Scientific Rationale: Pregnancy-related disorders (PRDs), such as pre-eclampsia, preterm birth, and gestational diabetes, represent a significant burden of maternal and neonatal morbidity. However, their underlying risk architecture and inter-relationships remain incompletely characterized. Current research is often constrained by a focus on single disorders, short-term follow-up, and limited data types, which fails to capture the complex interplay of risk factors and long-term health sequelae. This fragmented approach hinders the development of comprehensive prevention strategies. There is a pressing need to integrate multimodal data to delineate the longitudinal dynamics of PRDs and formulate holistic prediction and intervention strategies.
Research Questions: 1) What genetic, biochemical, clinical, and lifestyle factors serve as significant risk markers for specific PRDs and their co-occurrence? 2) What are the distinct and shared longitudinal progression patterns of these conditions, and what are their long-term maternal health consequences? 3) Can we develop integrated and accurate predictive models for these disorders by combining multimodal data?
Objectives: This project aims to: identify and validate novel risk markers through comprehensive association analyses; map disease trajectories and long-term outcomes (e.g., cardiovascular and metabolic health) utilizing extensive longitudinal follow-up data; and construct robust machine learning-based prediction models that integrate genetic, clinical, and lifestyle variables.
