Body Composition API for Fitness Apps — Complete Developer Guide (2026)

What Is a Body Composition API?

A body composition API is a REST service that accepts a photo alongside basic biometrics — height, weight, age, gender — and returns a structured JSON payload describing the subject's body in clinical terms: fat percentage, lean mass, muscle score, body type classification, posture flags, and circumference estimates.

Before camera-based body composition analysis existed, the only ways to measure body composition were expensive or inconvenient: DEXA scans in a radiology lab, hydrostatic weighing, bioelectrical impedance scales, or manual tape measurements by a fitness professional. A body composition API brings that analysis to any smartphone camera, making it practical at consumer scale.

How Camera-Based Body Scanning Works

Modern body composition APIs use a multi-stage computer vision and statistical pipeline:

• Pose estimation: A model (typically MediaPipe Pose or a fine-tuned equivalent) identifies 33+ skeletal landmarks — joints, limb endpoints, and torso reference points — from the 2D image.
• Anthropometric estimation: Landmark positions are combined with supplied height to calculate pixel-per-centimeter ratios, yielding circumference estimates for waist, hips, chest, thighs, arms, and more.
• Body fat formula ensemble: Multiple validated formulas (Deurenberg, Gallagher, US Navy method) are applied in parallel using the estimated measurements and supplied biometrics. Results are averaged and calibrated against a reference dataset.
• LLM vision pass: A vision model provides qualitative assessment — body type, fat distribution pattern, posture notes, and muscle definition score — adding nuance beyond pure formula outputs.

The combined pipeline ensembles multiple published anthropometric formulas so no single one dominates, and returns a confidence score alongside every estimate. Results are calibrated estimates for fitness tracking, not clinical measurements.

Key Metrics: What You Get Back

A full POST /v1/analyze response includes:

• Body composition (body_composition): body_fat_percentage, body_fat_category, body_fat_mass_kg, lean_mass_kg, body_type, bmi, bmi_category
• Body type: Ecto / Meso / Endomorph classification, plus a text description in ai_insights.body_type_detail
• Scores (scores): muscle_score (0–100), symmetry_score, fitness_score
• Measurements (estimated_measurements): shoulder_width_cm, waist_circumference_cm, hip_circumference_cm, waist_to_hip_ratio
• AI insights (ai_insights): posture_notes, fat_distribution pattern, muscle_definition_score

REST API Integration

The WorkoutX Body Scan API accepts multipart/form-data POST requests to https://amused-mercy-production-aeaa.up.railway.app/v1/analyze. Authentication uses a single header: X-BodyScan-Key, with the bsk_ key from your dashboard.

cURL

curl -X POST https://amused-mercy-production-aeaa.up.railway.app/v1/analyze \
  -H "X-BodyScan-Key: bsk_your_key_here" \
  -F "front_photo=@front_photo.jpg" \
  -F "height_cm=178" \
  -F "weight_kg=75" \
  -F "age=28" \
  -F "gender=male"

JavaScript (fetch)

const form = new FormData();
form.append('front_photo', photoFile); // File object from <input type="file"> or camera capture
form.append('height_cm', 178);
form.append('weight_kg', 75);
form.append('age', 28);
form.append('gender', 'male');

const res = await fetch('https://amused-mercy-production-aeaa.up.railway.app/v1/analyze', {
  method: 'POST',
  headers: { 'X-BodyScan-Key': 'bsk_your_key_here' },
  body: form
});

const scan = await res.json();
console.log(`Body fat: ${scan.body_composition.body_fat_percentage}%`);
console.log(`Muscle score: ${scan.scores.muscle_score}/100`);

Python

import requests

url = "https://amused-mercy-production-aeaa.up.railway.app/v1/analyze"
headers = {"X-BodyScan-Key": "bsk_your_key_here"}

with open("front_photo.jpg", "rb") as f:
    res = requests.post(url, headers=headers, files={
        "front_photo": f
    }, data={
        "height_cm": 178,
        "weight_kg": 75,
        "age": 28,
        "gender": "male"
    })

scan = res.json()
print(f"Body fat: {scan['body_composition']['body_fat_percentage']}%")
print(f"Waist: {scan['estimated_measurements']['waist_circumference_cm']} cm")

Comparison: Camera API vs DEXA vs 3D Scanner

Understanding the trade-offs between body composition measurement methods is essential for choosing the right approach for your app.

Method	Accuracy (Fat %)	Hardware cost	User convenience	API available	Best for
Camera API This	±2–3%	None	Any smartphone	Yes — REST	Consumer fitness apps
DEXA Scan	±1–2%	$100–$200/session	Lab visit required	No	Clinical / research
3D Body Scanner	±1–2%	$10,000–$50,000	Dedicated hardware	Proprietary only	Gyms, clinics
BIA Scale	±3–5%	$30–$300	Good (home)	Rarely	Home tracking
Calipers	±3–4%	<$20	Requires technician	No	Personal trainers

Use Cases Across Fitness Verticals

Body composition APIs are being adopted across every fitness and health vertical:

Consumer Fitness Apps

Replace the manual "enter your body fat %" input with an AI-powered camera scan on onboarding. Show progress over time with side-by-side scan comparisons. Gamify improvement with muscle and fitness scores.

Personal Training Platforms

Enable trainers to conduct remote assessments. Automate the initial consultation body scan and generate a PDF report. Track client progress without requiring in-person measurements.

Telehealth & Digital Health

Augment weight management programs with body composition tracking. HIPAA-ready architecture (no photo retention) makes this deployable in clinical workflows. Combine with exercise API data for holistic wellness scoring.

Corporate Wellness

Run company-wide wellness challenges with body composition benchmarks. Private, no-hardware, accessible from any device — reducing the friction of traditional biometric screenings.

Getting Started with WorkoutX Body Scan API

Getting up and running takes three steps:

1. Start with 5 free scans, then purchase a credit pack at workoutxapp.com/bodyscan.html. Packs start at $39 for 20 scans — no subscription required.
2. Add a camera capture UI to your app. Best results come from a front-facing photo taken 2–3 metres from the camera in form-fitting clothing with even lighting.
3. POST to https://amused-mercy-production-aeaa.up.railway.app/v1/analyze with the X-BodyScan-Key header, the photo (front_photo) and biometrics. Parse the JSON response and display the metrics to your users.

Most developers complete the full integration — including the UI — in under two hours. The API returns consistent JSON across all plans, so switching from the free scans to Starter to Growth requires no code changes.