Research Intern in Queen's University - Vector institute affiliated Lab (Advisor: Anna Panchenko)
📍 Kingston, Canada
📧 sunghwan.moon.ai@gmail.com
🛄 Linkedin

Kyung Hee University, Seoul, South Korea (GPA: 3.95 / 4.0)
M.S. in Software Convergence
Advisor: Prof. Won Hee Lee
Mar. 2022 – Feb. 2025

University of Toronto, Ontario, Canada
Visiting Graduate Student, Dept. of Mechanical and Industrial Engineering
Industry-academia joint project with LGE Toronto AI Lab (Adviosr: Dr. Thi Ha Kyaw)
Jan. 2024 – Jun. 2024

Kyung Hee University, Seoul, South Korea (GPA: 3.65 / 4.0)
B.S. in Software Convergence
B.S. in International Business
Mar. 2015 – Feb. 2022

Building AI systems for early diagnosis and personalized treatment.
Bridging basic research and clinical application using generative models and reliability metrics.

• 🧠 Generative Models for Healthcare

  • Medical report generation for disease progression and personalized treatment
  • Leveraging longitudinal imaging and other modality (e.g., EHR)

• 🏥 Model Reliability for Clinical Application

  • Developing evaluation metrics for clinical fidelity
  • Ensuring trustworthiness and interpretability in real-world settings

• Developing a GNN-based representation learning framework for residue interaction networks to identify oncogenic cancer drivers through classification and clustering as downstream tasks
• Conducting comprehensive biological network analysis utilizing graph theory to find cancer hotspots and quantify the propagation of mutation effects within chromatin structures

• Developed a diffusion-based generative model to simulate future brain MRIs
• Investigating potential for early Alzheimer’s diagnosis

• Proposed the multi-modal based brain age prediction model using structural MRI and diffusion MRI
• Demonstrated the multi-modal model outperformed single-modality model in terms of accuracy, generalizability, reproducibility, and consistency
• Identified the key associations between BrainPAD and clinical assessment score

• Proposed an anatomically informed 3D diffusion model for brain MRI
• Designed a framework for evaluating morphological preservation in synthetic images

• Sunghwan Moon, Junhyuk Lee, Won Hee Lee.
  “Predicting brain age with global-local attention network from multimodal neuroimaging data: accuracy, generalizability, and behavioral associations”, Computers in Biology and Medicine, 2025.
  (IF: 7.0, Top 2.3% in Mathematical & Computational Biology)

• Sunghwan Moon, Tae Seong Kim, Jihye Ryu, Won Hee Lee.
  “Federated Learning for Sleep Stage Classification on Edge Devices via a Model-Agnostic Meta-Learning-Based Pre-Trained Model”, IEEE ICCE-Berlin, 2023. (Oral Presentation)

Python

PyTorch

MONAI

Git

LangChain

W&B