You are comparing InBody and ShapeScale for a clinic or a fitness business. InBody is multi-frequency bioelectrical impedance analysis (BIA). The setup is hand and foot electrodes, a result sheet, and a trend line if the protocol is locked. ShapeScale is optical 3D. The person stands still, the sensor moves, and composition is predicted from the full mesh by a deep-learning model trained on paired scans vs GE Lunar iDXA. This page is for a gym or studio, med spa, plastics practice, wellness clinic, or GLP-1 program. Not the patient looking up "InBody near me."

Who should buy which

• Keep a professional InBody (270 / 570 / 770) if you already run one. Lock time-of-day, fasting, and no same-day training. Use the fat and lean numbers as a trend only. Gyms that sell a 4-week challenge printout already do this.
• Buy ShapeScale if the consult or trainer walkthrough has to show a photoreal body plus girths, volumes, composition trained on iDXA, and a PDF they can take, on iPhone or iPad, in the room.
• Do not treat the InBody Dial H20 number in the March 2026 white paper (mean absolute error, MAE, 5.01 pp vs iDXA) as the MAE of a 770. We ran Dial H20, not 270 / 570 / 770. Independent 770 papers still land several points off DXA, with a systematic fat under-read.

InBody vs ShapeScale

Girths ShapeScale lists today: neck, shoulders, chest, biceps, waist, hips, thighs, calves. Volumes: arms (one combined arm volume in Compare / Body / Heatmaps / PDF), torso, legs. Composition extras that ship: fat mass in lb or kg, lean mass index (LMI), ALM, visceral fat as a mass, Health Score, Shape Score. Height is typed in, not scanned.

What an InBody scan measures

InBody is multi-frequency BIA. Current at several frequencies goes through hand and foot electrodes. The device estimates intracellular and extracellular water, then fat mass, lean mass, and visceral fat from those resistances.

Professional units (270, 570, 770) add more frequencies and a tighter electrode setup than a consumer scale. That improves repeatability under a locked protocol. It does not turn BIA into DXA.

The number still moves with hydration (the largest lever), food, caffeine, alcohol, same-day exercise, menstrual cycle, skin temperature, and electrode contact. A bad scan is usually a protocol miss. Two InBody scans a week apart can disagree more than the treatment you are trying to prove.

How accurate is InBody vs DXA?

DXA is the clinical reference for fat and lean. InBody does not measure the same physical quantity. It estimates composition from water.

The March 2026 white paper compared optical 3D and a 4-point BIA unit (InBody Dial H20) to GE Lunar iDXA on 1,000+ adults:

• ShapeScale (optical 3D, current composition model): body fat MAE 1.89 pp, R2 0.91
• InBody Dial H20 (consumer BIA, same study): MAE 5.01 pp, R2 0.44

Lean mass (white-paper abstract): MAE 3.3 lb (1.49 kg), R2 0.98. ALM: MAE 2.1 lb (0.97 kg), R2 0.96.

We validated Dial H20 (consumer / home BIA), not InBody 270 / 570 / 770, so 5.01 pp is not the MAE of a 770. InBody's own Dial language is 8-point tactile electrodes. The paper labeled the comparator as 4-point BIA. In this post we just call it the Dial H20 we actually ran.

No single 770 MAE is universal. Independent papers on professional InBody still land several percentage points off, with a systematic fat under-read:

• Potter et al., BMJ Nutrition 2022. InBody 770 vs GE Lunar iDXA in 795 fit US Marines. %BF MAE 3.9 pp, bias -3.4 (under-predicts fat). The authors applied a ~3-point correction to get closer to DXA.
• Lahav et al., European Journal of Clinical Nutrition 2021. InBody 770 vs DXA in 226 clinic patients (BMI 18-56). Mean BF% -3.1 pp vs DXA. 95% limits of agreement +2.6 to -8.9. Bias was worst in the normal-weight group.
• Antonio et al., Journal of Functional Morphology and Kinesiology 2019. InBody 770 vs DXA over a 4-week diet. Absolute fat was low vs DXA at both time points, but 4-week change tracked.
• McLester et al., 2018. InBody 230, 720, and 770 vs DXA. All three under-read BF% / fat mass. Research-grade models added minimal benefit over the portable InBody230.
• ESPEN BIA guidance (Kyle et al.): do not use BIA routinely at BMI extremes or abnormal hydration.

