Skip to content

srinathsankara/scoliosis-brace-coach

BranchesTags

Repository files navigation

Scoliosis Brace Coach AI

AI-powered posture monitoring that helps avoid unnecessary surgery

Python Flask MediaPipe License: MIT

A web application that uses computer vision to analyze posture from photos and videos, track scoliosis treatment progress, and generate clinical reports — running entirely on your local machine with no cloud dependency.

Created by Srinath Sankara

Important Medical Disclaimer — This application is designed to help parents and caregivers monitor scolosis treatment progress between clinical visits. It is not a medical device, does not provide medical diagnosis, and does not replace professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions regarding a medical condition. Never disregard professional medical advice or delay in seeking it because of something you have read or seen in this application. The author assumes no responsibility or liability for any errors or omissions in the content of this application. Use of this application is entirely at your own risk.

What is this? | Who is this for? | Quick Start | How to Use | Features | For Developers

Table of Contents

What is this?

Scoliosis Brace Coach AI is a tool that helps families and clinicians monitor scoliosis treatment between office visits. It answers three critical questions:

  1. Is the brace actually working? — Measures how much posture improves when wearing the brace
  2. Is the curve getting worse? — Tracks asymmetry over time and alerts when progression is detected
  3. Is the brace being worn enough? — Logs wear-time and pressure distribution

The Problem It Solves

Adolescent Idiopathic Scoliosis (AIS) affects 2-3% of adolescents. The standard treatment is bracing combined with physical therapy (Schroth method). But there's a gap:

  • Clinic visits happen every 4-6 months
  • Curves can progress 1-2 degrees per month during growth
  • By the time progression is noticed at a clinic visit, it may be too late to avoid surgery
  • Spinal fusion surgery costs $100,000–$300,000 and permanently limits mobility

This tool fills that gap by giving families objective, quantitative data at home — so problems are caught early and treatment can be adjusted before surgery becomes necessary.

The Evidence

The landmark BrAIST trial (New England Journal of Medicine, 2013) proved:

  • Bracing reduces the risk of surgery by 72%
  • The more hours per day the brace is worn, the better the outcome
  • 13+ hours/day showed a 90% success rate

This tool helps families achieve those compliance numbers by making progress visible and measurable.

Who is this for?

For Patients and Parents

"I want to know if the brace is actually helping my child."

  • Upload a photo after each brace fitting to see exactly how much correction is being achieved
  • Track posture improvement week by week with easy-to-read charts
  • Get alerted if metrics worsen — before the next scheduled clinic visit
  • Download PDF reports to bring to orthopedic appointments
  • Stay motivated by seeing measurable progress

For Orthopedic Specialists and Orthotists

"I need objective data to make better treatment decisions."

  • Review patient data remotely through the clinician dashboard
  • See brace pressure distribution to verify corrective force at the curve apex
  • Identify patients with progression alerts who need urgent attention
  • Export clinical-grade PDF reports for documentation and insurance
  • Make data-driven decisions about brace adjustments vs. surgical referral

For Physical Therapists (Schroth-Certified)

"I need to verify my patients are doing their exercises correctly."

  • Monitor exercise form between in-person sessions
  • Track posture metrics to validate treatment effectiveness
  • Compare with-brace vs. without-brace sessions to demonstrate brace value
  • Use trend data to adjust exercise programs

Quick Start

Prerequisites

  • Python 3.11 or higher (download Python)
  • pip (comes with Python)
  • A webcam or smartphone camera

Installation

Option 1: Automatic (Windows)

git clone https://github.com/yourusername/scoliosis-brace-coach.git
cd scoliosis-brace-coach
install.bat

Option 2: Manual (All Platforms)

git clone https://github.com/yourusername/scoliosis-brace-coach.git
cd scoliosis-brace-coach
pip install -r requirements.txt

Running the Application

python app.py

Or on Windows, double-click start.bat.

