AI is making some rapid strides in medical science. Researchers from the Mass General Birmingham have developed an AI-powered tool that can predict the future risk of cognitive impairment, including mild cognitive impairment and dementia, using sleep electroencephalography (EEG), a non-invasive process that records the electrical activity of the brain. According to the paper, researchers were able to predict future cognitive impairments with 77 per cent accuracy. The AI tool was developed by using sleep study data from a group of women over 65 who were tracked for five years. Reportedly, the researchers identified subtle changes in brain wave patterns that predicted which participants would be diagnosed with cognitive impairment later. The research, published in the Journal of Alzheimer’s Disease, states that wearable EEG devices may likely help in identifying individuals at risk of dementia. As per the research, the AI model detected changes in brain wave patterns for 85 per cent of the patients who went on to develop dementia. This is a significant milestone as it could improve Alzheimer’s outcomes by letting at-risk patients start treatment years earlier than usual. Shahab Haghayegh, one of the researchers, said that with the help of novel sophisticated analysis, advanced information theory tools, and AI, it was possible to detect subtle changes in the brain wave patterns during sleep that signal future cognitive impairment. The key findings of the research revealed that the best predictive model, using EEG features from deep sleep (N3 stage) and gamma frequency bands, achieved an accuracy score (AUC) of 0.76. Most importantly, this approach could enable at-home monitoring with wearable EEG devices. How can this help? The research can help in early detection. According to experts, identifying cognitive decline years in advance before symptoms appear can allow for early interventions. Furthermore, non-invasive monitoring with wearable EEG devices offers a convenient and affordable way to monitor brain health. Besides, early identification of at-risk individuals could lead to lifestyle changes or treatments to slow cognitive decline.
