류창우 연구실은 신경진단방사선학과 뇌기능영상을 기반으로 알츠하이머병과 경도인지장애의 구조적·기능적 뇌 변화를 정량 MRI, 수초영상, 전도도 맵핑, 미세혈관 영상기법으로 분석하고, 이를 혈액 바이오마커 및 인지기능 지표와 연계하여 조기 진단과 질환 분류를 위한 영상 바이오마커를 개발하는 연구를 수행하고 있다.
Mapping High‐Frequency Conductivity Using MRI in the Brains of Alzheimer’s Disease Patients
Geon‐Ho Jahng, Soonchan Park, Mun Bae Lee, Hak Young Rhee, Chang‐Woo Ryu, Oh In Kwon
IF 11.1
Alzheimer s & Dementia
Abstract Background Previous studies reported increased concentrations of metallic ions and imbalanced Na+ and K+ ions in patients with Alzheimer’s disease (AD) and the increased mobility of protons by microstructural disruptions in AD. The purposes of this study were: 1)to apply a high‐frequency conductivity (HFC) mapping technique using a clinical 3T MRI system and 2)to compare HFC values among participant groups. Method This prospective study included 74 participants (23 AD patients, 27 amnestic mild cognitive impairment (MCI) patients, and 24 cognitively normal (CN) elderly people) to explore the clinical application of HFC mapping in the brain. For the brain MREPT images, a multi‐echo turbo spin‐echo pulse sequence was used. A homemade software was used to map the HFC at the Larmor frequency of 128 MHz at 3T. We performed statistical analyses to 1) compare HFC maps between the three participant groups and 2) to evaluate the association of HFC maps with Mini‐Mental State Examination (MMSE) scores. Result The HFC value was higher in the AD group than in the CN and MCI groups. MMSE scores were negatively associated with HFC values, but age was positively associated with HFC values. The HFC value in the insula has a high area under the ROC curve (AUC) value to differentiate AD patients from the CN participants (Sensitivity (SE) = 82, Specificity (SP) = 97, AUC = 0.902, p < 0.0001), better than GMV in hippocampus (SE = 79, SP = 83, AUC = 0.880, p < 0.0001). Conclusion HFC values were significantly increased in the AD group compared to the CN group and increased with age and disease severity. HFC values of the insula along with the GMV of the hippocampus can be used as an imaging biomarker to improve the differentiation of AD from CN.
Myelin‐weighted MR Imaging in Patients with Alzheimer’s Disease
Seung‐Hyun Lim, Ji Yoon Lee, Sumin Jung, Hak Young Rhee, Soonchan Park, Ah Rang Cho, Chang‐Woo Ryu, Geon‐Ho Jahng
IF 11.1
Alzheimer s & Dementia
Abstract Background Developing an imaging technique to map myelin changes in the brain within a short scan time is needed due to the inability of Alzheimer’s disease (AD) patients to remain motionless during MRI scanning. Purpose To investigate myelin loss in both AD and mild cognitive impairment (MCI) patients with a new myelin water mapping technique within reasonable scan time and evaluate the clinical relevance of the apparent myelin water fraction (MWF) values by assessing the relationship between decreases in myelin water and the degree of memory decline or aging. Method In this prospective study from March 2019 to October 2021, the participants were allocated as follows: 29 individuals were assigned to the cognitively normal (CN) elderly group (mean ± SD, 73 ± 5 years, 16 females), 32 participants were assigned to the MCI group (74 ± 5 years, 21 females), and 31 patients were assigned to the AD group (74 ± 6 years, 26 females). A 3D visualization of the short transverse relaxation time component (ViSTa)‐gradient and spin‐echo (GraSE) sequence was developed to map apparent MWF. Then, the MWF values were compared between the three participant groups, and was evaluated the relationship with the degree of memory loss. Result A total of 92 participants were evaluated. The AD group showed a reduced apparent MWF compared to the CN and MCI groups. Moreover, the MWF was decreased with age and positively correlated with Mini‐Mental State Examination scores. The largest AUC value (0.799) was in the corpus callosum and used to classify the CN and AD groups using the apparent MWF. Conclusion The ViSTa‐GraSE sequence can be a useful tool to map the MWF in a reasonable scan time. Combining the MWF in the corpus callosum with the detection of atrophy in the hippocampus can be valuable for group classification. Funding The research was supported by the National Research Foundation of Korea (NRF) grant funded by Ministry of Science and ICT (No. 2020R1A2C1004749, GHJ), Republic of Korea.
