DeepExposureNet

# DeepExposureNet **DeepExposureNet** is a temporal deep learning framework for modeling labor induction outcomes using multimodal clinical data. It incorporates latent temporal dynamics, multivariate outcome prediction, and policy optimization modules to support accurate, personalized delivery forecasting and decision support. ## 🧠 Project Highlights - **Latent Temporal Dynamics**: Captures evolving physiological states via continuous-time dynamics.- **Multivariate Outcome Modeling**: Jointly predicts time-to-event, delivery mode, and neonatal scores.- **Policy Optimization**: Computes utility-driven treatment adjustments in real-time.- **Clinical Interpretability**: Supports feature saliency and risk-aware recommendations. ## 📁 Project Structure