Skip To Content
Cambridge University Science Magazine

The staggering complexity of the human body has resulted in a shift in the medical landscape. Big data fields and machine learning are increasingly empowering a new, information-based medicine. Most recently, machine learning has infiltrated one of the most guarded mysteries of the mind’s pathologies — dementia.

Affecting 50 million worldwide, dementia causes the progressive loss of language, memory, problem-solving and thinking abilities. The early symptoms are subtle and the accumulation of abnormal protein structures in the brain associated with dementia takes years to occur, so diagnosis can take months or years. Trained machine algorithms, however, look elsewhere.

Machine learning models developed by Professor Zoe Kourtzi’s team at the University of Cambridge and The Alan Turing Institute mine large-scale data on brain structure from neuroimaging scans of people living with Alzheimer’s. Alzheimer’s disease involves cognitive decline; however, this can range from mild cognitive impairment (MCI) to dementia. Using patterns of grey matter loss — a well-studied biomarker for Alzheimer’s disease — or performance scores on cognitive measurements, the algorithms could predict with over 80% accuracy whether patients will have stable or progressive MCI. Furthermore, an algorithm that combined these measurements could estimate the rate of future cognitive decline for individual patients.

“We’ve even been able to identify some patients who were not yet showing any symptoms, but went on to develop Alzheimer’s." Encouraged by the success of machine learning models, Professor Kourtzi is working to further identify different types of dementia by its characteristic patterns of grey matter loss. "In time, we hope to be able to identify patients as early as five to ten years before they show symptoms as part of a health check.” Such early detection would open up new medical research opportunities, as well as allow preventative treatment and care that can make life-changing improvements to patients and their families.