Research in progress

The questions
driving the next chapter.

Funded research programs and ongoing investigations. Active work is presented as a research direction, not as a completed or clinically validated result.

Funded research

Principal Investigator · NRF Basic Science Research Program

Active

National Research Foundation of Korea (NRF)

Principal Investigator
Sep 2026 — Aug 2029

RS-2026-25580544

Explainable AI for long-term brain age trajectories

Research on multimodal, explainable models that connect diagnostic estimation with long-term prognostic trajectories. This is an active research objective, not a claim of completed clinical validation.

Official project title
Development of an Explainable AI Model for Long-term Brain Age Trajectory Prediction Based on Dual-stream Diagnostic-Prognostic Estimation Using Multimodal Data

Basic Science Research Program · Ministry of Education, Republic of Korea

Completed funding period

National Research Foundation of Korea (NRF)

Principal Investigator
Sep 2022 — Aug 2025

2022R1A6A3A01086784

Multimodal learning for dementia progression

A research program on multimodal and multi-domain learning methods for dementia progression prediction.

Official project title
Multi-Modal and Multi-Domain based Machine Learning Algorithms for Dementia Progression Prediction

Basic Science Research Program · Ministry of Education, Republic of Korea

Ongoing investigations

Work in progress.

Titles carried over from the existing public website. No publication or code-release status is implied.

Ongoing · Graph learning

Causal Diffusion on Multiplex Graphs for Robust Out-of-Distribution Generalization

Investigating causal diffusion on multiplex graphs for generalization beyond the training distribution.

Sunghong Park, Hyun Goo Woo*, Hyunjung Shin*

Ongoing · Representation learning

Generalized Discriminant Analysis for Explainable Concept Learning within Structurally-Aware Geometric Space

Studying structurally aware geometric representations for explainable concept learning.

Sunghong Park, Hyun Goo Woo*, Hyunjung Shin*

Ongoing · Optimization

Balancing the Unbalanceable: Harnessing Productive Tension via Dynamic Pareto-Guided Optimization

Exploring dynamic Pareto-guided optimization for competing learning objectives.

Sunghong Park†, Junghyun Kim†, JeongHeun Yeon, Hyun Goo Woo*, Hyunjung Shin*

Ongoing · Computational pathology

Causally-Infused Complexity-Guided Learning on Computational Pathology

Studying causal and complexity-guided approaches to computational pathology.

Sunghong Park, Hyun Goo Woo*

Ongoing · Graph learning

Distributionally Robust Manifold Alignment Framework for Geometric Information Fusion for Multiplex Graphs under Extreme Label Scarcity

Investigating robust manifold alignment and graph information fusion when labels are extremely limited.

Sunghong Park†, Junghyun Kim†, JeongHeun Yeon, Hyun Goo Woo*, Hyunjung Shin*

Ongoing · Biomedical AI

Artificial Intelligence-based Foundation-to-Precision Framework for Robust Survival Prediction and Therapeutic Hypothesis Generation in Hepatocellular Carcinoma

Exploring survival prediction and therapeutic hypothesis generation in hepatocellular carcinoma.

Sunghong Park†, Ji-Hye Choi†, Hyun Goo Woo*

Ongoing · Computational pathology

SPOT-MIL: Spatially-Guided Prototype-Based Optimal Transport Alignment for Trustworthy Multiple Instance Learning in Whole Slide Image Analysis

Investigating spatial guidance and optimal transport in multiple instance learning for whole slide images.

Sunghong Park, Julien Calderaro, Hyun Goo Woo*

Ongoing · Domain adaptation

Robust Multiplex Graph Domain Adaptation via Pseudo-Label-Free Manifold Alignment under Extreme Label Scarcity

Studying pseudo-label-free manifold alignment for multiplex graph domain adaptation.

Sunghong Park, JeongHeun Yeon, Hyun Goo Woo*, Hyunjung Shin*

† Co-first author · * Corresponding author. Please contact me for discussions about ongoing research.

Start a conversation

Good research begins
with a connection.

For research discussions and collaboration in graph learning, domain adaptation, and biomedical AI.

Life Science Research Institute · KAIST