If a vendor quotes 3-5% accuracy without naming the reference method, the population, and MAE vs DXA, it is marketing. A 4-6 point body-fat swing is larger than most 4-week GLP-1 or contouring changes you will show a patient.

What operators and coaches run into

Aly Khalifa (Physiqonomics), writing for coaches after a reader's InBody flipped from lost fat to gained fat / lost muscle in a month with no diet or training change:

If your InBody scan says you're 20% body fat, your actual body fat percentage could plausibly be closer to 14% or 30%.

Source: Physiqonomics

That range is from Brewer et al. (InBody 770 vs the 4-compartment model). Group error looks small (~1.5 pp). Individual limits of agreement run about +6 to -10 pp, with some cases near -20. Oliver et al.'s 2026 review of BIA vs 4C, including multiple InBody models, puts individual error in the 7-20 pp band. Only off by 3-4% is the group average.

Even as a trend tool, the error is not a constant offset. Same person, lab protocol, ten scans: fat mass still moved ~3.3 lb (1.5 kg) with no composition change. Minimum detectable change on InBody 230/720/770 under a locked protocol is about 2.1-2.7 pp of fat, or 3.3-5.3 lb (1.5-2.4 kg) of fat mass. A typical 4-week GLP-1 or contouring delta sits inside that noise.

Gyms like Fitness Quest 10 sell InBody as a feeder-program closer and a paid retest, which fits a 4-week challenge printout.

The result sheet vs the consult

InBody's product is a result sheet and an app. US App Store, checked 16 Aug 2026: InBody app 3.9 out of 5 from 1.8K ratings. Google Play, same day: 4.1 out of 5 from 12.9K reviews. The listing says the app is not intended for medical diagnosis or treatment. Public reviews cluster on setup friction and a printout sheet of a bunch of numbers.

How to read the result sheet

A professional 770 can print two papers: a Body Composition Result Sheet and a Body Water Result Sheet. The app and LookinBody are the same numbers, not a second measurement. The scan below is the 770 Jane Doe 2015.05.04 hybrid. Composition blocks and water blocks sit on one paper. Every fat, lean, visceral, and "protein" figure is a resistance / water estimate, not DXA tissue.

Keep four things. The header (ID, typed height, time) is the protocol log. A 09:46 fasted scan and an 18:10 post-class scan are not the same test. Body Composition History is the only block that matches how InBody should be used: a locked-protocol trend. Muscle-Fat (weight / skeletal muscle / fat bars) is the gym-floor money chart as a trend, not as week-four proof. ECW/TBW is the honest hydration flag. It tells you whether a lean bar jumped because of water.

Treat the rest as noise, with a reason. BMI is weight / height². Any scale already printed it. InBody Score is a vendor index InBody created, not a clinical endpoint. Weight Control is arithmetic on the same biased fat/lean estimates, not a prescription. Segmental lean is segmental fat-free mass (water + protein + minerals), not skeletal muscle in the limb. InBody says so. Segmental fat is a derivative of the other composition results. It is not a regional fat scan and not a heatmap. Visceral fat as cm² is a trunk-impedance estimate, not CT / MRI / DXA VAT. Phase angle is a raw 50 kHz electrical index, not "your cells improved." Protein and minerals on the sheet are water-linked compartments, not a blood protein or a DXA bone scan.

A 4-week GLP-1 or contouring delta often sits inside the 770's minimum detectable change under a locked protocol (about 2.1-2.7 pp fat, or 3.3-5.3 lb / 1.5-2.4 kg fat mass). It also sits inside the individual limits already walked above. Group MAE can look small. The consult has to survive the person in front of you. Keep a professional InBody if the protocol is locked and you only need that trend. Do not buy another BIA column if the bottleneck is trust in the consult.

Where InBody falls short for contouring, tightening, red light, and plastics

InBody is the right tool for a locked-protocol fat and lean trend. It is the wrong tool when the patient paid for a local change.