Then open your browser to: http://127.0.0.1:5000

First run note: The application will automatically download the AI pose detection model (~5 MB) from Google Cloud Storage. This only happens once.

How to Use

Step 1: Take a Photo or Video

Stand or walk directly facing away from the camera (posterior view). The person's entire body should be visible from head to ankles.

Good photo:

  • Person centered in frame, facing away from camera
  • Good lighting, plain background
  • Fitted clothing so body contours are visible

Bad photo:

  • Person facing the camera or at an angle
  • Shadows or cluttered background
  • Loose clothing hiding body shape

Step 2: Upload and Select Options

  1. Click Choose File and select your photo or video
  2. Select the Session Type:
    • Standing (Without Brace) — baseline measurement
    • Standing (With Brace) — measures brace correction
    • Walking (Without Brace) — gait analysis
    • Walking (With Brace) — braced gait analysis
    • Schroth Exercise — exercise form check
  3. Select the Age Group (Under 12, Under 15, Under 18, or Adult)
  4. Click Analyze Session

Step 3: Review Results

The analysis typically takes 2-5 seconds. Results include:

  • Shoulder Asymmetry — Height difference between left and right shoulders (pixels)
  • Hip Asymmetry — Height difference between left and right hips (pixels)
  • Trunk Lean Angle — How much the torso tilts to one side (degrees)
  • Head Tilt — How far the head deviates from center (pixels)
  • Spine Deviation — Lateral shift of the spine (pixels)
  • Trunk Rotation — Estimated vertebral rotation (degrees)
  • Rib Hump Proxy — Estimated rib prominence asymmetry (pixels)
  • Rotation Risk Score — Composite score from 0-100 (higher = worse)

Each metric shows a status badge: green for good, yellow for needs improvement.

Step 4: Compare Sessions

Go to Compare Sessions to see side-by-side how the brace improves posture. This is the most powerful feature — it shows exactly how many degrees of correction the brace provides.

Step 5: Track Trends

Go to Trends to see charts of all metrics over time. The system automatically detects:

  • Stable — metrics are consistent (good!)
  • Improving — asymmetry is decreasing (treatment working)
  • Worsening — asymmetry is increasing (may need attention)

Step 6: Export Reports

Go to Clinician Dashboard and click Export PDF to download a clinical-grade report with:

  • Session summary
  • Brace effectiveness percentage
  • Longitudinal trend analysis
  • Progression alerts
  • Evidence-based treatment rationale
  • Recommendations

Bring this PDF to your next orthopedic appointment.

Example Results

How the Tool Works

The tool uses MediaPipe pose detection to find 33 body landmarks, then calculates asymmetry metrics:

Upload Photo -> Detect Landmarks -> Calculate Metrics -> Compare Thresholds -> Report

Key Measurements

Shoulder Asymmetry - Height difference between left and right shoulders:

Left Shoulder *=================* Right Shoulder
              | delta_h (pixels) |

Trunk Lean Angle - How much the torso tilts:

     * Mid-Shoulder
    /|
   / | delta_x
  / theta
 *----* Mid-Hip
theta = arctan(delta_x / delta_y)

Trunk Rotation - Vertebral rotation from shoulder vs hip angles:

Rotation = |shoulder_angle - hip_angle|

Real Analysis Examples

Image: scoliosis_back2.jpg (Openverse CC-licensed photo) Mode: Standing (Without Brace) Age Group: Under 15

Metric Value Status What It Means
Shoulder Asymmetry 19 px NEEDS_IMPROVEMENT Shoulders uneven
Trunk Lean Angle 14.04 degrees NEEDS_IMPROVEMENT Lateral lean present
Trunk Rotation 7.27 degrees NEEDS_IMPROVEMENT Vertebral rotation present
Rotation Risk Score 57.3/100 NEEDS_IMPROVEMENT Moderate-high risk

Interpretation: Moderate asymmetry with shoulder imbalance, trunk lean, and rotation. Risk score of 57.3 suggests clinical evaluation needed.

