# Hyeong Jun Lee (이형준)

Postdoctoral Researcher · Multiscale Fluid Dynamics Lab, Yonsei University · computational hemodynamics · physics-informed AI · wearable biosignals · medical devices

I turn clinical questions into engineering problems and bring back answers clinicians can act on, using simulation, AI and wearable signals.

- First author · DMD 2026 · CMPB 2025 · CMPB 2022 (co-first) · Merit Academic Paper Award 2026
- PINO · 85 patients, 10-fold cross-validation · CFD up to 35 min on 64 CPU cores · 1.78 s on one GPU
- 16 patents (2 registered) · 7 funded projects (participating researcher) · 4 awards

Project Researcher, Yonsei University Health System, 4D CT-FFR project.

[Email](mailto:hjunlee96@gmail.com) · [CV (PDF)](https://hyeongjunlee.me/uploads/resume.pdf) · [LinkedIn](https://www.linkedin.com/in/hyeongjunlee/) · [Google Scholar](https://scholar.google.com/citations?user=1NxM3T0AAAAJ&hl=en) · [ORCID](https://orcid.org/0009-0001-5691-3787)

## Recent

- Sep 2026: Started as a postdoctoral researcher at Yonsei University.
- Aug 2026: Completed the Ph.D. in Mechanical Engineering at Yonsei University.
- 2026: A co-authored paper appeared in CMPB.
- 2026: A first-author paper appeared in DMD.
- 2026: A co-authored paper appeared in DMD.
- 2026: Received the Merit Academic Paper Award from the School of Mechanical Engineering, Yonsei University.
- Dec 2025: The patent “AI-based noninvasive urodynamics test method and apparatus” was registered.
- Nov 2025: The patent “Optimization system and method of AI algorithm for prediction coronary artery lesions based on FFR” was registered.

## The plate

CFD: FFR 0.83, up to 35 minutes on 64 CPU cores. PINO: FFR 0.82, 1.78 seconds on one GPU.

## What I work on

- **[Simulation](https://hyeongjunlee.me/research/#lbm-ctffr)**: Patient-specific CFD from CT, read out as FFR and wall shear stress; also a tricuspid valve device.
- **[Physics-informed AI](https://hyeongjunlee.me/research/#pino)**: Neural operators give a vessel's pressure field in 1.78 s instead of 35 minutes; synthesized 4D CT gives FFR across the cycle.
- **[Wearable biosignals](https://hyeongjunlee.me/research/#viscosity)**: Smartwatch PPG gives blood viscosity and predicts a fluid bolus within 24 h (AUC 0.895, under revision); patents cover blood pressure, glucose and glaucoma.
- **[Devices and translation](https://hyeongjunlee.me/research/#ffr-simulator)**: FFR simulator, tricuspid implant, negative-pressure tunnel, AI urodynamics; 16 patents, 2 registered; parts ship in CARDIOS and EASYCHECK.

## Selected

- 2026 · **Patient-Specific Coronary Flow Field Prediction Using Physics-Informed Neural Operators** · Proc. ASME 2026 Design of Medical Devices Conference (DMD2026) · First author · doi:[10.1115/DMD2026-1091](https://doi.org/10.1115/DMD2026-1091)
- 2025 · **A Physics-Integrated Deep Learning Approach for Patient-Specific Non-Newtonian Blood Viscosity Assessment using PPG** · Computer Methods and Programs in Biomedicine · First author · doi:[10.1016/j.cmpb.2025.108740](https://doi.org/10.1016/j.cmpb.2025.108740)
- 2022 · **Optimization of FFR prediction algorithm for gray zone by hemodynamic features with synthetic model and biometric data** · Computer Methods and Programs in Biomedicine · Co-first author · doi:[10.1016/j.cmpb.2022.106827](https://doi.org/10.1016/j.cmpb.2022.106827)

Full record: [CV](https://hyeongjunlee.me/cv/).

---

Open to collaboration, positions and paper requests.

[Email](mailto:hjunlee96@gmail.com) · [CV (PDF)](https://hyeongjunlee.me/uploads/resume.pdf) · [LinkedIn](https://www.linkedin.com/in/hyeongjunlee/) · [Google Scholar](https://scholar.google.com/citations?user=1NxM3T0AAAAJ&hl=en) · [ORCID](https://orcid.org/0009-0001-5691-3787)

Updated 2026-10-21 · hyeongjunlee.me