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: | Norashikin Yahya |
| Institution: | Universiti Teknologi Petronas |
| Department: | Electrical & Electronic Engineering |
| Country: | |
| Proposed Analysis: | Alzheimer's is a progressive neurological disorder causing the brain cells to degenerate and die. Alzheimer's disease is the most common cause of dementia a general term for memory loss and other cognitive abilities that disrupts a person's ability to function independently. There has been continuous research effort to identify objective biomarkers that may assist diagnosis of early-onset Alzheimer's. The researches focused at investigating the neural patterns that contributed most to Alzheimer's classification. The study on patterns of activation for Alzheimer's with associated neural and psychological components contributes to the understanding of the etiology of neurological disorders. Previous study has shown that Resting-state functional Magnetic Resonance Image (RS-fMRI) is very useful to objectively discriminate neuropsychiatric disorders from healthy controls. However, low signal-to-noise ratio, high dimension, and small sample size are still the main challenges in Alzheimer's diagnostics using RS-fMRI. To overcome the difficulties, various signal processing and machine learning techniques have been experimented for data pre-processing, dimension reduction and accurate classification Alzheimer's patients. Functional network connectivity (FNC) estimated from RS-fMRI provides a powerful tool to assess brain functional architecture in health and disease states. It is established that there are FNC differences among patients, diagnosed with neuropsychiatric disorder such as bipolar and schizophrenia disorder. Thus, an accurate characterization of FNC benefits from understanding both normal and abnormal brain function in patients. In this work, classification framework based on deep neural network (DNN) and layer-wise relevance propagation (LRP) will be used to differentiate between individuals with Alzheimer's from healthy controls. The used of LRP is not only expected to improve classification accuracy but will also enables the identification of the most contributing FNC pattern related to Alzheimer's. It is expected that the classification phase will unveil different FNC patterns associated with Alzheimer's that can be used to objectively identify the early-onset Alzheimer's. |
| Additional Investigators | |
| Investigator's Name: | Alishba Sadiq |
| Proposed Analysis: | Alzheimer's is a progressive neurological disorder causing the brain cells to degenerate and die. Alzheimer's disease is the most common cause of dementia a general term for memory loss and other cognitive abilities that disrupts a person's ability to function independently. There has been continuous research effort to identify objective biomarkers that may assist diagnosis of early-onset Alzheimer's. The researches focused at investigating the neural patterns that contributed most to Alzheimer's classification. The study on patterns of activation for Alzheimer's with associated neural and psychological components contributes to the understanding of the etiology of neurological disorders. Previous study has shown that Resting-state functional Magnetic Resonance Image (RS-fMRI) is very useful to objectively discriminate neuropsychiatric disorders from healthy controls. However, low signal-to-noise ratio, high dimension, and small sample size are still the main challenges in Alzheimer's diagnostics using RS-fMRI. To overcome the difficulties, various signal processing and machine learning techniques have been experimented for data pre-processing, dimension reduction and accurate classification Alzheimer's patients. Functional network connectivity (FNC) estimated from RS-fMRI provides a powerful tool to assess brain functional architecture in health and disease states. It is established that there are FNC differences among patients, diagnosed with neuropsychiatric disorder such as bipolar and schizophrenia disorder. Thus, an accurate characterization of FNC benefits from understanding both normal and abnormal brain function in patients. In this work, classification framework based on deep neural network (DNN) and layer-wise relevance propagation (LRP) will be used to differentiate between individuals with Alzheimer's from healthy controls. The used of LRP is not only expected to improve classification accuracy but will also enables the identification of the most contributing FNC pattern related to Alzheimer's. It is expected that the classification phase will unveil different FNC patterns associated with Alzheimer's that can be used to objectively identify the early-onset Alzheimer's. |
| Investigator's Name: | Tong Boon Tang |
| Proposed Analysis: | Alzheimer’s disease is a progressive neurological disorder that causes brain cells to degenerate and die. Alzheimer’s disease is the most common cause of dementia, a general term for memory loss and other cognitive abilities that disrupts a person’s ability to function independently. There has been continuous research effort by neuropsychiatric imaging community to identify objective bio-markers that may assist the diagnosis of early-onset Alzheimer’s. Their research is focused at investigating the neural patterns that contributed most to Alzheimer’s classification. The study on patterns of activation for Alzheimer’s disease with associated neural and psychological components contributes to the understanding of the etiology of mental disorders. Previous study has shown that Resting-state functional Magnetic Resonance Image (rs-fMRI) is very useful to objectively discriminate neuropsychiatric disorders from healthy controls. However, low signal-to-noise ratio, high dimension, and small sample size are still the main challenges in Alzheimer’s diagnostics using rs-fMRI. To overcome the difficulties, various signal processing and machine learning techniques have been experimented for data pre-processing, dimension reduction and accurate classification Alzheimer’s patients. The local and global connectivity and amplitude of Resting state- functional magnetic resonance imaging (rs-fMRI) provides a powerful tool to assess human brain functional architecture in health, disease, and developmental states. It is established that there are significant changes in the amplitudes of rs-fMRI signals, diagnosed with brain related disorders. Thus, an accurate characterization of neuronal patterns may provide better benefits from understanding both normal and abnormal brain development or function in patients. The activation map of brain is commonly obtained by using the specific frequency band within the interval of 0.01-0.08 Hz. However, different frequency bands within the same neuronal network reveal different features of spontaneous fluctuations in physiological, and using one frequency band may only estimate the part of brain spontaneous pattern. In this work, multi-frequency based feature extraction (i.e., slow-5: 0.01-0.027 Hz, slow-4: 0.027-0.073 Hz, slow-3: 0.073-0.198 Hz, and slow-2: 0.198-0.25 Hz) from the rs-fMRI is proposed for the analysis of neuronal patterns and it will be used to differentiate between individuals with Alzheimer’s from healthy controls. We hypothesize that the multi-frequency based bio-markers may provide more accurate description of the neuronal pattern to Alzheimer’s disease and hence might help identify the early onset of Alzheimer’s disease objectively. |
