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: | Flavia Azevedo |
| Institution: | State University of Campinas (UNICAMP) |
| Department: | Institute of Computing |
| Country: | |
| Proposed Analysis: | The problem addressed in this research project refers to the need for advances in Alzheimer's disease detection methods to accurately detect the disease in its incipient stages. Our ultimate research goal is to design and develop an end-to-end neuroimaging-based solution for early Alzheimer's disease detection by applying Explainable Artificial Intelligence (XAI) techniques in a human-in-the-loop approach. Additionally, our secondary goals include, but are not limited to: 1. Studying the interplay between structural and functional brain patterns by integrating sMRI and fMRI neuroimaging into a joint solution framework. 2. Investigating machine learning techniques do deal with few training examples and algorithmic biases, since sMRI and fMRI datasets from the same patients are limited and and often comprising few instances. 3. Analyzing the progression of Alzheimer's disease over time through longitudinal data from the same patient. 4. Enabling domain shift reduction through transfer knowledge techniques to mitigate the lack of large datasets of specific interests, especially longitudinal ones. 5. Emphasizing mild cognitive impairment detection over cognitively normal vs. advanced Alzheimer's disease to encourage the methods to capture minimal initial levels of brain impairment, just above the detection threshold, thus advancing the frontiers of early diagnosis. 6. Innovating in the devise of explainable machine learning techniques to identify patterns of interest in images of specific regions of the brain for domain knowledge generation, in a process guided by neuroimaging specialists. 7. Eliciting domain experts' critical medical and biological questions that will drive the design of the XAI strategies. |
| Additional Investigators | |
| Investigator's Name: | Anderson Rocha |
| Proposed Analysis: | The problem addressed in this research project refers to the need for advances in Alzheimer's disease detection methods to accurately detect the disease in its incipient stages. Our ultimate research goal is to design and develop an end-to-end neuroimaging-based solution for early Alzheimer's disease detection by applying Explainable Artificial Intelligence (XAI) techniques in a human-in-the-loop approach. Additionally, our secondary goals include, but are not limited to: 1. Studying the interplay between structural and functional brain patterns by integrating sMRI and fMRI neuroimaging into a joint solution framework. 2. Analyzing the progression of Alzheimer's disease over time through longitudinal data from the same patient. 3. Investigating machine learning techniques do deal with few training examples and algorithmic biases, since sMRI and fMRI datasets from the same patients are limited and and often comprising few instances. 4. Enabling domain shift reduction through transfer knowledge techniques to mitigate the lack of large datasets of specific interests, especially longitudinal ones. 5. Emphasizing mild cognitive impairment detection over cognitively normal vs. advanced Alzheimer's disease to encourage the methods to capture minimal initial levels of brain impairment, just above the detection threshold, thus advancing the frontiers of early diagnosis. 6. Innovating in the devise of explainable machine learning techniques to identify patterns of interest in images of specific regions of the brain for domain knowledge generation, in a process guided by neuroimaging specialists. 7. Eliciting domain experts' critical medical and biological questions that will drive the design of the XAI strategies. |
