Sungduk Cho

ML Engineer @ VUNO
newhigen@gmail.com
Machine learning engineer researching and building medical AI products for 4 years. Experienced in full-cycle model development for chest X-ray and lung CT, product deployment and optimization, and designing reusable data and evaluation pipelines.

Experience

ML Engineer, VUNO Mar 2022 – Present (4 yrs 7 mos)
PXI (Chest) Team Sep 2024 – Present
Diagnostic AI products for chest X-ray (VUNO Med®-Chest X-ray™)
  • Model R&D, ONNX/OpenVINO serialization, and deployment optimization
  • Built reusable DB-based training and evaluation pipelines with standardized metrics
  • Developed internal developer tools and established code review practices
BLU3 (Lung) Team Nov 2022 – Aug 2024
Diagnostic AI products for lung CT (VUNO Med®-LungCT AI™)
  • Built shared 3D lung imaging pipelines to eliminate duplicate code across models
  • Quantitatively analyzed label errors in external evaluation datasets to guide relabeling by overseas hospitals
  • Introduced team MongoDB for data integrity and performed on-site hospital validation
Digital Oncology Team Mar 2022 – Oct 2022
Digital pathology research and development
  • 1st place in TIGER Challenge survival analysis track (inference pipeline & TIL analysis)[J1]
  • Built containerized pipeline connecting multi-stage models and released open-source
  • Designed modular reusable library for whole slide image (WSI) processing
Research Assistant, Korea University High Performance Visual Computing Lab Jun 2019 – Feb 2022
Interactive Segmentation Research
  • Deep learning models and interactive tools to reduce physician annotation effort[C1]
Organized International Medical Data Challenges (2 Editions)
  • Web tool design, implementation, and qualitative evaluation for 3 hospitals[J2]
  • Built and operated a digital platform for sharing large-scale pathology datasets[J3]

Skills

ML & Serving Python, PyTorch, ONNX, OpenVINO
Data & MLOps MongoDB, Airflow, MLflow, W&B
DevOps & CI/CD Docker, Git, GitHub Actions
AI Tools & Agents Claude Code, Codex, Gemini

Education

M.E. in Computer Science and Engineering, Korea University 2020 – 2022
B.S. in Computer Science & Engineering, UNIST 2015 – 2020

Awards

2022
1st Place, International Pathology Image Challenge (TIGER Challenge)
Radboud University
2021
1st Place, Korea University Startup Competition
Seoul Metropolitan Government, Seongbuk-gu, SH Corporation, Korea University
2020
1st Place, Dental Panoramic X-ray AI Challenge
Healthhub Datathon Track B, Seoul National University Dental Hospital

Publications

Journals
[J1] 2025
Analysis of computational tumor-infiltrating lymphocytes in breast cancer from the results of the TIGER challenge
TIGER Challenge Consortium
Nature Communications
[J2] 2023
PAIP 2020: Microsatellite instability prediction in colorectal cancer
Kyungmo Kim*, Kyoungbun Lee*, Sungduk Cho*, Dong Un Kang, Seongkeun Park, et al.
Medical Image Analysis
[J3] 2021
Development and operation of a digital platform for sharing pathology image data
Yunsook Kang, Yoo Jung Kim, Seongkeun Park, Gun Ro, Choyeon Hong, Hyungjoon Jang, Sungduk Cho, et al.
BMC Medical Informatics and Decision Making
Conferences
[C1] 2021
DeepScribble: Interactive Pathology Image Segmentation Using Deep Neural Networks with Scribbles
Sungduk Cho, Hyungjoon Jang, Jing Wei Tan, Won-Ki Jeong
IEEE International Symposium on Biomedical Imaging (ISBI)
Preprint
[P1] 2024
Joint-Embedding Predictive Architecture for Self-Supervised Learning of Mask Classification Architecture
Dong-Hee Kim, Sungduk Cho, Hyeonwoo Cho, Chanmin Park, Jinyoung Kim, Won Hwa Kim
arXiv
* Equal contribution

Languages

English OPIc Advanced Low Jul 2025