Hyeongsub Kim
AI Research Engineer
LG CNS, Seoul, South Korea
Contact: E14, LG Science Park, 71, Magokjungang 8-ro, Gangseo-gu, Seoul, South Korea
Email: hyeongsub.kim@lgcns.com
Greetings!
My name is Hyeongsub Kim, and I am currently working as an AI Research Scientist at LG CNS in South Korea. I am actively engaged in a project with Daehan Steel, a leading steel manufacturing company in South Korea, focused on deep learning-based steel scrap classification. My specific interest lies in sensor fusion algorithms, and I am driven to enhance the performance of AI models through this expertise.
My career has predominantly centered around the medical domain. I have undertaken projects in artificial intelligence-based medical image analysis, particularly in pathology images, as well as bio-signal analysis. These experiences have broadened my understanding of various forms of data analysis. Building on my past endeavors, I am enthusiastic about exploring artificial intelligence research based on multi-modal data. Furthermore, I aspire to apply my skills to fields beyond the medical domain, contributing to making a positive impact on the world.
News
[Oct 2023] Paper on early prediction of respiratory failure in the neonatal intensive care unit using electronic health records has been accepted to BMC pediatrics.
[Sep 2023] I joined LG CNS Vision AI Engineering Team as an AI research scientist.
[May 2023] My first bio-signal paper has been submitted to BMC pediatrics.
[Oct 2022] I joined VUNO Bio-signal Team as a research scientist.
[Sep 2022] Paper on medical image learning with limited and noisy data has been accepted to MICCAI 2022 workshop as a poster presentation.
[Nov 2021] Paper on compressed domain segmentation has been accepted to Scientific Report.
[Aug 2021] I joined VUNO Digital Oncology Team as a research scientist.
Publications
Early Prediction of Need for Invasive Mechanical Ventilation in the Neonatal Intensive Care Unit using Artificial Intelligence and Electronic Health Records: A Clinical Study
Young A Kim†, Hyeongsub Kim†, Jaewoo Choi, Kyungjae Cho, Dongjoon Yoo, Yeha Lee, Su Jeong Park, Mun Hui Jeaong, Seong Hee Jeong, Kyung Hee Park, Shin-Yun Byun, Taehwa Kim, Sung-Ho Ahn, Woo Hyun Cho, Narae Lee.
In BMC Pediatrics 23.1 (2023): 525.
Deep learning-based computed tomographic image super-resolution via wavelet embedding
Hyeongsub Kim, Haenghwa Lee, Donghoon Lee.
In Radiation Physics and Chemistry 205 (2023): 110718.
Abstraction in Pixel-wise Noisy Annotation Can Guide Attention to Improve Prostate Cancer Assessment
Hyeongsub Kim, Seo Taek Kong, Hongseok Lee, Kyungdoc kim, Kyu-Hwan Jung.
In MICCAI workshop, 2022 poster presentation.
Deep learning-based histopathological segmentation for whole slide images of colorectal cancer in a compressed domain
Hyeongsub Kim, Hongjoon Yoon, Nishant Thakur, Gyoyeon Hwang, Eun Jung Lee, Chulhong Kim, Yosep Chong.
In Scientific Report 11 (1), 22520.
GPU-accelerated 3D volumetric X-ray-induced acoustic computed tomography
Donghyun Lee, Eun-Yeong Park, Seongwook Choi, Hyeongsub Kim, Jung-joon Min, Changho Lee, Chulhong Kim.
In Biomedical optics express 11 (2), 752-761.
Education
Mar 2018 - Feb 2021
Pohang University of Science and Technology (POSTECH), South Korea
M.S. in i-bio
Mar 2009 - Feb 2018
Yonsei University, South Korea
B.S. in Radiological Science, Electrical Engineering