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: | Dan Pan |
| Institution: | Guangdong Construction Polytechnic |
| Department: | Guangdong Polytechnic Normal University |
| Country: | |
| Proposed Analysis: | This study focuses on the early diagnosis of Alzheimer's disease (AD) based on imaging genetics. In order to acquire relatively objective biomarkers which are helpful in clinical diagnosis, we intend to make full use of all available data collected in clinics (e.g. neuroimaging data, genetic data, clinical neuropsychological assessment, epidemiological data etc.) with the help of the multimodal fusion techniques based on deep learning. We will research on the mechanisms and principles of deep learning to design a series of specific algorithms suitable for processing the above-mentioned data. We would like to consider “deep models” as a “brain” and to design a novel “visualization method” to observe and track the changes of weights and bias in the “deep models” with the changes of outer stimulus (i.e. input data). Based on the deep understanding of deep learning, we will try to design algorithms to tune and optimize the weights and bias in deep models to achieve better performances in feature selections and classifications. Based on the acquired deep models, we will construct a computer-aided system for the early diagnosis of AD and try to test this system in the related clinics affiliated to Liu-Hua-Hu Hospital located in Guangzhou, Guangdong province, China. In this study, we collaborate with Dr. Xiaowei Song, a Clinical Neuroimaging Senior Scientist, Health Sciences and Innovation in Fraser Health Authority and a Scientific Lead for MRI Program at ImageTech Laboratory in Surrey Memorial Hospital, Surrey, British Columbia, V3V 1Z2; Cell: 778.899.1681; Tel: 604.585.5666 (ext. 774.986), email: Xiaowei.Song@fraserhealth.ca and Prof. An Zeng at computer faculty in Guangdong University of Technology and a former postdoc at Faculty of Computer and Faculty of Medicine in Dalhousie University, email: zengan2010@126.com. Our research team has multiple years of experience in aging and dementia research and has finished several MRI research projects to promote brain health and physical health in aging. We thank you in advance for approving our applications. |
| Additional Investigators | |
| Investigator's Name: | Xiaowei Song |
| Proposed Analysis: | This study focuses on the early diagnosis of Alzheimer's disease (AD) based on imaging genetics. In order to acquire relatively objective biomarkers which are helpful in clinical diagnosis, we intend to make full use of all available data collected in clinics (e.g. neuroimaging data, genetic data, clinical neuropsychological assessment, epidemiological data etc.) with the help of the multimodal fusion techniques based on deep learning. We will research on the mechanisms and principles of deep learning to design a series of specific algorithms suitable for processing the above-mentioned data. We would like to consider “deep models” as a “brain” and to design a novel “visualization method” to observe and track the changes of weights and bias in the “deep models” with the changes of outer stimulus (i.e. input data). Based on the deep understanding of deep learning, we will try to design algorithms to tune and optimize the weights and bias in deep models to achieve better performances in feature selections and classifications. Based on the acquired deep models, we will construct a computer-aided system for the early diagnosis of AD and try to test this system in the related clinics affiliated to Liu-Hua-Hu Hospital located in Guangzhou, Guangdong province, China. |
| Investigator's Name: | An Zeng |
| Proposed Analysis: | This study focuses on the early diagnosis of Alzheimer's disease (AD) based on imaging genetics. In order to acquire relatively objective biomarkers which are helpful in clinical diagnosis, we intend to make full use of all available data collected in clinics (e.g. neuroimaging data, genetic data, clinical neuropsychological assessment, epidemiological data etc.) with the help of the multimodal fusion techniques based on deep learning. We will research on the mechanisms and principles of deep learning to design a series of specific algorithms suitable for processing the above-mentioned data. We would like to consider “deep models” as a “brain” and to design a novel “visualization method” to observe and track the changes of weights and bias in the “deep models” with the changes of outer stimulus (i.e. input data). Based on the deep understanding of deep learning, we will try to design algorithms to tune and optimize the weights and bias in deep models to achieve better performances in feature selections and classifications. Based on the acquired deep models, we will construct a computer-aided system for the early diagnosis of AD and try to test this system in the related clinics affiliated to Liu-Hua-Hu Hospital located in Guangzhou, Guangdong province, China. |
| Investigator's Name: | An Zeng |
| Proposed Analysis: | This study focuses on the early diagnosis of Alzheimer's disease (AD) based on imaging genetics. In order to acquire relatively objective biomarkers which are helpful in clinical diagnosis, we intend to make full use of all available data collected in clinics (e.g. neuroimaging data, genetic data, clinical neuropsychological assessment, epidemiological data etc.) with the help of the multimodal fusion techniques based on deep learning. We will research on the mechanisms and principles of deep learning to design a series of specific algorithms suitable for processing the above-mentioned data. We would like to consider “deep models” as a “brain” and to design a novel “visualization method” to observe and track the changes of weights and bias in the “deep models” with the changes of outer stimulus (i.e. input data). Based on the deep understanding of deep learning, we will try to design algorithms to tune and optimize the weights and bias in deep models to achieve better performances in feature selections and classifications. Based on the acquired deep models, we will construct a computer-aided system for the early diagnosis of AD and try to test this system in the related clinics affiliated to Liu-Hua-Hu Hospital located in Guangzhou, Guangdong province, China. We thank you in advance for approving our applications. |
| Investigator's Name: | Baoyao Yang |
| Proposed Analysis: | We will take advantage the proposed CNN ensemble framework to disclose the dynamic changes in the affected brain regions with the progress of AD. At the same time, we will try to utilize the unsupervised learning methods and/or advanced deep learning methods to identify the similarities and differences among the subjects in the datasets while optimizing the performance of AD classification. |
