Developing Reasonable Forecasts for Post-HCT Survival using Kaplan-Meier and Machine Learning Approaches

DOI

10.2991/978-94-6463-940-7_20How to use a DOI?

Keywords

Kaplan–Meier Survival Analysis; Random Forest; Clinical Decision Support; Survival Prediction; Fairness in Healthcare AI; Medical Data Mining; Feature Importance

Abstract

Hematopoietic Cell Transplantation (HCT) is a critical therapy for hematologic disorders, yet post-transplant survival prediction remains complex due to heterogeneous clinical and genetic factors. This study presents a machine learning–driven framework that integrates Kaplan–Meier survival estimation with supervised classification models to forecast patient outcomes. The publicly available Equity in Post-HCT Survival Predictions dataset, comprising approximately 5,000 patient records with over 60 demographic, genetic, and clinical features, was utilized for analysis. Different machine learning models such as Logistic Regression, Decision Tree, Support Vector Machine, and Random Forest are trained and evaluated. Among these, the Random Forest model achieved the highest performance with 98.73% accuracy and an F1-score of 0.99, while feature importance analysis emphasized clinically relevant predictors such as age, donor attributes, comorbidity scores, and HLA matching. Kaplan–Meier survival curves provided temporal interpretability of outcomes, and fairness metrics, including demographic parity, were assessed to ensure equitable predictions across patient subgroups. Despite strong performance, the unusually high accuracies may indicate dataset imbalance and the absence of external validation, highlighting the need for further testing on multi-center cohorts. A lightweight application was also implemented to facilitate interactive survival forecasting. This framework demonstrates the potential of combining machine learning and survival analysis to enhance clinical decision support in HCT.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - D. N. S. B. Kavitha
AU  - M. Venkata Subbarao
AU  - Divya Lanka
AU  - K. Vijaya Naga Valli
AU  - T. Gayatri
AU  - K. Veera Raju
PY  - 2025
DA  - 2025/12/31
TI  - Developing Reasonable Forecasts for Post-HCT Survival using Kaplan-Meier and Machine Learning Approaches
BT  - Proceedings of the Conference on Social and Sustainable Innovation in Technology & Engineering (SASI-ITE 2025)
PB  - Atlantis Press
SP  - 271
EP  - 281
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-940-7_20
DO  - 10.2991/978-94-6463-940-7_20
ID  - Kavitha2025
ER  -