Using Artificial Intelligence for Early Diagnosis of Alzheimer’s Disease
Alzheimer’s Disease (AD) is the second most common reason for death in the Netherlands and has a strong impact on patients’ cognitive and physical functioning. More than 44 million people worldwide suffer from AD and this is projected to triple by 2050. Apart from the effects on patients, there is also a strong societal and economic impact, as global costs of AD are estimated at US$818 billion and keep rising. It can take more than 20 years for AD symptoms to appear, making it difficult to diagnose the disease in early stages. This project aims at building an Artificial Intelligence (AI) computer aided diagnosis (CAD) framework using image recognition with deep neural networks (NN) to early diagnose AD from magnetic resonance imaging (MRI). An AI-CAD framework can improve AD diagnosis at early stages and enable practitioners to further explore the development of the disease and study the factors that cause its deterioration. Novel treatments for AD focus on the early stages of the disease before dementia is present, making early diagnosis essential. If an AI-CAD framework is successful for the early diagnosis of AD, it creates the potential to also assist clinicians in other disease diagnoses. Therefore, in the short term, the research project will help the field to better diagnose AD at its early stages, and in the longer term, will use the AI-CAD framework to assist clinicians in the diagnosis of other diseases on their early stages. These objectives resonate with the aim of We Care to strengthen the scientific collaboration program of ETZ and TiU in health (care) research to improve patients care provision with the aid of AI.
Heising, L.M., and Angelopoulos, S. (2021). “Early Diagnosis of Mild Cognitive Impairment with 2-Dimensional Convolutional Neural Network Classification of Magnetic Resonance Images“, 54th Hawaii International Conference on System Sciences (HICSS).