| Title: | Development of artificial intelligence model for dementia precision medicine and diversification of diagnosis |
| Lead Institution: | Ajou University |
| Principal investigator: | Professor Hyunjung Shin |
Application 79011
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Diagnosis of dementia is mostly dependent on clinical tests such as neuropsychological testing tools. Diagnosis using brain imaging (MRI, PET) is only a means of confirmation because it diagnoses patients who have already progressed brain atrophy. Due to the nature of dementia, there is no concept of a cure, and prevention is the top priority, so an early diagnosis method with an index that can predict the onset time is required. For a diversified diagnosis of dementia, it is necessary to comprehensively reflect the characteristics of each degenerative brain disease. Therefore, it is necessary to utilize and integrate various information such as the genome, brain imaging, and mental health test. However, each test is reflected in the diagnosis through the opinion of the relevant expert, and it is somewhat inefficient in terms of time and cost because it requires a number of experts (clinical, radiologist, pathologist) as a whole. Therefore, it is necessary to use a machine learning algorithm that learns various information as multi-modal data through artificial intelligence convergence. By learning the expertise of each domain, it is possible to diversify and refine the diagnosis that comprehensively reflects the individual characteristics and complex onset patterns of dementia diseases. In addition, it is possible to establish elaborate treatment and prescription plans for patients, which can be used as a basis for precision medicine for dementia. This study starts with constructing an artificial intelligence model by integrating various information of dementia patients. Various types of information are divided into fields and used to diagnose diseases based on the knowledge of experts and doctors. This study replaces this with machine learning algorithms and constructs diagnostic models for each field. In addition, the diagnostic models constructed earlier are synthesized by developing an integrated module that imitates the consensus process of experts from various fields. In this case, the diagnostic model derives diversified results by reflecting the complex onset patterns of degenerative brain diseases. Compared to the existing fragmentary diagnosis, it is possible to diagnose a number of diseases, thereby increasing the prescribing and treatment efficiency. This study is research for the development of medical artificial intelligence, and the research is carried out with the goal of contributing to overcoming dementia. The research consists of discovering the key genes for dementia, selecting candidate therapeutics for re-creating new drugs, developing diversified diagnostic models, and developing precision medicine.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 15664 | PPIxGPN: plasma proteomic profiling of neurodegenerative biomarkers with protein-protein interaction-based eXplainable graph propagational network | Sunghong Park (+6) | 2025 | Briefings in Bioinformatics |
Enabling scientific discoveries that improve human health