Image: scoliosis_back3.jpg (Openverse CC-licensed photo) Mode: Standing (Without Brace) Age Group: Under 15

Metric Value Status What It Means
Shoulder Asymmetry 27 px NEEDS_IMPROVEMENT Shoulders significantly uneven
Trunk Lean Angle 6.98 degrees - Mild lateral lean
Head Tilt 59.5 px - Head significantly tilted
Spine Deviation 75.5 px - Significant spine shift
Rotation Risk Score 30/100 GOOD Low-moderate risk

Interpretation: Significant shoulder and hip asymmetry with notable head tilt and spine deviation. Combined patterns suggest scoliosis requiring clinical assessment.

How to Read Results

  1. Status Badges:

    • GOOD = Within normal range for age group
    • NEEDS_IMPROVEMENT = Outside normal range - monitor closely

  2. Risk Score (0-100):

    • 0-29: Low risk
    • 30-59: Moderate risk
    • 60-100: High risk - seek clinical evaluation

  3. Key Patterns to Watch:

    • Shoulder asymmetry > Trunk lean > Rotation (combined indicates scoliosis)
    • High rotation risk score with multiple "needs_improvement" metrics
    • Consistent asymmetry across multiple sessions

For detailed interpretation, see docs/HOW_IT_WORKS.md.

Features

Core Analysis

Feature What It Measures Why It Matters
Pose Estimation 33 body landmarks from a single photo Foundation for all other measurements
Posture Metrics Shoulder, hip, trunk, head, spine asymmetry Primary indicators of scoliosis severity
Rotation Detection Trunk rotation, rib hump, pelvic obliquity Scoliosis is 3D — rotation matters as much as lateral curve
Brace Detection Whether a brace is present in the image Automatically categorizes with/without brace sessions
Gait Analysis Pelvic tilt, step symmetry during walking Walking patterns reveal functional impact

Treatment Monitoring

Feature What It Does Why It Matters
Session Comparison Side-by-side with/without brace analysis Proves the brace is actually correcting posture
Trend Analysis Linear regression across all sessions Detects slow progression that single snapshots miss
Progression Alerts Critical/warning alerts at clinical thresholds Early warning before curve crosses surgical threshold
Age-Specific Thresholds Different standards for under 12/15/18/adult Growing spines have different risk profiles

Brace Intelligence

Feature What It Does Why It Matters
Pressure Evaluation Scores force at 3 correction points Verifies brace applies force at the curve apex
Compliance Tracking Logs wear-time via temperature detection 13+ hours/day = 90% success rate (BrAIST)
Pressure History Time-series of pressure readings Shows if brace fit changes over time

Clinical Tools

Feature What It Does Why It Matters
Clinician Dashboard KPIs, alerts, trend overview At-a-glance view of patient status
PDF Reports A4 clinical reports with evidence citations Ready for doctor visits and insurance
Educational Content AIS treatment evidence and surgery prevention Helps families understand the importance of compliance

How It Works

The Analysis Pipeline

Photo/Video Upload
       │
       ▼
┌──────────────────┐
│  File Validation  │  Check extension, size (max 500MB)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Frame Sampling   │  Videos: extract 10 evenly-spaced frames
│                   │  Images: use directly
└────────┬─────────┘
         │
         ▼
┌──────────────────┐     ┌─────────────────┐
│  MediaPipe Pose   │────▶│  33 Body         │  Normalized (x,y,z) coordinates
│  Landmark Detect  │     │  Landmarks       │  for each landmark
└────────┬─────────┘     └─────────────────┘
         │
         ▼
┌──────────────────┐
│  Brace Detection  │  HSV color analysis on torso region
│  (if applicable)  │  White pixels > 20% = brace present
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Metric Compute   │  6 posture + 6 rotation metrics
│  (age-adjusted)   │  Thresholds vary by age group
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Store & Display  │  Save to SQLite, render results page
└──────────────────┘

