With the advancement of an aging society, extending healthy life expectancy has become a critical challenge for managing healthcare costs and sustaining a healthy population. Achieving this objective requires the development of personalized preventive strategies and sophisticated medical interventions tailored to individual patient profiles, particularly for prevalent conditions such as cardiovascular diseases, diabetes, malignant tumors, and dementia.
Recent technological innovations have enabled high-throughput proteomic analyses capable of quantifying thousands of proteins simultaneously. This capability provides a more precise and dynamic understanding of human biology. Proteins are closely linked to the current disease state, future risk of disease onset, and the efficacy of interventions such as lifestyle modifications and pharmacological treatments. Therefore, proteomic data serve as a valuable resource for designing personalized preventive strategies.
This study is guided by the following key research questions:
* Can novel protein biomarkers be identified from large-scale proteomic datasets to predict the onset of multiple chronic diseases simultaneously?
* How can proteomic data be integrated with clinical outcomes to construct robust predictive models for disease classification, risk stratification, and intervention efficacy?
* What is the added value of combining proteomic data with other modalities (e.g., clinical information, lifestyle) in enhancing predictive accuracy?
Based on the research questions, the objectives of this study are:
* To discover novel biomarkers associated with disease onset and progression using proteomic datasets.
* To develop and validate predictive mathematical models using NEC’s proprietary AI algorithms, including survival analysis and explainable AI techniques.
* To assess the utility of these models in estimating health status, classifying disease subtyp