CoolSculpting, RF (radiofrequency) tightening, red light, Emsculpt-class muscle work, and lipo or fat grafting all live on a body region. The printed % can hold still while the waist, hip, thigh, or abdomen changes. Segmental lean on the sheet is segmental fat-free mass (water + protein + minerals). That is not skeletal muscle in the limb, and not a regional fat scan. InBody says so. There is no surface inches line and no local volume. You cannot point at the treated flank on a resistance estimate.

The other failure is session noise. Hydration is the largest lever. Food, caffeine, alcohol, same-day exercise, menstrual cycle, skin temperature, and electrode contact move the number. A 4-week contouring or GLP-1 delta often sits inside the 770 minimum detectable change already on this page (about 2.1-2.7 pp fat, or 3.3-5.3 lb / 1.5-2.4 kg fat mass). It also sits inside individual limits of agreement. Staff then spend the consult defending water, not the applicator.

ShapeScale is the consult layer for those jobs. Optical 3D. Still person. Circumferences come from the mesh: neck, shoulders, chest, biceps, waist, hips, thighs, calves. Regional volumes: arms (one combined arm volume in Compare / Body / Heatmaps / PDF), torso, legs. A tape is one line. Volume is the whole region. That is why plastics training walks lipo and fat grafting on volume, not on a single girth.

The heatmap overlays two scans. Blue is volume down. Yellow to red is volume up. Grey is little change. You rotate the body and point at the treated abdomen, flank, thigh, arm, or buttock. Pair that with the composition trend from the March 2026 white paper (current composition model vs GE Lunar iDXA) when the question is fat vs lean, not whether this zone moved.

What we do not claim: ShapeScale does not grade skin, collagen, redness, or joules. Red light and RF still need photos for texture. Keep the photos. Add girth, regional volume, and a heatmap so the conversation is about the treated area, not a whole-body %. Fit3D also does girths and volumes. Keep Fit3D if posture is the product. See the Fit3D page.

Keep the 770 if the protocol is locked and you only need the trend. Do not buy another BIA column if the bottleneck is showing that the contouring, tightening, light, or surgical plan landed where you said it would.

Dr. Johnny Franco (Austin Plastic Surgeon):

ShapeScale has truly revolutionized how we guide patients through their weight loss journey. It makes it so much easier for them to visualize their transformation and track progress, because it is no longer just about losing weight.

Source: ShapeScale vs Styku

Dr. Jake Deutsch (Oakwood Precision Medicine): The interactive ShapeScale app allows patients have access to the data and scans which certainly helps motivate clients and results in better achieved success. Source: ShapeScale vs Styku

Related: Evolt 360 for clinics (same measurement class). Styku is optical, not BIA: ShapeScale vs Styku.

When InBody is enough

Keep InBody if you already run a tight protocol and only need a trend line. Do not buy another BIA column if the bottleneck is trust in the consult.

Thurston Pym (Resistanz Studio): ShapeScale is like having a DEXA scan in your studio, but faster, less intimidating, and far more engaging.

Source: Resistanz Studio case study

Book a ShapeScale demo

FAQ

Is InBody a 3D scanner?

No. InBody is BIA. Current at several frequencies goes through hand and foot electrodes. ShapeScale is optical 3D. Fit3D and Styku are turntable optical. Do not treat an InBody column as a Fit3D avatar.

Is InBody a DXA scan?

No, and neither is ShapeScale. DXA (here, GE Lunar iDXA in the white paper) is the clinical reference for fat and lean. InBody estimates composition from water. ShapeScale predicts composition from the full 3D mesh with a model trained on paired scans vs that iDXA.

How does InBody get a body-fat number?

From resistance, not from tissue and not from inches. Professional units (270, 570, 770) add frequencies and a tighter electrode setup than a consumer scale. That improves repeatability under a locked protocol. It does not turn BIA into DXA.

Did you validate a 770 in the white paper?

No. The March 2026 white paper ran InBody Dial H20 (consumer / home BIA) vs the same GE Lunar iDXA: MAE 5.01 pp, R2 0.44. We call it the Dial H20 we actually ran. Do not paste 5.01 pp onto a 770.

What do the 770 papers actually say?

