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: | Rahul Sarkar |
| Institution: | Calcutta University |
| Department: | Computer Science |
| Country: | |
| Proposed Analysis: | We aim to develop a machine learning algorithm that can accurately detect Alzheimer's disease (AD) based on neuroimaging data. Specifically, we plan to use magnetic resonance imaging (MRI) and positron emission tomography (PET) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to train and validate our algorithm. Our analysis will consist of the following steps: Data Preprocessing: We will preprocess the MRI and PET data to remove any noise, artifacts, or unwanted variability. This will involve steps such as skull stripping, normalization, and spatial registration. Feature Extraction: We will extract a set of features from the preprocessed MRI and PET data that capture the relevant information for AD detection. These features will include measures of brain structure, such as volume and thickness, as well as measures of functional connectivity, such as correlations between brain regions. Feature Selection: We will use various feature selection techniques to identify the most relevant and informative features for AD detection. These techniques may include correlation analysis, mutual information, or machine learning-based methods such as recursive feature elimination. Model Training and Validation: We will train and validate our machine learning algorithm using a subset of the ADNI dataset. We will use various machine learning algorithms such as Support Vector Machines (SVM), Random Forest (RF), and Deep Learning (DL) to classify AD patients and healthy controls based on the selected features. Model Evaluation: We will evaluate the performance of our machine learning algorithm using various metrics such as sensitivity, specificity, accuracy, and area under the curve (AUC). We will also compare the performance of our algorithm with existing methods for AD detection. Clinical Application: We will explore the clinical application of our machine learning algorithm by using it to predict AD status in a separate dataset. We will also investigate the potential use of our algorithm for early detection and monitoring of AD. |
| Additional Investigators | |
| Investigator's Name: | Sayan Das |
| Proposed Analysis: | The proposed analysis aims to develop a machine learning algorithm that can accurately detect Alzheimer's disease (AD) based on neuroimaging data. Specifically, we plan to use magnetic resonance imaging (MRI) and positron emission tomography (PET) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to train and validate our algorithm. Our analysis will consist of the following steps: Data Preprocessing: We will preprocess the MRI and PET data to remove any noise, artifacts, or unwanted variability. This will involve steps such as skull stripping, normalization, and spatial registration. Feature Extraction: We will extract a set of features from the preprocessed MRI and PET data that capture the relevant information for AD detection. These features will include measures of brain structure, such as volume and thickness, as well as measures of functional connectivity, such as correlations between brain regions. Feature Selection: We will use various feature selection techniques to identify the most relevant and informative features for AD detection. These techniques may include correlation analysis, mutual information, or machine learning-based methods such as recursive feature elimination. Model Training and Validation: We will train and validate our machine learning algorithm using a subset of the ADNI dataset. We will use various machine learning algorithms such as Support Vector Machines (SVM), Random Forest (RF), and Deep Learning (DL) to classify AD patients and healthy controls based on the selected features. Model Evaluation: We will evaluate the performance of our machine learning algorithm using various metrics such as sensitivity, specificity, accuracy, and area under the curve (AUC). We will also compare the performance of our algorithm with existing methods for AD detection. Clinical Application: We will explore the clinical application of our machine learning algorithm by using it to predict AD status in a separate dataset. We will also investigate the potential use of our algorithm for early detection and monitoring of AD. |
