Survival analysis is a branch of statistics focused on modeling time-to-event data, playing a critical role in healthcare by informing prognosis, treatment planning, and risk stratification. Here at AIRLab, we develop deep learning architectures that capture complex, nonlinear relationships in multimodal patient records—such as genomic arrays and whole-slide images—to improve the accuracy and interpretability of survival predictions. By integrating representation learning with censoring-aware loss functions, we build computational survival models that scale to millions of records and competing events. Ultimately, our work pushes the boundaries of predictive medicine, enabling clinicians to tailor interventions based on precise, data-driven risk assessments.
Survival Analysis
Back to: Deep Learning