Microvascular Morphology MR Imaging Using Relaxation Rate Change with Gadolinium‐Based Contrast Agent in Patients with Alzheimer’s Disease
Xiao‐Yi Guo, HyeokJung Kwon, Hak Young Rhee, Soonchan Park, Ah Rang Cho, Chang‐Woo Ryu, Geon‐Ho Jahng
IF 11.1
Alzheimer s & Dementia
Abstract Background Conventional MRI techniques cannot demonstrate microvascular alterations in mild and moderate Alzheimer’s disease (AD). Thus, the diagnosis of microvascular pathology commonly relies on postmortem. The purpose of this study was to evaluate alterations of microvascular structures in patients with AD using a 3T clinical MRI system with a commercially available contrast agent. Method Eleven patients with AD and 11 non‐AD controls were included in this cross‐sectional prospective study. R2 and R2* relaxation rate changes (∆R2 and ∆R2*) before and after a Gd‐based contrast agent injection were calculated from images obtained with a multi‐echo turbo spin‐echo sequence and multi‐echo gradient‐echo sequence to obtain microvascular index maps of blood volume fraction (BVf), mean vessel diameter (mVD), vessel size index (VSI), mean vessel density (Q), and microvessel‐weighted imaging (MvWI). Two‐sample t‐test was used to compare those values between the two groups. Correlation analysis was performed to evaluate the relationship between those values and the mini‐mental state examination (MMSE) score. Result Compared with the non‐AD group, BVf was significantly increased in the AD group at the parahippocampal gyrus (p = 0.033), gray matter area (p = 0.008), and white matter area (p = 0.010). MvWI was also significantly increased in the AD group at the hippocampus (p = 0.010), parahippocampal gyrus (p = 0.010), precuneus (p = 0.045), and globus pallidus (p = 0.008). Both mVD and VSI were only significantly decreased in the AD group at the white matter hyperintensity (WMHI) area (p = 0.045 with mVD; p = 0.033 with VSI). BVf and MvWI were significantly positively correlated with age at some areas. VSI was significantly positively correlated with MMSE at caudate (rho = 0.475/ p = 0.030) and corpus callosum (rho = 0.435/ p = 0.049). Conclusion Microvascular index was a useful non‐invasive method to evaluate microvascular morphology alteration. The microvascular morphology of AD was manifested as vessel constriction and increasing blood volume fraction. It was associated with MMSE score.
본 과제는 경증외상성뇌손상의 미세한 뇌 손상을 영상으로 정확히 구분하기 어려운 문제를 해결하기 위해, 뇌네트워크(커넥톰)를 기반으로 한 머신러닝 진단모델을 개발하는 연구임.
연구 목표는 뇌네트워크(커넥톰)기반 경증 외상성뇌손상 머신러닝 진단모델 개발이며, 3T MRI 휴지기/작업기반 기능성 MRI와 확산텐서 MRI의 최적화 및 원클릭 프로세싱 파이프라인, 외상성뇌손상환자코호트의 커넥톰 데이터베이스 구축, 선택적 Feature Guide DNN과 Semi-supervised Learning으로 진단·예측 모델을 학습 및 검증함. 기대 효과는 객관적 비침습 진단법 제공, AI 기반 뇌질환 자율진단도구 상용화, 장기 자연경과 연구 기반 확보임.