How It Works
Our system simplifies the survival prediction process into three straightforward steps, ensuring clarity and accuracy.
Step 1: Input Data
Step 1: Input Data
The system accepts various input parameters, including patient demographics, medical history, and current health status. This data is crucial for generating accurate survival predictions.
Step 2: AI Processing
Step 2: AI Processing
Our advanced AI algorithms analyze the provided data, identifying complex patterns and correlations that are invisible to the human eye to build a highly accurate predictive model.
Step 3: Prediction
Step 3: Prediction
Based on the analysis, the system generates a clear and concise survival prediction, presented with confidence scores and explainable insights to aid in decision-making.
Project Context
This application is developed in collaboration with Dr. Wonsuk Yoo, Research Associate Professor at the School of Nursing and Health Studies, as part of his ongoing research on survival prediction in oncology. Leveraging retrospective randomized controlled trial (RCT) single-arm datasets, the system implements advanced artificial neural network (ANN) architectures to predict median progression-free survival (mPFS) and six-month progression-free survival (PFS6).
By integrating multi-task learning, calibrated outputs, and robust validation strategies, the platform provides a research-grade predictive environment that aligns computational modeling with clinical data. This work reflects a translational effort to bridge statistical evidence from prior trials with modern machine learning, supporting more rigorous prognostic assessment and decision-making in cancer research.