Ryan Vo
Used AI to optimize 3D-bioprinted skin graft scaffolds for burn and chronic wound healing. Published at IEEE ICIEA26 — the IEEE Industrial Electronics Society conference.
Where Ryan Started
His Background
- • 10th grader at St. John's School, Houston
- • Passionate about medicine and healthcare accessibility
- • Nationally ranked junior tennis player
- • Interested in biomimetic bioprinting and regenerative medicine
- • No prior research or programming experience
His Goals
- • Contribute to making medicine more affordable and accessible
- • Explore biomimetic bioprinting in depth
- • Publish research in a peer-reviewed venue
- • Build a competitive edge for college admissions
- • Pave the way for future life-changing therapies
The Problem He Wanted to Solve
Burns are among the most common injuries worldwide, often resulting in prolonged hospitalization and significant impairment. Current skin graft approaches rely on manual parameter tuning with broad, imprecise ranges — leading to suboptimal healing outcomes. Ryan wanted to use AI to optimize scaffold design and make 3D-bioprinted skin grafts more effective for burn and wound patients.
The Research
Working with YRI mentor Dr. Swetha MP, Ryan developed an AI-driven framework that integrates Multivariate Normal Distribution and polynomial regression with a 3D Generative Adversarial Network (GAN) to optimize scaffold designs for bioprinted skin grafts. The model was trained on synthetic datasets derived from real-world burn wound data from leprosy patients.
AI-Driven Scaffold Design Optimization of 3D-Bioprinted Skin Grafts for Improved Burn and Chronic Wound Healing
Conventional scaffold designs use broad, imprecise parameter ranges that limit healing effectiveness for burn and chronic wound patients
Multivariate Normal Distribution + polynomial regression logic combined with a 3D-GAN to generate optimized voxel-based scaffold designs
Synthetic datasets from real-world burn wound data (CO2Wounds-V2 Extended Chronic Wounds Dataset from Leprosy Patients)
R² score of 0.59, identified optimized parameters for porosity, pore size, thickness, and biomaterial selection
AI + Regenerative Medicine
What makes Ryan's work exceptional is the integration of AI with biomedical scaffold design — a cutting-edge intersection that most researchers don't tackle until graduate school. His framework enables patient-specific customization through data-driven scaffold generation, moving beyond one-size-fits-all approaches. The 3D-GAN generates complex structural variations in pore geometry and distribution, while a supervised regression model predicts healing efficiency — creating a complete pipeline from wound data to optimized graft design.
Parameters Optimized
GAN Architecture
R² Score
From Zero Experience
The Outcome
Accepted at IEEE International Conference on Industrial Electronics and Applications
IEEE ICIEA26, IEEE Industrial Electronics Society
2026
ICIEA26-000288
Accepted for Presentation & Publication
No research or programming experience, passionate about medicine and healthcare accessibility
Published at IEEE conference with AI-driven biomedical research, learned Python, built a complete ML pipeline from scratch
The Bigger Picture
Diabetes cases globally driving demand for advanced skin graft and wound healing technologies
Grade — Ryan tackled graduate-level biomedical AI research as a sophomore with zero prior experience
Accepted by the IEEE Industrial Electronics Society — one of IEEE's most established technical communities
Ready to Start Your Research Journey?
Join the YRI Fellowship and work with expert mentors to conduct original research, get published, and build a profile that stands out.
Apply Now