Dynamic Survival Analysis

Dynamic survival analysis (DSA) focuses on modeling the time-to-event for individuals, allowing predictions to be updated as more data become available. Current research emphasizes improving prediction accuracy, particularly for early event prediction, using methods like neural controlled differential equations and signature-based estimators, as well as incorporating competing risks and temporal dependencies into model architectures. DSA's applications span diverse fields, including healthcare (e.g., predicting patient outcomes), finance, and predictive maintenance, offering more nuanced and timely predictions than traditional survival analysis. The improved accuracy and timeliness of DSA models are leading to more effective decision-making in various domains.