Korea University | BREIN Lab

Brain Reverse Engineering by Intelligent Neuroimaging

BREIN Lab은 뇌영상, EHR, 인지척도, 분자 바이오마커를 통합하는 Medical AI와 multimodal biomarkers를 통해 뇌질환의 진행과 예후를 예측합니다.

Core Focus
Medical AI
Methods
MRI, PET, EHR, CDR
Location
Hana Science Hall B

Research Pillars

Medical AI for multimodal brain health data.

Medical AI, multimodal biomarkers, disease progression modeling, interpretable decision support를 중심으로 연구를 확장합니다.

Medical AI

임상 데이터, 뇌영상, EHR을 활용해 diagnosis prediction, cognitive staging, hospital-data foundation model을 연구합니다.

Multimodal Biomarkers

MRI, PET, blood biomarkers, CDR/MMSE, omics, digital behavior를 통합해 Alzheimer’s disease 및 관련 질환을 해석합니다.

Disease Progression & Prognosis

Tau/ND mismatch, resilience, vulnerability, future cognitive decline을 통합해 patient-tailored trajectory를 모델링합니다.

Interpretable AI

Feature attribution, modality contribution, structured risk explanation을 통해 모델 결과를 임상적으로 해석 가능한 근거로 변환합니다.

Featured Publications

Clinical questions and computational methods, together.

All publications
A peak-oriented diffusion model for high-fidelity electrocardiogram reconstruction from photoplethysmogram: development and usability study
2026 Medical AI

A peak-oriented diffusion model for high-fidelity electrocardiogram reconstruction from photoplethysmogram: development and usability study

광용적맥파(PPG) 데이터로부터 고충실도 심전도(ECG)를 재구성하는 peak-oriented diffusion model을 제안합니다.

BioData Mining 19 (1), 48, 2026

Distinct spatial patterns of perivascular spaces enlargement for multiple and Co-existing pathologies of cognitive impairment
2026 Neuroimaging Biomarkers

Distinct spatial patterns of perivascular spaces enlargement for multiple and Co-existing pathologies of cognitive impairment

인지장애와 관련된 여러 병리 및 공존 병리에서 perivascular spaces enlargement의 공간 패턴을 분석합니다.

The Journal of Prevention of Alzheimer's Disease 13 (8), 100631, 2026

Development of head-to-head and longitudinal CycleGAN algorithm for MRI harmonization: validation in follow-up MRI evaluation in patients with brain metastasis
2026 Medical AI

Development of head-to-head and longitudinal CycleGAN algorithm for MRI harmonization: validation in follow-up MRI evaluation in patients with brain metastasis

추적 MRI 평가를 위한 head-to-head 및 longitudinal CycleGAN 기반 MRI harmonization 방법을 개발하고 검증합니다.

Scientific Reports 16 (1), 13163, 2026

High-order attention mechanism of MR imaging features predicts differential diagnosis for dementia
2025 Medical AI

High-order attention mechanism of MR imaging features predicts differential diagnosis for dementia

MR 영상 특징의 high-order attention mechanism을 활용해 치매 감별진단을 예측합니다.

Alzheimer's & Dementia 21, e106312, 2025

Validation of criteria for frontotemporal dementia with right anterior temporal lobe predominance
2025 Dementia AI / 치매 AI

Validation of criteria for frontotemporal dementia with right anterior temporal lobe predominance

우측 전측두엽 우세 양상을 보이는 전두측두치매 진단 기준을 검증합니다.

Alzheimer's & Dementia 21 (7), e70443, 2025

Temporal dynamics and biological variability of Alzheimer biomarkers
2025 Neuroimaging Biomarkers

Temporal dynamics and biological variability of Alzheimer biomarkers

알츠하이머병 바이오마커의 시간적 변화와 생물학적 변동성을 분석합니다.

JAMA Neurology 82 (4), 384-396, 2025

Professor

Joon-Kyung Seong

Quantitative neuroimaging, brain network analysis, shape analysis, medical AI를 기반으로 뇌질환과 치매 연구를 수행합니다.

Professor Profile
Joon-Kyung Seong

Join Us

Medical AI, neuroimaging, dementia research에 관심 있는 학생을 기다립니다.

Medical AI, multimodal biomarker analysis, deep learning, statistics, neuroscience, clinical data에 관심이 있다면 연락해 주세요.

Contact BREIN