No single 770 MAE is universal. Potter 2022: n=795 Marines, %BF MAE 3.9 pp, bias -3.4. Lahav 2021: n=226, mean BF% -3.1 pp, limits +2.6 to -8.9. Antonio 2019: absolute fat was low vs DXA, but 4-week change tracked. McLester 2018: 230 / 720 / 770 all under-read BF%; research-grade models added minimal benefit over the portable InBody230. ESPEN / Kyle: do not use BIA routinely at BMI extremes or abnormal hydration.

What do MAE and R2 mean?

MAE is the average size of a miss versus the reference. Lower is better. R2 says how tightly the estimate tracks the reference. Higher is better. Root-mean-square error (RMSE) weights larger misses more; we use it when a paper reports it and we do not invent an MAE from it.

What do ALM, LMI, and FFM mean?

ALM is lean tissue in the arms and legs, where most skeletal muscle sits. A whole-body lean or fat-free mass (FFM) number is everything that is not fat and can move when someone drinks or eats. ALM is the number a GLP-1 or training program actually wants. LMI is lean mass relative to height. ShapeScale reports ALM and LMI today. InBody reports segmental lean as a BIA estimate. We do not have a ShapeScale-run 770 ALM MAE.

Does ShapeScale do posture analysis?

No. Neither does InBody. Fit3D reports tilt, shift, and ProScanner balance. Keep a ProScanner if posture is the product you sell. See the Fit3D page.

Who should keep a professional InBody?

A clinic or gym that already runs one, locks the protocol, and will only use the number as a trend. Fitness Quest 10-style feeder programs and paid retests fit a 4-week challenge printout.

Who should buy ShapeScale instead?

A gym or studio, med spa, plastics practice, wellness clinic, or GLP-1 program that needs a still-person photoreal mesh, girths, volumes, and composition trained on GE Lunar iDXA, shown on iPhone or iPad during the consult, plus a PDF they can take.

Can staff walk a patient through results in the room?

On InBody, the product is a result sheet and a number app. On ShapeScale, photoreal 3D, girths, volumes, and heatmaps are on iPhone or iPad during the consult. Staff can export a PDF of one scan or a two-scan compare. The app is iOS only. Android is not available.

Does a still on this page mean the consult is iPad-only?

No. The app runs on iPhone or iPad.

Do staff have to tap anything to scan a return client?

No. First scan uses a QR code from the app. After that: rainbow lights, face and body recognition. No app. No buttons. Works across even hundreds of users.

Does the app help personalize a treatment plan?

Yes. AI guidance on goals is in the app. That saves consult time. Compare + heatmap + PDF are still the visual.

Does recognition work with a lot of clients?

Yes, after the first QR scan. Face and body recognition starts the return scan across even hundreds of users.

Can both the provider and the client use the app?

Yes, if staff share or invite. Both sides see the scans on iOS. Staff work from the Clients tab. Android clients stay managed and get the PDF. Android app is not available.

Does the client have to create their own account?

No. Staff can create a managed account (Menu -> Clients -> New Client -> Create). HIPAA-private clients still create their own private account.

Can InBody show CoolSculpting, RF, red light, or surgical change in a treated area?

No. InBody estimates composition from impedance. It does not measure surface shape, local inches, or local volume. ShapeScale reports those girths and regional volumes plus a volume heatmap. It does not measure skin quality.

Sources

• Potter et al., BMJ Nutr Prev Health 2022 (InBody 770 vs iDXA)
• Lahav et al., Eur J Clin Nutr 2021 (InBody 770 vs DXA)
• Antonio et al., 2019 (InBody 770 vs DXA, 4-week diet)
• McLester et al., 2018 (InBody 230 / 720 / 770 vs DXA)
• Kyle et al., ESPEN BIA Part II, 2004
• Aly Khalifa, Physiqonomics (Brewer 770 vs 4C; Oliver 2026 review)
• ShapeScale composition white paper, March 2026
• InBody app, US App Store (3.9 / 5, 1.8K, 16 Aug 2026)
• InBody app, Google Play (4.1 / 5, 12.9K, 16 Aug 2026)
• ShapeScale App Store (4.0 / 5, 43, iOS, 16 Aug 2026)
• ShapeScale vs Styku