Research Background!Neuroimmune and cardiocerebrovascular diseases are a major part of the global disease burden. With population aging, lifecycle-based health management for these diseases is a key challenge for policymakers and medical experts. Large-scale cohort studies with genetic data offer a unique chance to find new biomarkers and develop accurate predictive models.
Aims of the Study!!1!To explore risk factors: Use generalized mixed linear models, GWAS and Mendelian Randomization to find environmental, clinical and genetic risk factors related to the morbidity, recurrence and mortality of these diseases.!2!To analyze MRI data: Use automated pipelines to analyze MRI data. Quantifying image data can support subjective findings and aid early disease detection.!3!To build predictive models: Apply Bayesian networks to predict the risk of morbidity, recurrence and mortality by integrating genetic, environmental and clinical data.!4!To offer scientific advice: Provide advice for preventing these diseases, improving patients’ prognosis and reducing the health burden.
Scientific Basis
These diseases contribute significantly to the global disease burden. Epidemiological studies have found many environmental risk factors and risk genes. Clinical researchers have studied the relationship between prognosis and clinical characteristics. However, few studies have built predictive models for the morbidity, recurrence and mortality of these diseases using high-dimensional data like UK Biobank. Also, few studies have integrated the onset and prognosis of the same disease into a lifecycle health management framework. Since disease onset and prognosis may share risk factors such as genetic background, lifestyle, or nutrition status, this is a key area to explore. Moreover, analyzing high-dimensional data is challenging. Most epidemiological studies lack MRI data, which is crucial for objective measurement and has predictive value. Fortunately, UK Biobank offers rich data.
