If you run a gym, studio, med spa, plastics practice, wellness clinic, GLP-1 program, or nutrition practice, InBody versus DEXA is a method choice, not a two-box shopping list. InBody is bioelectrical impedance (BIA). DEXA is dual-energy X-ray. Optical 3D is a third method. They do not measure the same physical thing.

Keep the DEXA table when bone density or a research reference is the product. Keep a professional InBody when a locked-protocol trend on a gym or studio floor is enough. Buy optical 3D when the consult has to show a body, inches, and volume in the room. Updated 2 Oct 2026 (PT).

The longer method page is already live: best body scanner for a clinic. Named InBody papers stay on InBody scan accuracy. This URL is the leftover versus-query.

Who should buy which

Keep DEXA when bone density or a research-grade reference is what you sell. Neither BIA nor optical 3D measures bone. Endocrinology, osteoporosis work, and a protocol that has to match a GE Lunar or Hologic table still buy a table.

Keep professional InBody if you already run a column, you lock time of day, food, and same-day training, and you will only use fat mass and lean mass as a trend. The product is a fast result sheet.

Buy optical 3D when the consult has to show shape plus composition without a radiation visit every time the patient returns. ShapeScale is the consult tool on this site: a photoreal mesh, circumferences, regional volumes, a heatmap, and composition trained on GE Lunar iDXA. Styku and Fit3D are the same optical class with a different capture. Their pages stay as they are.

When to keep DEXA

DEXA sends two X-ray energies through the body and separates bone, fat, and lean from how much each energy is absorbed. It is the clinical reference most composition papers use, including the March 2026 ShapeScale white paper (GE Lunar iDXA). The Lunar iDXA product page (checked 2 Oct 2026) describes it as GE HealthCare's system for bone health and body composition. It does not publish a list price.

Keep the table when the paid work is bone, a research protocol, or a baseline you will defend against another DXA. That job needs a dedicated room, a trained operator, and ionizing radiation. You do not put a Lunar iDXA in the consult hallway and scan every GLP-1 or contouring return.

DEXA is not a photograph. The patient leaves with a clinical report. Useful as a reference. Weak as the thing you open on an iPad during the follow-up.

When BIA is enough

InBody is BIA. Current at several frequencies goes through hand and foot electrodes. The device estimates water, then fat mass and lean mass from those resistances. That is not DXA, and it is not optical 3D.

InBody's test page (checked 2 Oct 2026) tells the person to test every 2 to 4 weeks under the same conditions, at the same time of day, at least 3 hours after a meal, and only after they have recovered from exercise (they write 1 to 3 days). That protocol is the product. If staff will not lock it, the number moves more than the program.

Keep the column if the room is a gym or studio, staff already know the handles, and the product you sell is a locked-protocol trend or a 4-week challenge printout. Same class as Evolt 360.

The March 2026 white paper also ran an InBody Dial H20, a consumer BIA scale, against the same GE Lunar iDXA: body-fat mean absolute error (MAE, the average size of a miss) 5.01 percentage points, R² 0.44. That is a home scale in the study, not a 770. This page does not invent a professional 770 MAE. Named papers stay on InBody scan accuracy.

When optical 3D is the consult tool

Optical 3D captures surface geometry. Composition is estimated from the mesh, not from electrical resistance and not from X-ray attenuation. It does not measure bone. It does not replace a DEXA table for osteoporosis or a research protocol that has to match a Lunar or Hologic printout.

What it can do in a hallway, which a DEXA table usually cannot, is sit in the room and show the body. No ionizing radiation. A retest the staff will actually run. A photoreal mesh the patient will look at next to a photo.

ShapeScale predicts composition from the full 3D mesh with a deep-learning model trained on paired scans versus GE Lunar iDXA. March 2026 white paper, n=1,000+ adults, no participant overlap across train, validation, and test: body fat MAE 1.89 percentage points, R² 0.91. Lean mass MAE 3.3 lb (1.49 kg), R² 0.98. Appendicular lean mass (ALM, lean tissue in the arms and legs) MAE 2.1 lb (0.97 kg), R² 0.96. Those figures are the abstract. This page does not use the Results lean-mass figure from that paper. Limitations in that paper are clothing, hair, and stance, not hydration.

That is a ShapeScale number versus iDXA. It is not a Styku number, a Fit3D number, or an InBody 770 number. Those units were not on that iDXA protocol. Do not invent one.

Styku and Fit3D ProScanner are the same optical class with a different capture: the person turns on a plate in front of a fixed camera. Their composition is often a girth equation. Those comparisons stay on ShapeScale vs Styku and Fit3D. Fit3D SNAP is tablet photogrammetry. There is no SNAP-versus-DXA paper on this page.

What each method shows

Method

DEXA: dual-energy X-ray. Bone, fat, and lean from attenuation. GE Lunar iDXA is the reference in the March 2026 paper.

InBody: multi-frequency BIA. Current through hand and foot electrodes.

ShapeScale: optical 3D. The person stands still. The sensor head moves. Composition comes from the full mesh, trained on paired scans versus GE Lunar iDXA.

What you show in the room

DEXA: a clinical report. Bone plus fat plus lean. Usually a referral out, not a hallway retest.

InBody: a result sheet. Weight, estimated fat mass, estimated lean mass, water, a visceral-fat estimate.

ShapeScale: a photoreal 3D body on iPhone or iPad, circumferences, regional volumes, a volume heatmap, and a PDF of one scan or a two-scan compare. The first scan uses a QR code from the app. Later scans can start from face and body recognition. The app is iOS only. Android is not available.

