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: | Gao Hanxing |
| Institution: | Fujian Normal University |
| Department: | College of Optoelectronics and information enginee |
| Country: | |
| Proposed Analysis: | Well preprocessed data along with proper classifier could bring an accurate computational model for supporting clinicians to diagnose the severity of Alzheimer’s disease (AD). This work proposed a hybrid framework containing 2688 combinations of 8 methods of handling missing values, 11 feature selection algorithms, 6 strategies of balancing imbalanced data, followed by 4 classifiers, addressing the problem of data-driven modelling by extensively using multiple preprocessing and classification workflows based on per-class diagnostic classification accuracy of healthy controls, mild cognitive impairment, and AD. All approaches are easily interpretable. Taking the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data as an example, by visualising the experimental results we found the outperformed combination of preprocessing methods and classifier not only for the whole class but also focusing on per-class classification, with 93% ~ 98% accuracy. More importantly, we further developed an interactive toolbox with human-centred graphical user interface (GUI) to evaluate the user-chosen preprocessing and classifier. This provides non-technical users, e.g. clinicians or other stakeholders, a platform to fully make use of various data mining and machine learning techniques in AD diagnosis. The potential is that our work generalises to other biomedical research for building up a diagnostic decision support system. |
| Additional Investigators |
