Detecting and preventing Alzheimer’s disease with machine learning

Our AEHRC researchers are using artificial intelligence and machine learning methods to combat Alzheimer’s disease, a deadly neurodegenerative disorder that accounts for 60-80% of dementia cases. As our population ages, the number of Australians suffering from dementia is set to increase.

Scientists and health professionals worldwide are engaged in efforts to identify Alzheimer’s risk factors and develop early detection methods. A major barrier in this research is a lack of data. Dr Rosita Shishergar, CSIRO Research Scientist, is solving this issue one line of code at a time. She created an algorithm that pulls together Alzheimer’s disease datasets from across Australia and around the globe, producing the largest Alzheimer’s dataset in the world.

Scientists like the AEHRC’s Dr Pierrick Bourgeat can use the dataset to train artificial intelligence algorithms. The Principal Research Scientist and his team are harnessing the power of machine learning to investigate the development and progression of Alzheimer’s disease. Their research into extracting information from medical images presents opportunities for use in measuring neurodegeneration and testing the accuracy of novel Alzheimer’s detection methods.

Dr Filip Rusak, CERC Fellow at CSIRO, is also interested in using medical images to monitor neurodegeneration. He used machine learning methods to generate a set of ‘fake’ brain MRIs with predefined signs of neurodegeneration. The dataset can be used to evaluate the sensitivity of methods that quantify the loss of brain tissue.

Watch the video below to hear directly from these innovate researchers.