Metric Computation

Posture Metrics (from posture_rules.py):

Metric Calculation Units
Shoulder Asymmetry |left_shoulder_Y - right_shoulder_Y| pixels
Hip Asymmetry |left_hip_Y - right_hip_Y| pixels
Trunk Lean Angle arctan2(mid_shoulder_X - mid_hip_X, mid_hip_Y - mid_shoulder_Y) degrees
Head Tilt nose_X - mid_shoulder_X pixels
Spine Deviation mid_shoulder_X - mid_hip_X pixels
Arm Hang Diff |(left_shoulder_Y - left_wrist_Y) - (right_shoulder_Y - right_wrist_Y)| pixels

Rotation Metrics (from rotation_rules.py):

Metric Calculation Units
Rib Hump Proxy |left_shoulder_width - right_shoulder_width| pixels
Axillary Fold Diff |left_elbow_Y - right_elbow_Y| pixels
Trunk Rotation Angle |shoulder_line_angle - hip_line_angle| degrees
Scapular Winging Diff |left_scapular_dist - right_scapular_dist| pixels
Pelvic Obliquity |left_hip_Y - right_hip_Y| pixels
Rotation Risk Score Weighted composite (0-100) score

Age-Group Thresholds

Metrics are compared against age-specific thresholds. What's "normal" for a 10-year-old is different from an adult:

Metric Under 12 Under 15 Under 18 Adult
Shoulder Asymmetry 12 px 15 px 18 px 20 px
Trunk Lean Angle 2.5° 3.0° 3.5° 4.0°
Rib Hump 5 px 7 px 9 px 10 px
Axillary Fold 4 px 5 px 6 px 7 px
Rotation Angle 3° 4° 5° 5°

Progression Alerts

The system uses linear regression to detect trends across sessions:

Alert Level Trigger Meaning
Critical Metric changes by ≥ threshold Significant progression — consult specialist
Warning Metric changes by ≥ threshold Worsening trend — schedule follow-up
Stable Slope < 0.1 No significant change (good!)

For Developers

Architecture

scoliosis-brace-coach/
│
├── app.py                          # Flask application (routes, job management, DB init)
│
├── analysis/                       # Core analysis engine
│   ├── pose_detector.py            # MediaPipe PoseLandmarker wrapper
│   ├── video_processor.py          # Orchestrates the full analysis pipeline
│   ├── brace_detector.py           # HSV color-space brace presence detection
│   ├── posture_rules.py            # 6 standing posture metrics + thresholds
│   ├── rotation_rules.py           # 6 rotation/rib hump metrics + risk score
│   ├── gait_rules.py               # Walking gait analysis (minimal)
│   ├── exercise_rules.py           # Schroth exercise form (placeholder)
│   ├── trend_analysis.py           # Linear regression, alerts, progression reports
│   └── clinician_report.py         # PDF generation with ReportLab
│
├── sensors/                        # Hardware integration
│   ├── ble_scanner.py              # Bluetooth LE device scanning (Bleak)
│   └── brace_pressure.py           # Pressure evaluation, compliance tracking
│
├── templates/                      # Jinja2 HTML templates
│   ├── index.html                  # Home page with upload form
│   ├── results.html                # Analysis results display
│   ├── dashboard.html              # Clinician dashboard
│   ├── history.html                # Session history
│   ├── compare.html                # Side-by-side comparison
│   ├── sensors.html                # Sensor pairing & pressure input
│   ├── trends.html                 # Longitudinal trend charts
│   └── about.html                  # AIS educational content
│
├── static/
│   ├── css/style.css               # Custom styles
│   └── js/compare.js               # Client-side comparison logic
│
├── requirements.txt                # Python dependencies
├── install.bat                     # Windows installer
├── start.bat                       # Windows launcher
├── .gitignore                      # Git ignore rules
└── README.md                       # This file

