GCE: A Framework for Interpretable Nonlinear Hazard Modeling in Cardiac Sarcoma Survival Using SEER Data
Muhammad Shoaib Kareem, Madiha Amjad, Saba Aslam, Abdur Rasool, Mutiullah Jamil, Hazrat Ali · 2026
Read GCE: A Framework for Interpretable Nonlinear Hazard Modeling in Cardiac Sarcoma Survival Using SEER Data by Muhammad Shoaib Kareem, Madiha Amjad, Saba Aslam, Abdur Rasool, Mutiullah Jamil, Hazrat Ali (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13509387.txt.
Description
Kareem and colleagues built a tool to predict survival for cardiac sarcoma, a rare, fast-growing heart cancer, using United States registry data, in a 2026 methods study. Prediction is hard because cohorts are tiny, follow-up is incomplete, and risk factors act in curved ways. Their tool fuses a neural network learner suited to patterns with the classic Cox proportional hazards model, weighted together, with safeguards against information leakage.
Kareem found 727 patients and kept 27 predictors after screening. In leakage-free cross-validation the combined model ranked patients' risk better than the Cox model alone. But the Cox model's risk estimates matched real outcomes better, a trade-off the authors state plainly. Follow-up summaries described which factors drove the combined model's scores. They stress the high scores are likely too rosy and that the tool is a research aid that needs outside validation and recalibration before any clinical use.
Catalog record
| CREATOR | Muhammad Shoaib Kareem, Madiha Amjad, Saba Aslam, Abdur Rasool, Mutiullah Jamil, Hazrat Ali |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |