ADNI data is made available to researchers around the world. As such, there are many active research projects accessing and applying the shared ADNI data. To further encourage Alzheimer’s disease research collaboration, and to help prevent duplicate efforts, the list below shows the specific research focus of the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Marc Dupuis |
| Institution: | Imperial College London |
| Department: | Biomedical Engineering |
| Country: | |
| Proposed Analysis: | IXICO Ltd. is an Imaging CRO, which is now moving into the healthcare domain. Their initial healthcare product uses image processing techniques that are already proven in the clinical trial space to calculate the volume of brain structures from a single T1 weighted MRI scan to support clinical diagnosis. The change in brain size is directly proportional to the advance of dementia and possibly Alzheimer’s in the patient. The information on brain size is then relayed to GP’s and pharmaceutical companies allowing them to track the advance of the disease. Currently, the quality of images is manually checked before further processing. Machine learning techniques need to be investigated for detecting and classifying poor quality images based on information derived automatically from the image data. This may involve signal-to-noise-ratio and histogram metrics or more advanced techniques. Some of the more advanced techniques to be considered include: minimum redundancy and maximum relevance and a support vector machine. Any automated classification will be supported by existing metadata checks to ensure that the right type of image has been uploaded to the system and used by the algorithm. As the volume measuring techniques are fully automatic and extremely robust, removal of the manual image processing step will facilitate wide scale rollout of Ixico’s software, ensuring that adequate quality image data can be acquired from all hospitals/imaging centres and processed in a completely automated solution. The project will involve investigation of techniques to extract metrics from the images and to apply machine learning algorithms (perhaps using those in the Weka framework or developing new techniques if required) to both discover metrics that are most suited for classification of images by image quality and to create classification models that can perform this classification automatically. The project will run alongside internal investigations along similar lines. A successful project could result in direct or indirect inclusion of the output into both Ixico’s healthcare and clinical trial products. |
| Additional Investigators |