← All models
Oral Cancer (OSCC) Survival & Screening
Primary ModelSurvBoost is an AI-powered clinical decision support model predicting two-year survival probability for oral cancer patients using a Fully Parametric Gradient Boosting architecture composed of Weibull, Log-Logistic, Log-Normal, and Generalized Gamma hazard distributions, coupled with SHAP-based Explainable AI (XAI) feature attributions.
Expected Input
The model is intended to work with oral images. When the inference interface becomes available, submissions are expected to follow these guidelines:
- Intraoral photographs of the oral cavity.
- The relevant region of interest clearly visible in the frame.
- Adequate lighting and reasonable focus, so image features are discernible.
- Images used only within this research prototype — nothing is a clinical report.
Clinical Screening & Survival Analysis
Interactive prediction powered by SurvBoost Parametric Gradient Boosting and Kernel SHAP attributions.
SurvBoost Clinical Survival Predictor
Parametric Gradient Boosting & SHAP Explainable AI for Oral Squamous Cell Carcinoma
Checking backend...