Tech Stack

Component Technology Version
Language Python 3.11+
Web Framework Flask ≥ 2.3.0
ML/CV Runtime Google MediaPipe Tasks ≥ 0.10.0
Image Processing OpenCV ≥ 4.8.0
Numerical NumPy ≥ 1.24.0
PDF Generation ReportLab ≥ 4.0.0
BLE (optional) Bleak ≥ 0.21.0
Database SQLite built-in
Frontend CSS Tailwind CSS CDN
Frontend Charts Chart.js CDN

Database Schema

sessions.db — Analysis session results

CREATE TABLE sessions (
    job_id     TEXT PRIMARY KEY,    -- 8-char UUID prefix
    created_at TEXT,                -- ISO datetime
    mode       TEXT,                -- standing_no_brace, standing_with_brace, etc.
    result_json TEXT                -- Full analysis result as JSON blob
);

compliance.db — Brace wear-time and pressure data

CREATE TABLE wear_sessions (
    id                   INTEGER PRIMARY KEY AUTOINCREMENT,
    session_id           TEXT,
    started_at           TEXT,     -- ISO datetime
    ended_at             TEXT,     -- ISO datetime
    total_minutes        INTEGER,
    temperature_readings TEXT,     -- JSON blob
    status               TEXT DEFAULT 'active'
);

CREATE TABLE pressure_log (
    id               INTEGER PRIMARY KEY AUTOINCREMENT,
    session_id       TEXT,
    timestamp        TEXT,         -- ISO datetime
    upper_support    REAL,         -- kPa
    middle_pressure  REAL,         -- kPa
    lower_support    REAL,         -- kPa
    skin_temp        REAL,         -- Celsius
    brace_detected   INTEGER       -- 0/1
);

Installation (Development)

# Clone
git clone https://github.com/yourusername/scoliosis-brace-coach.git
cd scoliosis-brace-coach

# Create virtual environment (recommended)
python -m venv venv
source venv/bin/activate        # Linux/Mac
venv\Scripts\activate           # Windows

# Install dependencies
pip install -r requirements.txt

# Run
python app.py

Key Configuration Constants

Constant Value Location Purpose
MAX_CONTENT_LENGTH 500 MB app.py Max upload file size
MODEL_URL Google Storage URL pose_detector.py MediaPipe model download
num_poses 1 pose_detector.py Detect single person
min_detection_confidence 0.5 pose_detector.py Minimum pose confidence
Brace white threshold > 20% brace_detector.py Torso pixel threshold for brace detection
Frame sample count 10 video_processor.py Frames extracted from video
Progression critical 5.0° (trunk) trend_analysis.py Alert threshold
Progression warning 3.0° (trunk) trend_analysis.py Alert threshold
Prescribed hours 23 hrs/day brace_pressure.py Compliance target
Pressure ranges 40-60 / 50-70 / 35-55 kPa brace_pressure.py Correction point ideals
Skin temp baseline 33.0°C brace_pressure.py Brace-worn detection threshold

Running Tests

The application includes an end-to-end test script:

python test_e2e.py

This tests all page routes, API endpoints, pressure evaluation, compliance tracking, and PDF generation.

Extending the Application

Add a new posture metric:

  1. Open analysis/posture_rules.py
  2. Add your metric calculation using detector.get_landmark_coords() for landmark access
  3. Add the metric to the return dictionary
  4. Add a threshold in AGE_THRESHOLDS

Add a new analysis mode:

  1. Open analysis/video_processor.py
  2. Add a new elif branch in process_media()
  3. Create a new rules file in analysis/
  4. Add the mode option to templates/index.html

Add a new API endpoint:

  1. Open app.py
  2. Add a new @app.route() function
  3. Follow the existing pattern for JSON responses

API Reference

Page Routes (HTML)

