Coventry University researchers have developed a new tool to help diagnose Alzheimer’s disease.
Alzheimer’s disease is one of the most common neurodegenerative diseases, affecting 50 million patients worldwide, and that figure is expected to increase by 50% by 2050.
Current Alzheimer’s disease diagnosis methods like cognitive, physical, and radiological assessments can often be subjective, time-consuming, and invasive to the patient.
This research hopes that patient experiences will be enhanced by receiving a more accurate and quicker diagnosis.
Dr Fei He and Dominik Klepl, researchers within the Centre for Computational Science and Mathematical Modelling at Coventry University, have developed a unique diagnosis approach which analyses the brain dynamics from electroencephalography signals which measure brain electrical activity.
This ground-breaking research uses an energy landscape concept from statistical physics to model the patients’ EEG signals. It suggests that the findings could be used to improve the diagnosis of Alzheimer’s disease. This approach performs significantly better than alternative baseline diagnosis models and offers high levels of accuracy.
The energy landscape of the brain is a method of analysis that can be used to quantify the dynamics of brain transitions between stable states. These brain states illustrate different patterns of brain activities, the activation or depression in different brain regions at a specific time.
The EEG dynamics in those suffering from Alzheimer’s are more constrained than in non-sufferers, with the energy landscape of the brain showing more localised activity.
The results indicate that Alzheimer’s patients’ EEG signals are less complex, showing the increased difficulty of changing brain states compared to non-sufferers. This approach could be used to analyse other neurological disorders, including Parkinson’s disease.
Dr Fei He, assistant professor, Research Centre for Computational Science and Mathematical Modelling, said: ‘Our research shows the importance of studying the global dynamics of the brain in characterising neurological disorders, such as Alzheimer’s disease.’
‘The energy landscape technique and EEG could offer promising tools to support the diagnosis and characterise the severity of Alzheimer’s disease of a patient.’
‘This work also demonstrates the importance of multi-disciplinary research, such as integrating techniques from statistical physics, signal processing and machine learning, in tackling global challenges like a neurodegenerative disease.’
This research forms part of a collaborative project featuring: Dr Ptolemaios Sarrigiannis, Department of Neurophysiology, Royal Devon and Exeter NHS Foundation Trust, Dr Daniel Blackburn, Department of Neuroscience, University of Sheffield, Dr Min Wu, Institute for Infocomm Research, A*STAR, Singapore and Dr Matteo De Marco, Department of Life Sciences, Brunel University London.