cv
This is a description of the page. You can modify it in '_pages/cv.md'. You can also change or remove the top pdf download button.
Basics
| Name | SangHyuk Kim |
| Label | Machine Psychology Fellow |
| sanghyuk.kim001@umb.edu | |
| Url | https://shkimmie-umb.github.io/aboutme/ |
| Summary | A Ph.D. student at the University of Massachusetts Boston, specializing in medical imaging and machine learning with over 4 years of experience in software engineering and interdisciplinary research. |
Work
-
2024.05 - 2024.12 Research Intern at NuSCI Research Group
Stanford Research Institute (SRI) International
Contributed to the SANSA project, focusing on reliable AI for healthcare applications.
- Enhanced melanoma detection by reducing misdiagnoses by 40.5% using uncertainty quantification.
- Developed a Python package achieving a 97% detection rate in melanoma detection, presented at IEEE ISBI 2025.
- Extended internship offer based on exceptional contributions.
-
2023.09 - Present Teaching/Research Assistant
Machine Psychology Lab, University of Massachusetts Boston
Conducting Ph.D. research in machine learning and medical imaging, focusing on uncertainty quantification and diagnostic reliability.
- Developed machine learning pipelines with 97.8% accuracy, reducing melanoma misdiagnoses by 40.5%.
- Teaching assistant for courses like C++, DBMS, Discrete Mathematics, and Theory of Computing.
- Authored a publication nominated for speaker presentation at IEEE ISBI 2025.
-
2017.02 - 2021.04 Software Engineer
TmaxOS
Fulfilled military service as a software developer with projects in kernel-level programming and smart factory solutions.
- Developed a graphics kernel to enable native execution of Windows applications on Ubuntu.
- Led smart factory planning for automated manufacturing
- Created a production-ready database system and web applications for enterprise decision-making.
Education
Publications
-
2025.04.15 Melanoma Detection with Uncertainty Quantification
2025 IEEE International Symposium on Biomedical Imaging (ISBI 2025)
Introduced a novel pipeline for melanoma detection, reducing misdiagnoses by 40.5% with uncertainty quantification.
-
2025.04.14 Boostlet.js: Medical Image Processing Plugins for the Web via JavaScript Injection
2025 IEEE International Symposium on Biomedical Imaging (ISBI 2025)
Co-author of a framework enabling medical image processing on web platforms using JavaScript injection.
-
2024.03.01 Web-based Melanoma Detection
arXiv preprint
Proposed a web-based solution integrating machine learning pipelines for melanoma detection.
-
2015.06.01 3D Face Modeling Using Face Image
Journal of International Society for Simulation Surgery
Explored techniques for constructing 3D face models from single facial images.
-
2012.06.01 Pattern Noise Removal Method Using Wavelet-FFT
Summer Annual Conference
Presented a pattern noise removal approach combining wavelet and FFT methodologies.
Skills
| Machine Learning | |
| Uncertainty Quantification | |
| Medical Imaging | |
| 2D/3D Detection Pipelines |
| Systems Programming | |
| Graphics Kernel Development | |
| Database Systems | |
| Smart Factory Solutions |
Languages
| Korean | |
| Native speaker |
| English | |
| Fluent |
References
| Daniel Haehn | |
| Assistant Professor at the University of Massachusetts Boston *Research Advisor |
| Brian Matejek | |
| Senior Computer Scientist at SRI International. *Internship Supervisor |
Interests
| Healthcare AI | |
| Medical Imaging | |
| Reliable Diagnostics | |
| Healthcare Applications |
| Systems Optimization | |
| Kernel Engineering | |
| Cross-Platform Compatibility | |
| Efficient Pipelines |