Route Method Description
/ GET Home page with upload form
/results/<job_id> GET Analysis results display
/sensors GET BLE sensor pairing and pressure monitoring
/history GET Session history table
/compare GET Side-by-side session comparison
/dashboard GET Clinician dashboard with KPIs
/trends GET Longitudinal trend charts
/about GET AIS treatment education

API Routes (JSON)

Route Method Body/Params Response
/upload POST media (file), mode, age_group {job_id}
/status/<job_id> GET — {status, result}
/api/session/<job_id> GET — Full session JSON
/api/trends GET ?mode= (optional) Trend analysis
/api/progression-report GET — Full progression report
/api/pdf-report GET — PDF binary download
/api/pressure/evaluate POST {upper_support, middle_pressure, lower_support, skin_temp} Pressure evaluation
/api/compliance/start POST {session_id} Session started
/api/compliance/log POST {session_id, upper_support, middle_pressure, lower_support, skin_temp, brace_detected} Reading logged
/api/compliance/end POST {session_id} Compliance summary
/api/compliance/summary/<id> GET — Compliance summary
/api/pressure/history/<id> GET — Array of pressure readings

Example: Upload and Analyze

# Upload a photo
curl -X POST http://localhost:5000/upload \
  -F "media=@photo.jpg" \
  -F "mode=standing_no_brace" \
  -F "age_group=under15"

# Response: {"job_id": "a1b2c3d4"}

# Check status
curl http://localhost:5000/status/a1b2c3d4

# Response when done:
# {"status": "done", "result": {"status": "success", "mode": "standing_no_brace", ...}}

Example: Evaluate Brace Pressure

curl -X POST http://localhost:5000/api/pressure/evaluate \
  -H "Content-Type: application/json" \
  -d '{"upper_support": 52, "middle_pressure": 63, "lower_support": 45, "skin_temp": 34.8}'

# Response:
# {
#   "status": "analyzed",
#   "score": 100.0,
#   "points": {
#     "upper_support": {"value": 52, "status": "optimal", "score": 100, ...},
#     "middle_pressure": {"value": 63, "status": "optimal", "score": 100, ...},
#     "lower_support": {"value": 45, "status": "optimal", "score": 100, ...}
#   },
#   "summary": "All correction points are at optimal pressure."
# }

Example: Compliance Tracking

# Start monitoring
curl -X POST http://localhost:5000/api/compliance/start \
  -H "Content-Type: application/json" \
  -d '{"session_id": "patient_001"}'

# Log a pressure reading
curl -X POST http://localhost:5000/api/compliance/log \
  -H "Content-Type: application/json" \
  -d '{"session_id": "patient_001", "upper_support": 50, "middle_pressure": 60, "lower_support": 40, "skin_temp": 34.5, "brace_detected": true}'

# End session and get compliance summary
curl -X POST http://localhost:5000/api/compliance/end \
  -H "Content-Type: application/json" \
  -d '{"session_id": "patient_001"}'

# Response:
# {"total_wear_minutes": 480, "total_wear_hours": 8.0, "compliance_percentage": 5.8, "status": "poor"}

Clinical Background

What is Adolescent Idiopathic Scoliosis?

AIS is an abnormal lateral curvature of the spine that develops in adolescents (ages 10-18) with no known cause. It affects 2-3% of adolescents and is the most common spinal deformity in children.

