Postdoctoral Research Associate - Generative AI for Fracture Modeling and Digital Twins

Summary

The Medical Simulation and Computer-Assisted Intervention Laboratory at Old Dominion University invites applications for a full-time Postdoctoral Research Associate to develop next-generation generative AI models for anatomically accurate bone fracture simulation as part of a multidisciplinary translational research program involving Old Dominion University, the University of Virginia, Virginia Tech, and Sonogen Medical. The project is supported through a Virginia Catalyst collaborative award focused on AI-enabled orthopedic technologies and digital twin development.

Project Overview

The successful candidate will lead the development of fracture-aware generative anatomical models capable of producing high-fidelity 3D representations of healthy and fractured skeletal anatomy. These models will support AI algorithm development, digital twins, biomechanical simulation, medical device validation, and future clinical translation.

The position offers an opportunity to work at the intersection of:

• Generative AI
• Computer vision
• Medical image analysis
• Biomechanical simulation
• Digital twins
• Orthopedic engineering
• Medical device innovation

Research Responsibilities

The postdoctoral fellow will execute a one-year research program consisting of four major phases.

Months 1–3: Foundation and Baseline Models

• Define target anatomical representations
• Assemble and curate CT/MRI/X-ray datasets
• Build computational infrastructure
• Develop baseline diffusion and/or GAN-based anatomical generation pipelines

Months 4–6: Fracture-Aware Generation

• Introduce fracture-conditioned generation
• Improve anatomical realism
• Incorporate biomechanical and clinical constraints
• Develop quantitative evaluation metrics

Months 7–9: High-Resolution Modeling and Validation

• Produce publication-quality anatomical models
• Integrate finite element and biomechanical plausibility
• Validate models for engineering and clinical applications
• Prepare conference and journal publications

Months 10–12: Generalization and Software Release

• Improve robustness and diversity
• Package reusable software
• Finalize curated datasets
• Prepare future NIH, NSF, and DoD proposals
• Deliver open, reusable research outputs

These activities align with the ODU milestones in the Virginia Catalyst project, including foundation generative models, fracture-aware generation, high-resolution validation, and refinement/generalization.

Opportunities

The successful candidate will have opportunities to

• Publish in leading AI and biomedical engineering journals
• Present research at premier international conferences
• Collaborate with clinicians, orthopedic surgeons, and AI researchers across multiple institutions
• Participate in development of next-generation AI-enabled medical devices

Appointment

• Full-time, anticipated to last 12-month renewable subject to funding availability
• Start date: Immediately available (preferred)

Job Responsibilities:

Literature review, including finding relevant open-source software
Designing architecture of image analysis system
Implementing and training image analysis system
Validation of system.
Publishing on innovations.

Education and/or training:

• Ph.D. in Biomedical Engineering, Computer Science, Electrical Engineering, Robotics, Medical Imaging, Applied Mathematics, or related discipline

Level and type ofexperience

Required

• Strong programming skills in Python
• Experience with PyTorch or TensorFlow
• Experience with deep learning and generative AI

Specific knowledge, skills, and abilities:

Preferred

Experience in one or more of:

• Diffusion models
• GANs
• 3D generative models
• Medical image segmentation
• Computer vision
• Medical image registration
• Mesh processing
• Point clouds
• Neural implicit representations
• Finite element modeling
• Biomechanics
• Digital twins
• Scientific computing (CUDA desirable)

Candidates with publications at MICCAI, CVPR, ICCV, ECCV, NeurIPS, ICLR, AAAI, RSNA, or similar venues are especially encouraged to apply.