Explainable Survival Modeling in Pediatric Hematopoietic Stem Cell Transplantation Using Landmark-Based Machine Learning

Scientifica

卷:2026

Research Article

Khawar Siddiqui¹  Abdulrahman Al-Musa¹  Mona Al-Saleh²  Hawazen Al Saedi¹  Abdullah Al-Jefri¹  Ibrahim Ghemlas¹  Awatif Al Anazi¹  Saadiya Khan¹  Mouhab Ayas¹  Edward De Vol³  Ali Al-Ahmari¹ 

[1] Research Administrative Operations Department, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia, kfshrc.edu.sa

[2] Department of Pediatrics, College of Medicine, Imam Abdulrahman Bin Faisal University, Alkhobar, Saudi Arabia, iau.edu.sa

[3] Department of Pediatric Hematology/Oncology, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia, kfshrc.edu.sa

关键词: acute GvHD;    artificial intelligence;    black-box model;    global explanation;    hematopoietic stem cell transplantation;    local explanation;    pediatric;    sinusoidal obstruction syndrome;    survex;    veno-occlusive disease;   

DOI  :  10.1155/sci5/7818238

 received in 2025-01-05, accepted in 2026-04-18,  发布年份 2026

来源: Hindawi Publishing Corporation

【 摘 要 】

BackgroundMachine learning survival models can outperform conventional Cox regression in heterogeneous pediatric HSCT cohorts, but clinical adoption is limited by concerns regarding information leakage, time-dependent structure, and interpretability. We evaluated explainable survival modeling aligned to real-world decision points before transplantation and after early post-transplant complications.Materials and MethodsIn a single-center cohort of 839 transplant-naïve children (< 14 years) undergoing HSCT (2010–2016), overall survival was modeled from infusion to death or last follow-up. A baseline model used only pretransplant variables (age at HSCT, primary diagnostic group, stem cell source, and donor HLA type). To prevent leakage from postinfusion events, a day + 100 landmark cohort (n = 721; 121 deaths beyond day + 100) was constructed and extended with early complications known by day + 100 (SOS/VOD and acute GvHD). Cox proportional hazards, random survival forests (ranger), and randomForestSRC (rfSRC) were benchmarked using C-index, integrated cumulative/dynamic AUC (iAUC), and integrated Brier score (IBS). Cox proportional hazard assumptions were tested using Schoenfeld residuals. Model-agnostic explainability (survex) included time-dependent permutation importance, partial dependence survival profiles, SurvSHAP(t), ceteris paribus profiles, and additive attributions (breakdown and SHAP).ResultsOverall mortality was 28.1% (236/839). Baseline Cox violated proportional hazards (global p=0.00181), whereas the day + 100 landmark Cox did not (global p=0.52). Baseline performance favored ML models (C-index: Cox 0.673, ranger 0.677, rfSRC 0.668; iAUC: 0.698, 0.866, 0.826; IBS: 0.159, 0.126, 0.135). In the landmark setting, discrimination and accuracy improved with ML (C-index: Cox 0.721, ranger 0.870, rfSRC 0.840; iAUC: 0.609, 0.668, 0.650; IBS: 0.125, 0.0969, 0.1016). Explainability analyses produced clinically coherent global and individualized risk narratives beyond day + 100.ConclusionAn information-leakage-free, landmark-based XAI framework allows clinically credible survival prediction in pediatric HSCT, with random survival forests outperforming Cox regression and providing transparent cohort-level and patient-level explanations suitable for clinical communication and risk updating.

【 授权许可】

CC BY   
Copyright © 2026 Khawar Siddiqui et al. Scientifica published by John Wiley & Sons Ltd. 2026

【 图 表 】

Figure 9

Figure 8

Figure 7

Figure 15

Figure 4

Figure 13

Figure 2

Figure 8

Figure 9