BACK TO LIST
Questions · Methods · Findings · Current work · Collaboration
진단·다중모달 바이오마커
인지검사와 뇌·신체 신호를 결합해 인지저하를 더 일찍, 더 정밀하게 평가하는 방법을 연구합니다.
Which cognitive and biological signals can identify impairment earlier and remain reliable across people and settings?
인지검사·뇌영상·생체신호에서 얻은 정보를 어떻게 신뢰할 수 있는 조기 진단으로 연결할 수 있는가?
Development, adaptation, standardization and normative research include CERAD-K, SMCQ, CERAD total scores, dementia screening and functional assessment. Machine-learning work uses clinical reference diagnoses to evaluate performance.
MRI texture, brain structure and white-matter-hyperintensity segmentation are studied alongside longitudinal outcomes. KLOSCAD-NI and research datasets support model development and validation.
Neurophysiological studies examine quantitative electrical activity and task-related responses, including visual-cognition paradigms.
Accelerometry, gyroscopes and gait measurements connect movement patterns with frailty and cognition. Existing work includes the Korean Frailty and Aging Cohort Study (KFACS) and gait-analysis projects.
Acoustic-feature research investigates markers of late-life depression. Voice, cognition and clinical context are evaluated together rather than treating a single signal as a diagnosis.
Biomarker studies examine blood-based and genomic information, including amyloid-related assays and plasma proteomics, with clinical and cognitive characterization.
In an ADNI-based development and validation study, MRI texture measures predicted progression from mild cognitive impairment to Alzheimer dementia better than hippocampal volume for the evaluated comparisons. Generalization requires validation in the intended population.
Magnetic resonance imaging texture predicts progression to dementia due to Alzheimer disease earlier than hippocampal volume
Read the study / 논문 원문 ↗The Subjective Memory Complaints Questionnaire established a structured approach to assessing perceived memory difficulties. It is a research and assessment component, interpreted with clinical and objective cognitive information.
Development of the Subjective Memory Complaints Questionnaire
Read the study / 논문 원문 ↗A KLOSCAD study combined gait speed and temporal variability from a single wearable inertial sensor to distinguish cognitive disorders from normal cognition. Performance in an internal validation sample was comparable to MMSE-based classification, supporting further evaluation of gait as a digital marker.
Development of a Gait Feature-Based Model for Classifying Cognitive Disorders Using a Single Wearable Inertial Sensor
Read the study / 논문 원문 ↗An international collaborative study found that motoric cognitive risk syndrome, defined by slow gait and cognitive complaints, was associated with subsequent cognitive impairment and dementia. The findings support gait assessment as a practical component of identifying older adults at elevated risk.
Motoric cognitive risk syndrome: multicountry prevalence and dementia risk
Read the study / 논문 원문 ↗Development of brain-imaging software for earlier Alzheimer disease assessment.
Machine-learning methods using quantitative neurophysiological signals for geriatric cognitive disorders.
Plasma-proteomic biomarker discovery in mild cognitive impairment.
Research inquiries · 연구 문의
Ki Woong Kim, M.D., Ph.D. · RECODE Principal Investigator
Discuss independent validation cohorts, harmonized assessments, multimodal analysis or clinically interpretable models. Please describe the target setting, reference diagnosis, available modalities and validation plan.
협업 문의 / Contact PI ↗PI profile →
The CERAD-K (Consortium to Establish a Registry for Alzheimer's Disease - Korean) neuropsychological assessment battery
B. NEUROIMAGING
■ The Korean Longitudinal Study on Cognitive Aging and Dementia – Neuroimaging Substudy (KLOSCAD-NI):
Principal investigator, ongoing since 2014,
supported by the Korean Health Technology R&D Project, Ministry of Health and Welfare
and Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government (MSIT)
■ Development and validation of a deep learning algorithm for assisting the diagnosis of Alzheimer's disease using brain magnetic resonance images (K-DASH)[link]:
Principal investigator, ongoing since 2018,
supported by Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government (MSIT)
■ Development of brain MRI texture as an early diagnostic marker of Alzheimer’s disease:
Principal investigator, ongoing since 2014,
supported by Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government (MSIT)
(https://www.ncbi.nlm.nih.gov/pubmed/31228173)
■ Development of an algorithm for automatically segmenting white matter hyperintensities from brain MRI:
Principal investigator, completed from 2009 to 2011,
supported by the Independent Research Grant (IRG) from Eisai Global Pharmaceuticals
(https://www.ncbi.nlm.nih.gov/pubmed/24493377)
AD prediction model using deep convolutional neural networks
Texture analysis of the hippocampus from a T1-weighted MRI
Segmentation of WMH and Perventricular/Deep WMH from FLAIR image.
Design of an object-word matching task with four conditions.
Topological distribution of the difference between different object-word matching conditions in normal elderly and mild cognitive impairment patients.
D. GAIT
■ Korean Frailty and Aging Cohort Study (KFACS)[link]:
Co-investigator, ongoing since 2015,
supported by the Korea Health Technology R&D Project through the Korean Health Industry Development Institute (KHIDI), Ministry of Health and Welfare (https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiOrteView.kci?sereArticleSearchBean.artiId=ART002152213)
■ Development of gait database applied with the Walk-in-place Interactive System:
Principal investigator, ongoing since 2019,
supported by Korea Institute of Science and Technology
■ Development of gait analysis algorithm using the accelerometry in Korean elderly:
Principal investigator, completed from 2011 to 2012,
supported by Seoul National University Bundang Hospital
(https://www.ncbi.nlm.nih.gov/pubmed/27427965)
Walking posture of elderly
Attached sensors
Signal data of 3axial accelerator and gyrosensor
E. VOICE
■ Development of a gender-specific diagnostic algorithm for geriatric depression using acoustic features of voice:
Principal investigator, completed from 2015 to 2019,
supported by the Seoul National University Bundang Hospital
■ Development of a diagnostic algorithm for geriatric depression using acoustic features of voice:
Principal investigator, completed from 2014 to 2019,
supported by the Brain Fusion Program, Seoul National University
F. SEROLOGIC / GENETIC
■ Development of a gene-chip for diagnosing Alzheimer’s disease:
Collaborator, the Korean Brain Aging Study for the Elderly (KBASE) and the National Research Center for Dementia,
supported by the National Research Foundation of Korea, Ministry of Science and ICT.
