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: | Zainab Loukil |
| Institution: | University of Gloucestershire |
| Department: | Technical and Applied Computing |
| Country: | |
| Proposed Analysis: | Recently, Machine learning and physiological markers have been used by different researchers in order to enhance the early disease detection, which could potentially enhance the survival and recovery rate in patients. These mechanisms are useful but to a limited extent and therefore, make their integration with conventional mechanisms quite difficult. This research presents a framework that can accurately detect and predict different diseases. Generally, the diseases that require prediction have multiple stages and can progress differently with respect to each patient. DenCeption, combines Densenet and Inception network to be used on unique physiological markers in order to accurately predict and detect different diseases accurately. DenCeption is a three dimensional generic framework that can be used alongside conventional disease detection mechanisms to detect different stage-based diseases. To detect brain tumours, this novel framework incorporates EEG, MRI, along with age, gender, and disease progression in different individuals to accurately predict any possibility of different forms of tumour in unique individuals. The proposed hybrid DL-based network, namely, DenCeption is composed of two different architectures including DensNet and Inception-V4. |
| Additional Investigators | |
| Investigator's Name: | Qublai Ali Mirza |
| Proposed Analysis: | Towards disease prevention, medical scientists have been facing serious challenging scenarios related mainly to early disease diagnosis at a first step and stage progression tracking at a second. Despite all these effortful situations, including uncertainty of disease classification and decision making, the medical field is being in a progress towards improving existing physiological analysis methods that help in early disease detection and prediction. The latter has been of significant interest for medical and cross-domain researchers, particularly, disease prediction which presents one of the most crucial challenges, compared to disease detection, knowing that it requires disease progression tracking through time and multiple subjects’ factors. The problem is being currently handled by human interpretation or classic prediction methods, which neither are presenting as an accurate outcome. The aim of this project is to design and implement an automated mechanism that receives diverse physiological markers as input data and learn from it through extracting relevant features that will make the prediction (as an output) accurate and efficient. Towards achieving that, the system will be using Artificial Intelligence technique, specifically, Deep Learning (DL). The latter has been evolved significantly due to its ability to support medical applications, particularly prediction, by efficiently mapping the inputs, extracting the features, learning from them, and make the decision. |
