As Alzheimer’s disease continues to account for 60-80% of dementia cases in an ageing Australian population, researchers at Australian e-Health Research Centre (AEHRC) are turning to artificial intelligence (AI) and machine learning for answers.
Researchers have broken through a major barrier in research, which is data scarcity issues, by developing an algorithm that amalgamates Alzheimer’s datasets globally and within Australia, resulting in the world's most comprehensive Alzheimer’s disease database.
To build on this, AEHRC is utilising this extensive database to train AI algorithms, focusing on the analysis of medical images to measure neurodegeneration and assess the accuracy of emerging Alzheimer's detection methods.
How does it work?
Scientists at the CSIRO have employed machine learning techniques to generate a set of synthetic brain Magnetic Resonance Images (MRIs) displaying predefined neurodegenerative signs.
This artificial dataset serves as a benchmark for assessing the efficacy of methods quantifying loss of brain tissue.
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.