Medical imaging of the brain plays a crucial role in the diagnosis of dementia. The rise of artificial intelligence in medical imaging in combination with the growth of clinical data and medical imaging made available for medical research provides opportunities for early diagnosis of cognitive decline and dementia. Diagnosing dementia in a preclinical phase of the disease is a steppingstone towards (targeted) prevention and intervention. However, major challenges in reaching early diagnosis are:
- The long preclinical phase of the disease before symptom onset.
- The heterogeneity of dementia.
- The large variation in healthy aging observed in a population.
- The strong association between age and the risk of developing dementia.
This project focuses on early diagnosis of dementia in primary care patients with subjective cognitive complaints. The project aims to predict whether someone with subjective cognitive complaints will develop cognitive decline or dementia, based on magnetic resonance imaging (MRI) of the brain using artificial intelligence.
As a PhD candidate on this project, you will perform state-of-the-art machine-learning based methodology, particularly deep neural networks and disease progression modeling, to predict cognitive decline and dementia based on the MRI of the brain. Your project will be embedded in the Scan2Go consortium, a national initiative which focusses on designing an easy-to-use autonomous MRI dedicated for diagnosing dementia in an early stage. A proof of principle of self-scanning and automatic quantification and reporting of brain damage from MRI will be developed with the ultimate aim to initiate a cultural change in the utilization of MRI, partially shifting from third line to first line care. In collaboration with other partners of the consortium, your role is to develop automated quantification and reporting of brain damage, based on the machine-learning based methodologies that support clinicians in diagnosing dementia.
You will be working with population-based as well as clinical data from different centers and will also contribute to the implementation of the automated quantification of brain damage within the Alzheimer Center of the Erasmus MC.