Hong JaeSeong (Jay Hong)
Portrait
Logo AI Researcher & Ph.D Candidate, Yonsei University
Research Interests

Federated Learning Multimodal AI Medical Informatics.

I aim to develop robust, trustworthy, and real-world applicable multimodal AI models by leveraging federated learning, which enables collaborative learning across multiple institutions while preserving data privacy.

Biography

I am currently a Ph.D candidate in the Department of Biomedical Systems Informatics at Yonsei University College of Medicine, supervised by Prof. Yu Rang Park. I also working as an AI Researcher at the Digital Healthcare Laboratory (DHLab) in Yonsei University College of Medicine and Severance Hospital in Seoul, Republic of Korea. Previously, I finished my B.B.A in Big Data Management and Statistics at Kookmin University in Seoul, Republic of Korea. From July to September in 2021, I worked as a Data Scientist at DACON, which is the largest data science competition platform in Korea.

Google Scholar GitHub LinkedIn ORCID
Education
  • Yonsei University
    Yonsei University
    Department of Biomedical Systems Informatics, College of Medicine
    Ph.D. Candidate
    September. 2022 - present
  • Kookmin University
    Kookmin University
    B.B.A in Big Data Management and Statistics
    March. 2018 - Febuary. 2022
Research Projects
News
2025
🤖 Two papers are under review
Oct 01
🎓 Completed Ph.D. coursework (GPA 4.11/4.5)
Aug 31
✍🏼 AAAI (Association for the Advancement of Artificial Intelligence) - Reviewer (2025)
Jul 25
🦴 Co-authored paper “Classification models for arthropathy grades of multiple joints based on hierarchical continual learning” published in La Radiologia Medica (IF: 4.8)
May 24
✍🏼 MICCAI (Medical Image Computing and Computer Assisted Intervention) -- Reviewer (2025)
Mar 28
🧠 Co–first author of “Retinal fundus imaging as biomarker for ADHD using machine learning for screening and visual attention stratification” published in npj Digital Medicine (IF: 12.4)
Mar 17
2024
🏥 Co-authored paper “PPFL: A personalized progressive federated learning method for leveraging different healthcare institution-specific features” published in iScience (IF: 4.6)
Dec 13
🧩 Co-authored paper “Comprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks–Based Deep Learning” published in JMIR Formative Research (IF: 2.0)
Feb 21
👁️ Co–first author of “Screening of moyamoya disease from retinal photographs: development and validation of deep learning algorithms” published in Stroke (IF: 7.9)
Jan 23
2023
👁️ Co–first author of “Development of deep ensembles for screening and severity of autism using retinal photographs” published in JAMA Network Open (IF: 13.9)
Dec 15
🧬 Co-authored paper “Three-dimensional label-free morphology of CD8+ T cells as a sepsis biomarker” published in Light: Science & Applications (IF: 20.6)
Dec 07
🧠 Co-authored paper “Development and Validation of a Joint Attention–Based Deep Learning System for Detection and Symptom Severity Assessment of Autism Spectrum Disorder” published in JAMA Network Open (IF: 13.9)
May 25
2022
🎓 Began Integrated Master’s and Ph.D. program
Sep 01
🎓 Earned B.B.A degree (Big Data Management and Statistics, GPA 3.86/4.5)
Apr 01
2021
💼 Joined DACON as a Data Scientist
Jul 01
Selected Publications (view all )
Retinal fundus imaging as biomarker for ADHD using machine learning for screening and visual attention stratification
Retinal fundus imaging as biomarker for ADHD using machine learning for screening and visual attention stratification

Hangnyoung Choi*, Hong JaeSeong*, Hyun Goo Kang*, Min-Hyeon Park, Sungji Ha, Junghan Lee, Sangchul Yoon, Daeseong Kim, Yu Rang Park, Keun-Ah Cheon (* equal contribution)

npj Digital Medicine 2025 Medical Dataset Deep+Machine Learning

This study demonstrated that machine learning analysis of retinal fundus photographs can serve as a noninvasive biomarker for screening attention-deficit/hyperactivity disorder (ADHD) and stratifying executive function deficits, achieving up to 96.9% AUROC and showing strong performance particularly in the visual attention domain.

[Paper]

Retinal fundus imaging as biomarker for ADHD using machine learning for screening and visual attention stratification

Hangnyoung Choi*, Hong JaeSeong*, Hyun Goo Kang*, Min-Hyeon Park, Sungji Ha, Junghan Lee, Sangchul Yoon, Daeseong Kim, Yu Rang Park, Keun-Ah Cheon (* equal contribution)

npj Digital Medicine 2025 Medical Dataset Deep+Machine Learning

This study demonstrated that machine learning analysis of retinal fundus photographs can serve as a noninvasive biomarker for screening attention-deficit/hyperactivity disorder (ADHD) and stratifying executive function deficits, achieving up to 96.9% AUROC and showing strong performance particularly in the visual attention domain.

[Paper]

PPFL: A personalized progressive federated learning method for leveraging different healthcare institution-specific features
PPFL: A personalized progressive federated learning method for leveraging different healthcare institution-specific features

Tae Hyun Kim*, Jae Yong Yu*, Won Seok Jang, Sun Cheol Heo, MinDong Sung, Hong JaeSeong, KyungSoo Chung, Yu Rang Park (* equal contribution)

iScience 2024 Medical Dataset Federated Learning

This study proposes a personalized progressive federated learning (PPFL) framework that incorporates client-specific feature information for heterogeneous healthcare data, achieving superior mortality prediction performance (accuracy = 0.941, AUROC = 0.948) compared with local and FedAvg models, and demonstrating the utility of personalized FL for vertically distributed clinical features.

[Paper]

PPFL: A personalized progressive federated learning method for leveraging different healthcare institution-specific features

Tae Hyun Kim*, Jae Yong Yu*, Won Seok Jang, Sun Cheol Heo, MinDong Sung, Hong JaeSeong, KyungSoo Chung, Yu Rang Park (* equal contribution)

iScience 2024 Medical Dataset Federated Learning

This study proposes a personalized progressive federated learning (PPFL) framework that incorporates client-specific feature information for heterogeneous healthcare data, achieving superior mortality prediction performance (accuracy = 0.941, AUROC = 0.948) compared with local and FedAvg models, and demonstrating the utility of personalized FL for vertically distributed clinical features.

[Paper]

Screening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms
Screening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms

Hong JaeSeong*, Sangchul Yoon*, Kyu Won Shim, Yu Rang Park (* equal contribution)

Stroke 2024 Medical Dataset Deep Learning

This study developed a deep learning model using retinal fundus photographs to screen for and stage Moyamoya disease (MMD), achieving high diagnostic performance (AUROC = 94.6%) and identifying retinal vascular regions as key features, suggesting that retinal imaging may serve as a noninvasive biomarker for MMD detection and progression assessment.

[Paper] [Patent]

Screening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms

Hong JaeSeong*, Sangchul Yoon*, Kyu Won Shim, Yu Rang Park (* equal contribution)

Stroke 2024 Medical Dataset Deep Learning

This study developed a deep learning model using retinal fundus photographs to screen for and stage Moyamoya disease (MMD), achieving high diagnostic performance (AUROC = 94.6%) and identifying retinal vascular regions as key features, suggesting that retinal imaging may serve as a noninvasive biomarker for MMD detection and progression assessment.

[Paper] [Patent]

All publications