Measuring Neuropsychological Change and Brain Structural Alteration for Cognitive Decline Prediction
Poster No:
136
Submission Type:
Abstract Submission
Authors:
Ling Yue1, Qiufeng Chen2, Bo Hong3, Tianli Tao4, Shifu Xiao3, Han Zhang5
Institutions:
1Shanghai Mental Health Center, Shanghai Jiaotong University, School of Medicine, Shanghai, Shanghai, 2College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou, Fujian, 3Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine, Shanghai, Shanghai, 4School of Biomedical Engineering, ShanghaiTech University, Shanghai, China, 5School of Biomedical Engineering, ShanghaiTech University, Shanghai, Shanghai
First Author:
Ling Yue
Shanghai Mental Health Center, Shanghai Jiaotong University, School of Medicine
Shanghai, Shanghai
Shanghai Mental Health Center, Shanghai Jiaotong University, School of Medicine
Shanghai, Shanghai
Co-Author(s):
Qiufeng Chen
College of Computer and Information Sciences, Fujian Agriculture and Forestry University
Fuzhou, Fujian
College of Computer and Information Sciences, Fujian Agriculture and Forestry University
Fuzhou, Fujian
Bo Hong
Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine
Shanghai, Shanghai
Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine
Shanghai, Shanghai
Shifu Xiao
Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine
Shanghai, Shanghai
Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine
Shanghai, Shanghai
Introduction:
Dementia is the common healthcare issue worldwide due to the prevalence of aging population expansion. However, very limited studies were initiated among cognitively normal (NC) elderly considered at risk of developing mild cognitive impairment (MCI) (Albert et al., 2018). Reliable screening for the NC who are more likely to convert to MCI many years later becomes critical. In this study, we conducted a seven-year baseline follow-up investigation of elderly subjects recruited from local communities to detect subtle alterations in clinical and neuroimaging markers as elderly transition from NC to MCI. We propose utilizing a machine learning algorithm to predict changes in brain MRI scans, based on baseline MRI features, clinical information, and comprehensive neuropsychological assessments (Roe et al., 2018; Xie et al., 2020).
Methods:
Participants were recruited through the Chinese Longitudinal Aging Study, initiated in 2011 (CLAS) (Yue et al., 2018). All subjects had signed written informed consent prior to enrollment and the study protocol was scrutinized and approved by the Ethical Committee. A total of 222 (aged 70.4±7.66, 105 males) community-dwelling adults who were from Shanghai and completed both the baseline T1-weighted MRI scan and took a battery of neuropsychological tests were recruited. The T1-weighted 3D structural MRI was acquired from a 3.0 T MRI scanner (Siemens, MAGNETOM VERIO) with 1*1*1 mm3. We used the FreeSurfer for the minimal structural preprocessing.
Figure. 1 depicted how we train and test the two-stage progression prediction model for the elderly subjects who will or will not progress to MCI 7 years later. In the training stage, we subtracted the features extracted from the follow-up data from those from the baseline to obtain within-subject longitudinal changes for each individual in the training set of the main dataset. By using the 7-year changes as targets, we trained a multi-task regression model with sparsity and global consistent constraints. This regression model established the mapping relationship between baseline features and 7-year changed features. In the predicting stage, we leveraged the pre-trained regression model to predict the longitudinal changes for the new coming subject at the baseline phase. Based on the predicted changed features, we employed the support vector machine (SVM) to classify this subject into progressive or non-progressive groups.
Figure. 1 depicted how we train and test the two-stage progression prediction model for the elderly subjects who will or will not progress to MCI 7 years later. In the training stage, we subtracted the features extracted from the follow-up data from those from the baseline to obtain within-subject longitudinal changes for each individual in the training set of the main dataset. By using the 7-year changes as targets, we trained a multi-task regression model with sparsity and global consistent constraints. This regression model established the mapping relationship between baseline features and 7-year changed features. In the predicting stage, we leveraged the pre-trained regression model to predict the longitudinal changes for the new coming subject at the baseline phase. Based on the predicted changed features, we employed the support vector machine (SVM) to classify this subject into progressive or non-progressive groups.
·The schematic diagram of two-stage MCI prediction
Results:
We selected the most discriminative features from clinical information (MoCA, digital span forward and backward, WAIS block design, visual recognition, and long-term auditory verbal learning test (AVLT)) and structural MRI (the middle frontal, superior occipital, superior parietal lobe, and ventral part of the temporal and occipital lobes). With predicted 7-year feature changes as inputs, we achieved promising conversion prediction results (accuracy=73.8%, sensitivity=75%, and specificity=73.2%). Furthermore, our analysis revealed significant correlations between cortical thickness changes and declines in specific cognitive domains among the progressive group. For instance, there was a strong correlation between fusiform gyrus thickness and WAIS block design performance (r=0.728, p=0.0252), as well as between right collateral-lingual sulcus thickness and visual functional recognition (r=0.677, p=0.0337). Interestingly, we also observed a negative correlation between right lateral sulcus thickness and AVLT performance (r=-0.729, p=0.0107).
Conclusions:
Our findings underscore the pivotal role of neuropsychological assessments in early prediction of cognitive decline, while further highlighting the significance of specific cortical thickness alterations and their associations with cognitive decline. These correlations contribute to advancing personalized risk prediction for cognitive impairment, even among cognitively normal elderly individuals within the community.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Novel Imaging Acquisition Methods:
Anatomical MRI 2
Keywords:
Machine Learning
MRI
STRUCTURAL MRI
Other - Alzheimer's
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Roe, C. M. (2018). Incident cognitive impairment: longitudinal changes in molecular, structural and cognitive biomarkers. Brain, 141(11), 3233-3248.
Xie, L. (2020). Longitudinal atrophy in early Braak regions in preclinical Alzheimer's disease. Human Brain Mapping, 41(16), 4704-4717.
Yue, L. (2018). Asymmetry of hippocampus and amygdala defect in subjective cognitive decline among the community dwelling Chinese. Frontiers in psychiatry, 9, 226.
Acknowledgement
This work is supported by the China Ministry of Science and Technology (STI2030-Major Projects-2022ZD0213100), Shanghai public health projects (GWVI-11.2-XD24), China Scholarship Council, National Natural Science Foundation of China (Grant No.62301160, 62131015, U23A20295), Nature Science Foundation of Fujian Province (Grant No.2022J01607), Shanghai Pilot Program for Basic Research - Chinese Academy of Science, Shanghai Branch (No. JCYJ-SHFY-2022-014) and Shenzhen Science and Technology Program (No. KCXFZ20211020163408012).