Eulerian Magnification of Jugular Pulse

By /

Eulerian Magnification of Jugular Pulse

ISEF Category: Biomedical Engineering

Ready to Turn This Idea Into a Real Project?

This guide was put together with the help of AI research tools to give you a solid starting point. But a competitive science fair project lives in the details: refining your research question, fine-tuning your variables, analyzing your data, and presenting your findings like a seasoned scientist.

For next steps tailored to your interests, skill level, and timeline, work one-on-one with a MehtA+ mentor. Learn more about MehtA+ Science & Engineering Research Mentorship →

Subcategory: Biomechanics  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

Doctors estimate central venous pressure by watching how high the neck-vein pulse climbs. The signal is so subtle that most clinicians find it hard to see. Eulerian video magnification amplifies tiny color changes in a smartphone video, making the pulse visible to anyone. Pair it with a simple 1D hemodynamic model and you have a non-contact pressure estimate.

What Is It?

Eulerian video magnification is a signal-processing technique developed at MIT. It boosts small color or motion changes between video frames. Skin pulsations from blood flow become visible after magnification.

The jugular vein lies just under the skin of the neck. Its pulse height relative to the sternum corresponds to central venous pressure. By magnifying that pulse and timing peaks, you derive a pressure proxy.

A 1D hemodynamic model treats arteries and veins as elastic tubes. Valsalva maneuvers (forced exhalation against a closed glottis) predictably raise central venous pressure. By comparing the magnified pulse during normal and Valsalva conditions to the model's prediction, you validate the pipeline.

Why This Is a Good Topic

Non-contact vitals monitoring is an active clinical area. The math is approachable and the implementation runs in free Python notebooks. You will learn video signal processing, hemodynamic modeling, and validation against a controlled maneuver.

Research Questions

  • How does magnification factor change detected pulse amplitude?
  • What is the effect of room lighting on signal quality?
  • Does Valsalva-induced change match the 1D model prediction?
  • To what extent does skin tone affect signal-to-noise ratio?
  • Which camera frame rate captures the pulse most reliably?
  • How does subject posture shift the apparent pulse height?
  • What is the effect of subject motion on magnification artifacts?

Basic Materials

  • Smartphone with manual-exposure camera app.
  • Tripod or fixed mount.
  • LED ring light.
  • Reflective neck-marker stickers for ROI tracking.
  • Stopwatch for Valsalva timing.
  • Informed-consent form.

Advanced Materials

  • High-frame-rate camera.
  • Pulse oximeter and ECG for ground-truth timing.
  • Standardized clinical exam table with calibrated head-of-bed angle.
  • Clinical mentor.

Software & Tools

  • Python (OpenCV and NumPy): Implements Eulerian magnification.
  • scikit-image: Tracks ROIs across frames.
  • SciPy: Fits the 1D hemodynamic model.
  • FFmpeg: Standardizes video preprocessing.

Experiment Steps

  1. Lock the camera, lighting, and subject posture before recording.
  2. Decide your magnification factor and frequency band before processing.
  3. Build a controlled Valsalva protocol with timed maneuvers.
  4. Plan controls (no Valsalva, breath-hold) to disentangle effects.
  5. Process video and extract pulse amplitude per condition.
  6. Compare amplitude changes to 1D model predictions.

Common Pitfalls

  • Recording under flickering room light, which contaminates the magnified signal.
  • Selecting an ROI that drifts as the subject moves.
  • Pushing magnification too high and amplifying noise.
  • Skipping a within-subject baseline.
  • Treating correlation as agreement without a Bland-Altman analysis.

What Makes This Competitive

A competitive project includes informed consent, an IRB-light protocol, calibrated lighting, repeated trials per subject, and a direct comparison of estimated pressure changes to the 1D model's prediction. Report agreement intervals, not just correlation, and audit fairness across skin tones.

Project Variations

  • Apply the method to wrist instead of neck and compare.
  • Add a head-tilt sweep to vary central venous pressure.
  • Replace 1D model with a multi-compartment lumped-parameter model.

Learn More

  • MIT Eulerian video magnification project page: Free code and reference.
  • PubMed: Search jugular pulse non-contact reviews.
  • NIH PubMed Central: Open-access hemodynamic modeling papers.
  • OpenCV documentation: Free tutorials on video processing.
  • MIT OpenCourseWare: Course 6.869 Advances in Computer Vision.

For next steps tailored to your interests, skill level, and timeline, work one-on-one with a MehtA+ mentor. Learn more about MehtA+ Science & Engineering Research Mentorship →

To discover more projects, visit the MehtA+ Science Fair Project Discovery Hub​ →

Shopping Cart