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: | Mohammed Rajab |
Institution: | The Sheffield University |
Department: | Computer Science |
Country: | |
Proposed Analysis: | There are a lot of factors and measures from looking at the brain that may be associated with dementia, but it is difficult to know which of these are the most important for diagnosis. Different pathologists would use different sets of features and that is where the discrepancies between diagnosis may occur. Yet, this explains why pathologist classification may not match with clinical classification. A pathologist examines features of the brain to resolve if the brain may have dementia or not. The issue is that different pathologists study the brain’s features in distinctive ways approaching different conclusions about dementia classification. What we can do with machine learning (ML) is that we can teach the computer to objectively look at the brain in a certain way to determine if the brain is likely to have dementia or not. Thus, potential automated approaches that can help the medical professional carefully select relevant factors/features would be based on computational intelligence and machine learning approaches. Both of these approaches can explore datasets attempting to bring up hidden useful patterns, in our case, features that have a clear impact on detecting dementia and hence reducing a tedious manual process. We aim to use the ADNI dataset, along with the clinical and neuropathology information, to improve the performance of dementia detection by automatically selecting features for the pathologist to use. This allows for the identification of clinical subtypes and molecular markers for each subtype. Furthermore, we want to compare how a pathologist thinks and what features he would pick up to determine dementia. This project may provide a new approach for dementia detection that considers the relationships among brain features themselves and relationships among the different types of cognitive features in the ADNI dataset and the class. More importantly, a new machine learning system can guide the clinician to use the minimum number of features from the patient. |
Additional Investigators | |
Investigator's Name: | Dennis Wang |
Proposed Analysis: | supervisor, Professor |
Investigator's Name: | Teruka Taketa |
Proposed Analysis: | Master student |