Thank you. Listen to this article using the player above.
An international research team led by the Hong Kong University of Science and Technology (HKUST) has developed an artificial intelligence (AI)-based model that uses genetic information to predict an individual’s risk of developing Alzheimer’s disease (AD) well before symptoms appear. This groundbreaking study paves the way for using deep learning methods to predict disease risks and uncover their molecular mechanisms; this could revolutionize diagnosis, interventions and clinical research in AD and other common diseases such as cardiovascular disease.
The researchers led by the president of the HKUST, prof. Nancy IP, in collaboration with chair professor and director of the Big Data Institute of HKUST, prof. The team established one of the first deep learning models to estimate polygenic risks of AD in both populations of European and Chinese ancestry. Compared to other models, these deep learning models more accurately classify AD patients and stratify individuals into distinct groups based on disease risks associated with alterations in various biological processes.
In current day-to-day practice, AD is diagnosed clinically, using various means including cognitive testing and brain imaging, but often by the time patients show symptoms, it is already well beyond the optimal intervention window. Thus, early prediction of AD risk can greatly aid diagnosis and development of intervention strategies. By combining the new deep learning model with genetic testing, it is possible to estimate an individual’s lifetime risk of developing AD with greater than 70% accuracy.
Want more breaking news?
Subscribe to Technology networks daily newsletter, delivering the latest science news straight to your inbox every day.
Sign up for free
AD is an inherited disease that can be attributed to genomic variants. Because these variants are present from birth and remain constant throughout life, examining information about an individual’s DNA can help predict their relative risk of developing AD, thus allowing for early intervention and prompt management. Although FDA-approved genetic testing for the APOE-4 gene variant can estimate Alzheimer’s risk, they may not be sufficient to identify high-risk individuals, because multiple genetic risks contribute to the disease. Therefore, it is essential to develop tests that integrate information from multiple AD risk genes to accurately determine an individual’s relative risk of developing AD over their lifetime.
Our study demonstrates the effectiveness of deep learning methods for genetic research and risk prediction for Alzheimer’s disease. This breakthrough will significantly accelerate population-scale screening and risk staging for Alzheimer’s disease. In addition to risk prediction, this approach supports grouping of individuals based on their disease risk and provides insights into the mechanisms that contribute to disease onset and progression, said Prof. Nancy Ip.
Meanwhile, Prof. Chen Lei noted that this study exemplifies how the application of artificial intelligence to life sciences can bring significant benefits to biomedical and disease-related studies. Using a neural network, we effectively captured non-linearity in high-dimensional genomic data, which improved the accuracy of Alzheimer’s disease risk prediction. Furthermore, through AI-based data analysis without human supervision, we classified at-risk individuals into subgroups, which revealed insights into the underlying disease mechanisms. Our research also highlights how AI can address interdisciplinary challenges elegantly, efficiently and effectively. I strongly believe that AI will play a vital role in various healthcare sectors in the near future.
The study was conducted in collaboration with researchers from the Shenzhen Institute of Advanced Technology and University College London, as well as doctors from local hospitals in Hong Kong, including Prince of Wales Hospital and Queen Elizabeth Hospital. The results were recently published in Communication medicine. The research team is now refining the model and aims to eventually incorporate it into standard screening workflows.
AD, which affects more than 50 million people worldwide, is a deadly disease involving cognitive dysfunction and loss of brain cells. Its symptoms include progressive memory loss and impaired movement, reasoning, and judgment.
Reference: Zhou X, Chen Y, Ip FCF, et al. Deep learning-based polygenic risk analysis for Alzheimer’s disease prediction. Municipality Med. 2023;3(1):49. doi: 10.1038/s43856-023-00269-x
This article was republished from the following materials. Note: Material may have been modified in length and content. For more information, please contact the source cited.
#model #genetic #information #predict #risk #developing #Alzheimers #disease