What DEXA still owns

Bone density. Neither BIA nor optical 3D replaces that. If the consult is fracture risk or a research bone endpoint, keep the table.

What 3D owns in the consult

Shape. Circumferences and regional volume on a body the patient can rotate. A heatmap that points at the treated abdomen, flank, thigh, arm, or buttock. Composition you can walk without a radiation consent on every visit.

ShapeScale circumferences from the mesh: neck, shoulders, chest, biceps, waist, hips, thighs, calves. Volumes: arms (one combined arm volume), torso, legs. Composition that ships: fat mass, lean mass, lean mass index (LMI, lean mass relative to height), appendicular lean mass, visceral fat as a mass, Shape Score. Height is typed in. ShapeScale has no posture score and does not measure skin quality. Keep photos for texture. Keep a Fit3D ProScanner if posture is the product you sell.

Body fat versus iDXA

DEXA is the reference. ShapeScale in the March 2026 abstract: MAE 1.89 percentage points, R² 0.91, n=1,000+ versus GE Lunar iDXA. InBody Dial H20 in that same paper: MAE 5.01 percentage points, R² 0.44. That Dial is consumer BIA, not a 770. Professional InBody and Evolt do not get an invented MAE on this page. Evolt's published claim is repeatability, not a versus-DXA figure this page will stand on.

Lean mass and ALM

DEXA is the reference. ShapeScale in the abstract: lean mass MAE 3.3 lb (1.49 kg), R² 0.98; ALM MAE 2.1 lb (0.97 kg), R² 0.96. InBody lean mass is a BIA estimate from water. This page does not name a 770 lean-mass error.

Best use

DEXA: bone, research, a reference baseline. InBody: a locked-protocol trend on a gym or studio floor. ShapeScale: proof in a clinic consult. Photoreal body, inches, volume, heatmap, composition trained on iDXA. This page does not become a second copy of the method guide.

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.

Dr. Johnny Franco (Austin Plastic Surgeon). Source: ShapeScale vs Styku

ShapeScale is like having a DEXA scan in your studio, but faster, less intimidating, and far more engaging.

Thurston Pym (Resistanz Studio). Source: Resistanz Studio case study. That line is a consult analogy. ShapeScale is not a DXA table. It does not measure bone.

If the versus-query turns into a buy: the 12-month plan is $199/mo billed annually ($2,388/year) or $299/mo on monthly billing. Purchase is $9,990 plus shipping. US shipping is a $199 flat delivery and setup fee. See pricing. Book a demo.

FAQ

Is InBody a DEXA scan?

No. InBody is BIA. Current through hand and foot electrodes estimates water, fat mass, and lean mass. DEXA is dual-energy X-ray and separates bone, fat, and lean. They do not measure the same physical quantity.

Does optical 3D replace DEXA?

No. DEXA still owns bone density and the research reference. Optical 3D does not measure bone. Use 3D when the job is a retest in the room: shape, inches, volume, and composition, without ionizing radiation on every visit.

How does ShapeScale compare to DEXA?

ShapeScale is optical 3D. Composition is predicted from the full mesh versus GE Lunar iDXA in the March 2026 white paper abstract: body fat MAE 1.89 percentage points, R² 0.91; lean mass 3.3 lb (1.49 kg), R² 0.98; ALM 2.1 lb (0.97 kg), R² 0.96; n=1,000+. That is not a second DXA machine. Limitations named in the paper are clothing, hair, and stance.

Did you publish a Styku or Fit3D error versus iDXA?

No. The March 2026 paper did not include those units. This page does not invent a Styku, Fit3D, or InBody 770 error. Styku stays on its page. Fit3D stays on the Fit3D review. SNAP has no versus-DXA paper here.

Where are the InBody 770 papers?

On InBody scan accuracy. This page does not reprint that table and does not claim a 770 error.

Is this the method guide?

No. The live method guide is best body scanner for a clinic. This URL stays InBody versus DEXA: when to keep the table, when BIA is enough, and when optical 3D is the consult tool.

Does ShapeScale measure posture or skin?

No. ShapeScale has no posture score and does not measure skin quality. Keep a Fit3D ProScanner if posture is the product. Keep photos for texture.

How much does ShapeScale cost?

The 12-month plan is $199/mo billed annually ($2,388/year) or $299/mo on monthly billing. Purchase is $9,990 plus shipping. US shipping is a $199 flat delivery and setup fee. Both are on pricing (checked 2 Oct 2026).

Sources

• March 2026 composition white paper. Abstract versus GE Lunar iDXA, n=1,000+: body fat MAE 1.89 percentage points, R² 0.91; lean mass MAE 3.3 lb (1.49 kg), R² 0.98; ALM MAE 2.1 lb (0.97 kg), R² 0.96. Same paper, InBody Dial H20 (consumer BIA): body fat MAE 5.01 percentage points, R² 0.44.
• GE HealthCare Lunar iDXA (checked 2 Oct 2026). No public list price.
• InBody Test (checked 2 Oct 2026). Retest every 2 to 4 weeks, same time of day, at least 3 hours after a meal, exercise only after recovery (1 to 3 days).
• InBody scan accuracy
• Best body scanner for a clinic
• Evolt 360
• ShapeScale vs Styku
• Fit3D scanner review
• ShapeScale pricing (checked 2 Oct 2026)
• Book a demo