, because nothing found them on your body. Each section of the report carries its chart: where body fat sits against reference categories, how weight divides into fat and fat-free mass, the two sides compared, and how mass is distributed. The printed report includes three fixed views of the model and every chart.
Your scan never leaves your device
The file is read and analysed in your own browser and sent to no server; the site does not store it anywhere either, not even on your device. A 3D model of a body is sensitive data, and it reaches the clinic only if you send the report yourself.
The evidence behind the method
- How accurate are phone and tablet 3D scanners — a systematic review of 13 studies; LiDAR-equipped systems beat a single camera, and the limits are given the same weight as the findings.
- Why waist-to-height ratio beats BMI — a meta-analysis of 31 studies and more than 300,000 adults.
What this scan is not
It is not DXA, not a diagnosis, and no substitute for an examination. The body-fat categories the number is shown against are references, not grades: different authorities draw the lines slightly differently and none of them adjusts for age. The scan is at its best comparing a person with their own next scan under the same conditions — where the errors that repeat cancel out and real change shows through.
Frequently asked questions
What device do I need for a 3D body scan?
A phone or tablet with a LiDAR sensor — an iPhone or iPad Pro — and a 3D scanning app that exports OBJ, PLY, STL, GLB, glTF or FBX. The true scale of the measurements comes from the sensor, which is why a model built only from photographs may not have the right dimensions.
Is my scan file sent anywhere?
No. The file is read and analysed in your own browser and goes to no server. It reaches us only if you send the report to the clinic yourself.
How accurate is the body-fat figure?
It is an estimate with an error band of about 3.5 percentage points either way, and the band is what should be read. Entering neck and abdomen measured with a tape makes it more accurate, because the scan counts hair and the jawline as neck. To follow change, compare a scan with your own next one under the same conditions.
Why are some numbers missing from my report?
Because the scan could not give them honestly, and the reason is printed at the top of the report — arms held against the trunk, a waist whose dip was lost in the surface noise, or shorts filling the gap between the thighs, for example. Repeating the scan as the guide describes usually brings the number back.
Why is entering my weight required?
Because without a weight there is no way to tell whether the scan is the right size. The scan's volume is divided into the weight; a scan with the right shape but a third too big — from loose clothing, say — is caught no other way.
How is the body scan different from the posture analysis?
They complement each other. The posture analysis shows from photographs how the body stands — angles and deviations. The 3D scan shows how big the body is, what it is made of, how symmetrical its two sides are and where its weight sits.