Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 5, 2026npj Digital MedicineOpen Access

Development and explainable AI-driven characterization of a prognostic model for haploidentical transplantation outcomes

View Full Paper
Ask AI
Bookmark
Share

Authors

RMRohtesh S. MehtaYAYosra M. AljawaiPKPartow Kebriaei

Discussion

Loading...

Member takes

Overview

AI-driven model predicts haploidentical transplantation outcomes in a cohort, suggesting improved donor selection strategies.

Key Points

  • The aim is to develop a prognostic model for predicting outcomes in haploidentical transplantation using AI techniques.
  • Developed a prognostic model using a single-centre cohort of 668 patients.
  • Employed gradient boosting machine algorithms with explainable AI techniques.
  • Analyzed prognostic factors influencing overall survival, including donor and recipient age and HLA mismatches.
  • Identified recipient age as the dominant risk factor, affected by donor age and HLA factors.
  • Noted a U-shaped effect of donor age, with optimal outcomes for ages late 20s to early 40s.
  • Achieved a 3-year survival rate of 75% for lowest-risk patients, dropping to below 20% for highest-risk patients.

Cite This Study

Mehta et al. (2026) studied this question.

synapsesocial.com/papers/69a91df9d6127c7a504c1610https://doi.org/10.1038/s41746-026-02377-z
View Full Paper
Ask AI
Bookmark
Share