Generalized Association Plots and Structural Equation Modeling of ADNI Latent Structures
Poster No:
1605
Submission Type:
Abstract Submission
Authors:
Yi-Ju Lee1
Institutions:
1Academia Sinica, Taipei City, Taipei City
First Author:
Introduction:
Exploratory data analysis integrates statistics and computing to reveal the latent structure of multivariate data sets. While precision medicine and big medical data have advanced biological and methodological research, effective therapies for complex human diseases remain elusive, and adequate models for treatment validation are still lacking. For neurodegenerative diseases, current models based on physiological and neurobiological hypotheses offer limited insight into pathomechanisms.
In this research, we employ Generalized Association Plots (GAP) and Structural Equation Modeling (SEM) to explore Alzheimer's Disease. GAP provides advanced visualization of the ADNI dataset's multiple aspects, revealing non-linear relationships and complex patterns across disease progression that traditional approaches might miss. SEM complements this by providing a confirmatory framework to test theoretical disease models. This integrated approach aims to identify novel biomarker combinations for early diagnosis and support personalized treatment strategies.
In this research, we employ Generalized Association Plots (GAP) and Structural Equation Modeling (SEM) to explore Alzheimer's Disease. GAP provides advanced visualization of the ADNI dataset's multiple aspects, revealing non-linear relationships and complex patterns across disease progression that traditional approaches might miss. SEM complements this by providing a confirmatory framework to test theoretical disease models. This integrated approach aims to identify novel biomarker combinations for early diagnosis and support personalized treatment strategies.
Methods:
This study employs a two-stage analytical approach to investigate Alzheimer's disease mechanisms using the ADNI dataset. We analyzed data from 400 ADNI-2 participants: 100 each with early-onset mild cognitive impairment, late-onset mild cognitive impairment, and Alzheimer's disease, plus 100 healthy controls. The analysis included demographic data, family history, 23 clinical assessments, and three key features from brain MRI images. The use of data has been officially reviewed and approved by ADNI.
The first stage utilizes Generalized Association Plots (GAP v0.2.7) for exploratory data analysis, enabling simultaneous visualization of multiple data matrices through seriation techniques to reveal hidden patterns and relationships within the dataset. This method allows for comprehensive examination of complex interactions between clinical, demographic, and imaging variables.
The second stage employs Structural Equation Modeling using LISREL (v10) to test and validate the hypothesized relationships identified through GAP analysis. The SEM framework builds upon the discovered patterns, creating a comprehensive model that accounts for both observed variables and latent constructs in disease progression. This dual-methodology approach enables identification and validation of novel patterns across the spectrum from healthy controls to established Alzheimer's disease.
The first stage utilizes Generalized Association Plots (GAP v0.2.7) for exploratory data analysis, enabling simultaneous visualization of multiple data matrices through seriation techniques to reveal hidden patterns and relationships within the dataset. This method allows for comprehensive examination of complex interactions between clinical, demographic, and imaging variables.
The second stage employs Structural Equation Modeling using LISREL (v10) to test and validate the hypothesized relationships identified through GAP analysis. The SEM framework builds upon the discovered patterns, creating a comprehensive model that accounts for both observed variables and latent constructs in disease progression. This dual-methodology approach enables identification and validation of novel patterns across the spectrum from healthy controls to established Alzheimer's disease.
Results:
Our findings support the current hypothesis that abnormality of brain structure leads to disturbed human behavior, with GAP analysis identifying 7 distinct clusters of interconnected pathological patterns in Alzheimer's Disease. Two major pathways emerged from our structural equation modeling: Figure 1 (orange pathways) reveals significant correlations between physical activity, metabolic factors, and cognitive functions, with memory and speech serving as mediating functions between grey matter volumes and behavioral manifestations; Figure 2 (blue pathways) emphasizes socioeconomic influences on disease progression, highlighting relationships between lifestyle factors, social support, and disease state, while revealing a circular relationship between physical activity and metabolic factors. Both models demonstrate significant correlations between grey matter volume, white matter intensity, and cognitive function, supporting a cascade model of disease progression, with all pathways maintaining significance after Bonferroni correction.
·Figure 1 Female Adult Alzheimer's Disease Network Model
·Figure 2 Male Adult Alzheimer's Disease Network Model
Conclusions:
This research explored Alzheimer's disease relationships through GAP analysis and structural equation modeling, revealing pathways between brain structure, cognition, and behavior. We identified associations among metabolic factors, socioeconomic influences, and disease progression, with cognitive functions mediating between brain changes and behavior. Future work will integrate genetic and PET imaging data to enhance understanding and develop targeted therapies.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2
Modeling and Analysis Methods:
Methods Development
Multivariate Approaches 1
Other Methods
Novel Imaging Acquisition Methods:
Anatomical MRI
Keywords:
Computational Neuroscience
Data analysis
Degenerative Disease
Machine Learning
Modeling
MRI
Statistical Methods
Other - Data exploration
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.
Gomar, J. J., Bobes-Bascaran, M. T., Conejero-Goldberg, C., Davies, P., & Goldberg, T. E. (2021). Utility of combinations of biomarkers, cognitive markers, and risk factors to predict conversion from mild cognitive impairment to Alzheimer disease in patients in the Alzheimer's disease neuroimaging initiative. Archives of General Psychiatry, 68(9), 961-969.
Jack, C. R., & Holtzman, D. M. (2019). Biomarker modeling of Alzheimer's disease. Neuron, 80(6), 1347-1358.
Jack, C. R., Bernstein, M. A., Fox, N. C., Thompson, P., Alexander, G., Harvey, D., ... & Weiner, M. W. (2021). The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. Journal of Magnetic Resonance Imaging, 27(4), 685-691.
Kim, J. P., & Park, B. (2024). Structural equation modeling of ADNI biomarkers reveals temporal ordering of Alzheimer's disease progression. Neurobiology of Aging, 127, 22-34.
Li, X., Martinez-Murcia, F. J., & Johnson, K. A. (2024). Novel visual analytics frameworks for longitudinal brain imaging studies in Alzheimer's disease. Scientific Reports, 14, 1234-1248.
Martinez-Murcia, F. J., Gorriz, J. M., Ramirez, J., & Puntonet, C. G. (2020). Structure-based SEM for ADNI multimodal data: Diagnostic classification and disease progression. NeuroImage: Clinical, 28, 102-114.
McArdle, J. J., & Kadlec, K. M. (2019). Structural equation models of static and dynamic processes. Annual Review of Psychology, 64, 237-258.
Patel, A. B., & Ramirez, J. (2024). Interactive visualization tools for analyzing large-scale neuroimaging datasets: Applications in early Alzheimer's detection. Journal of Neuroscience Methods, 389, 109731.
Selkoe, D. J., & Hardy, J. (2020). The amyloid hypothesis of Alzheimer's disease at 25 years. EMBO Molecular Medicine, 8(6), 595-608.
Thompson, W. K., & Weiner, M. W. (2022). Structural equation models in longitudinal ADNI studies: A systematic review. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 14(1), e12247.