An international research team led by the Hong Kong University of Science and Technology (HKUST) has recently developed an artificial intelligence model that uses genetic information to accurately predict the risk of developing Alzheimer's disease before symptoms appear. This groundbreaking research opens the way for using deep learning methods to predict disease risk and reveal its molecular mechanisms. This will revolutionize the diagnosis, intervention, treatment and clinical research of Alzheimer's disease and other common diseases such as cardiovascular disease.
The research team led by Professor Ye Yuru, President of HKUST, and Chen Lei , Director of HKUST Data Research Institute, Dean of HKUST (Guangzhou) Information Hub, and Chair Professor, will study artificial intelligence models in this plan, especially to explore whether deep learning models can Using genetic information to assess Alzheimer's risk. The team built the first deep learning models to assess the polygenic risk of Alzheimer's disease in European and Chinese populations. Compared with other models, the deep learning model of HKUST can more accurately identify Alzheimer's patients, and also quantify the impact of genetic risk on various biological processes, and according to various diseases related to changes in biological processes Risk is graded and stratified for individuals.
Prof. Ye Yuru, President of HKUST (middle in the front row), Prof. Chen Lei, Director of the Data Institute of HKUST (second from the left in the front row), Prof. Fu Jieyu, Research Professor of the Department of Life Sciences of HKUST (first from the right in the front row), and Director of the Hong Kong Center for Neurodegenerative Diseases Scientist Dr. Ye Cuifen (front row, first from left) and the first author of the research paper, Professor Zhou Xiaopu (front row, second from right), took a group photo with other research team members.
At present, the clinical diagnosis of Alzheimer's disease is mainly carried out through doctor's judgment, cognitive ability scale test and brain scan, but it is usually carried out when the patient has symptoms, and the best intervention period is often missed. Therefore, early prediction of the risk of Alzheimer's disease can greatly help early diagnosis and development of intervention strategies. The study combined a novel deep learning model with genetic testing to estimate a person's lifetime risk of Alzheimer's disease with more than 70 percent accuracy.
Alzheimer's disease is a genetic disease that can be attributed to genetic variation. Since these genetic variations are inherited from parents at birth and remain unchanged throughout life, testing DNA information can effectively help predict the relative risk of Alzheimer's disease, so as to achieve early intervention and timely management of the disease. Although the U.S. Food and Drug Administration (FDA) has approved the use of mutations in the APOE-ε4 gene as a method for assessing the risk of Alzheimer's disease, because Alzheimer's disease is caused by multiple risk loci , testing for one risk gene alone may not be enough to identify high-risk individuals. Therefore, it is critical to develop a test that integrates information from multiple Alzheimer's disease risk genes to accurately assess an individual's relative risk of developing Alzheimer's disease during their lifetime.
Prof. Ye Yuru said: "Our study demonstrates the effectiveness of deep learning methods in genetic research and Alzheimer's disease risk prediction. This major breakthrough will speed up large-scale risk screening and risk classification for Alzheimer's disease. In addition to risk prediction, this method can also classify individuals according to disease risk, which provides new research ideas and insights into the pathogenesis and deterioration mechanisms of Alzheimer's disease."
Professor Chen Lei said: "This study demonstrates that the application of artificial intelligence in biological sciences can bring great benefits to biomedical and disease-related research. By using neural network models, we can effectively capture the abnormalities in high-dimensional genomic data. linear features, thereby improving the accuracy of Alzheimer's disease risk prediction. In addition, through artificial intelligence data analysis without human supervision, we divided at-risk individuals into various subgroups, revealing the underlying disease mechanism .This research highlights the potential of artificial intelligence to provide a powerful and efficient tool in solving interdisciplinary challenges. We firmly believe that artificial intelligence will play an important role in various medical fields in the near future."
The research was carried out in collaboration with researchers from the Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, University College London, and doctors from Prince of Wales Hospital and Queen Elizabeth Hospital in Hong Kong. The results of the study were recently published in Communications Medicine. The research team is currently further studying and refining the model, with the ultimate goal of incorporating it into routine screening procedures.
Alzheimer's disease, which affects more than 50 million people worldwide, is a fatal disease involving cognitive impairment and loss of brain cells. Symptoms include progressive memory loss and impaired reasoning and judgment.
-end-