There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Amna Saeed |
| Institution: | National University of Sciences & Technology (NUST) |
| Department: | Biomedical Sciences |
| Country: | |
| Proposed Analysis: | Title: Predicting Structural Changes in Alzheimer's Disease Patients Using Machine Learning Models Trained on Neuropsychological and MRI Data Proposed analysis: The objective of this study is to develop machine learning models that utilise neuropsychological data, specifically the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Rey Verbal Learning Test (RVLT), ADAA-Cog and Clinical Dementia Rating-Sum of boxes (CDR-SB), to predict structural changes in the brains of patients with Alzheimer's disease (AD). Methodology: We will utilise the ADNI database to access the necessary data, including the neuropsychological test scores (MMSE, MoCA, RVLT,ADAS-Cog, CDR-SB) and MRI scans. Our analysis will focus on a subset of patients diagnosed with AD based on clinical criteria. The dataset will be divided into training and testing sets, with a cross-validation strategy to ensure robust model performance. To predict structural changes in the brain, we will employ machine learning algorithms, such as support vector machines (SVM), random forests, or deep learning models. The training phase will involve using the neuropsychological test scores as input features and the corresponding MRI data as target labels. We will explore feature engineering techniques and select appropriate imaging features, such as volumetric measurements or cortical thickness, to enhance the predictive power of the models. Variables of interest: The main variables of interest include the MMSE, MoCA, RVLT, ADAS-Cog, CDR-SB scores as input features, and the structural MRI data (e.g., volumetric measurements, cortical thickness) as target labels. Additionally, demographic information (e.g., age, sex) and clinical diagnosis (AD) will be considered as potential covariates. Expected outcomes: We expect that our machine learning models will demonstrate the ability to predict structural changes in the brains of AD patients based on the neuropsychological data. The models will help identify specific cognitive tests or combinations of tests that are most strongly associated with structural alterations. Additionally, the predictive performance of the models will provide insights into the potential utility of these neuropsychological tests as non-invasive biomarkers for AD progression. Timeline: Considering the complexity of the analysis and model development, we estimate that the data preprocessing, model training, evaluation, and interpretation will be completed within a 6-8 month time frame from the date of data access approval. By utilising specific neuropsychological tests, MRI data, and machine learning techniques, this proposed analysis aims to investigate the potential relationship between cognitive performance and structural brain changes in AD patients. The findings may contribute to the development of early diagnostic tools or personalised interventions for AD and further our understanding of the disease pathology. |
| Additional Investigators | |
| Investigator's Name: | Maryam Nadeem |
| Proposed Analysis: | Title: Predicting Structural Changes in Alzheimer's Disease Patients Using Machine Learning Models Trained on Neuropsychological and MRI Data Proposed analysis: The objective of this study is to develop machine learning models that utilize neuropsychological data, specifically the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Rey Verbal Learning Test (RVLT), ADAA-Cog and Clinical Dementia Rating-Sum of boxes (CDR-SB), to predict structural changes in the brains of patients with Alzheimer's disease (AD). Methodology: We will utilize the ADNI database to access the necessary data, including the neuropsychological test scores (MMSE, MoCA, RVLT,ADAS-Cog, CDR-SB) and MRI scans. Our analysis will focus on a subset of patients diagnosed with AD based on clinical criteria. The dataset will be divided into training and testing sets, with a cross-validation strategy to ensure robust model performance. To predict structural changes in the brain, we will employ machine learning algorithms, such as support vector machines (SVM), random forests, or deep learning models. The training phase will involve using the neuropsychological test scores as input features and the corresponding MRI data as target labels. We will explore feature engineering techniques and select appropriate imaging features, such as volumetric measurements or cortical thickness, to enhance the predictive power of the models. Variables of interest: The main variables of interest include the MMSE, MoCA, RVLT, ADAS-Cog, CDR-SB scores as input features, and the structural MRI data (e.g., volumetric measurements, cortical thickness) as target labels. Additionally, demographic information (e.g., age, sex) and clinical diagnosis (AD) will be considered as potential covariates. Expected outcomes: We expect that our machine learning models will demonstrate the ability to predict structural changes in the brains of AD patients based on the neuropsychological data. The models will help identify specific cognitive tests or combinations of tests that are most strongly associated with structural alterations. Additionally, the predictive performance of the models will provide insights into the potential utility of these neuropsychological tests as non-invasive biomarkers for AD progression. Timeline: Considering the complexity of the analysis and model development, we estimate that the data preprocessing, model training, evaluation, and interpretation will be completed within a 6-8 month timeframe from the date of data access approval. By utilizing specific neuropsychological tests, MRI data, and machine learning techniques, this proposed analysis aims to investigate the potential relationship between cognitive performance and structural brain changes in AD patients. The findings may contribute to the development of early diagnostic tools or personalized interventions for AD and further our understanding of the disease pathology. |