Treatment Hierarchy

Detection (Cobb angle measured via X-ray)
    │
    ├── Mild curves (< 25°) → Observation + Schroth exercises
    │
    ├── Moderate curves (25°-40°) → Bracing + Schroth exercises
    │   └── Goal: Prevent progression to surgical threshold
    │
    └── Severe curves (> 40-50°) → Spinal fusion surgery
        └── This tool aims to keep patients in the bracing category

Why Bracing Works

The brace applies corrective forces at specific points on the torso to prevent the curve from worsening during growth. The BrAIST trial proved:

Brace Wear Time Surgery Avoidance Rate
< 6 hours/day 41%
6-12 hours/day 60%
13+ hours/day 90%
18+ hours/day 90-95%

Why Schroth Exercises Help

The Schroth method is a scoliosis-specific exercise approach that:

  • Rotational Angular Breathing (RAB) — Targets rotated ribs to derotate the spine
  • Muscle Cylinder Expansion — Activates weak concave-side muscles
  • Postural Retraining — Corrects habitual movement patterns

The 2025 SOSORT guidelines recommend Schroth PSSE over traditional physiotherapy for all curve patterns.

How This Tool Fits In

Patient at home                    Clinic (every 4-6 months)
─────────────                      ──────────────────────────
Upload weekly photos               X-ray + Cobb angle measurement
     │                                    │
     ▼                                    ▼
See brace correction %              Review PDF reports
Track trends over weeks             Compare with previous visits
Get progression alerts              Make treatment decisions
     │                                    │
     ▼                                    ▼
Adjust brace/PT early  ◄────────  Adjust treatment plan
     │
     ▼
Avoid crossing surgical threshold

Contributing

Contributions are welcome! Here's how to get started:

  1. Fork the repository
  2. Create a feature branch: git checkout -b feature/my-feature
  3. Commit your changes: git commit -m "Add my feature"
  4. Push to the branch: git push origin feature/my-feature
  5. Open a Pull Request

Ideas for Contributions

  • Schroth Exercise Validation — Pose classification for specific exercises
  • Real-Time Camera Overlay — WebRTC-based positioning guide
  • Lighting Normalization — Image preprocessing for consistent results
  • Cobb Angle Estimation — ML model trained on paired surface/X-ray data
  • Multi-Language Support — Internationalization for global use
  • Mobile App — React Native or Flutter wrapper
  • Dark Mode — UI theme toggle
  • Unit Tests — Pytest test suite for all analysis modules

License

MIT License — see LICENSE for details.

Disclaimer

Medical Disclaimer

This application is for educational and monitoring purposes only.

This application is designed to assist parents and caregivers in tracking scoliosis treatment progress between clinical visits. It is not intended to be used as a substitute for professional medical advice, diagnosis, or treatment.

  • This application is not a medical device and has not been reviewed or approved by the FDA or any medical regulatory body.
  • The metrics displayed are estimates based on computer vision analysis of 2D images and should not be used as a basis for medical decisions.
  • Always consult a qualified healthcare provider (physician, orthopedic specialist, or physical therapist) for clinical decisions regarding scoliosis treatment.
  • Never ignore professional medical advice or delay seeking it because of information obtained from this application.

Limitation of Liability

The author, Srinath Sankara, and contributors to this project assume no responsibility or liability for any direct, indirect, incidental, special, exemplary, or consequential damages arising from:

  • Use or misuse of this application
  • Any medical decisions made based on the output of this application
  • Any errors or omissions in the content of this application
  • Any interruptions or errors in the operation of this application

By using this application, you acknowledge that you have read this disclaimer, understand its contents, and agree to use this application solely at your own risk.

No Doctor-Patient Relationship

Use of this application does not create a doctor-patient, therapist-patient, or any other professional healthcare relationship between the user and the application author.

Created by Srinath Sankara

Built with care for the scoliosis community.

If this tool helps your family, please consider sharing it with other families affected by scoliosis.

About

AI-powered scoliosis brace monitoring web app using MediaPipe pose estimation. Tracks posture, gait, brace effectiveness, and generates clinical PDF reports. Educational platform for parents and clinicians.

github.com/srinathsankara/scoliosis-brace-coach

Topics

flask ai computer-vision web-app healthcare pose-estimation medical-device scoliosis mediapipe brace-monitoring

Resources

Readme

License

MIT license

Code of conduct

Code of conduct

Contributing

Contributing

Security policy

Security policy
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages