EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dynamic Prediction of Mortality Risk Following Allogeneic Hematopoietic Stem Cell Transplantation

Philipp Spohr, Rebecca C. Fröhlich, Sebastian Scharf, Anna Rommerskirchen, Jonathan Bobak, Sarah Selina Mathilde Schweier, Paul Jäger, Guido Kobbe, Sascha Dietrich, Alexander T. Dilthey, Birgit Henrich, Klaus Pfeffer und 2 weitere

Machine Learning: Health · 2025

DOI: 10.1088/3049-477x/adf74e ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Learning: Health, 2025. Verfasst von Philipp Spohr und 13 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Manchot Foundation: Research Group “Decision-making with the help of Artificial Intelligence”.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Allogeneic hematopoietic stem cell transplantation (alloHSCT) is a potentially curative treatment for high-risk hematological malignancies, but early mortality within 100 days remains a significant challenge, affecting approximately 18% of patients. Identifying high-risk patients early is critical for timely interventions. While static risk scores like the Hematopoietic Cell Transplantation-Specific Comorbidity Index (HCT-CI) and Endothelial Activation and Stress Index (EASIX) are valuable, they do not adapt to patients’ clinical trajectories. Leveraging the increasing availability of digital medical datasets, we developed a dynamic risk stratification system based on a random forest approach to continuously predict mortality risk from day 0 to day 30 post-transplantation. Analyzing data from 847 patients undergoing alloHSCT between 2004 and 2019, our model achieved high classification performance, with AUROC scores continuously exceeding 0.8 from day 17, and effectively stratified patients into high, moderate, and low-risk groups. Feature importance analysis revealed a shift over time, with early predictions utilizing diverse variables and later stages relying more on inflammation-related bloodwork parameters. This dynamic approach surpasses static methods, demonstrating improved predictive performance and adaptability across treatment phases. To promote reproducibility and broader application, we provide open access to the software and accompanying bloodwork time-series dataset.

Förderung

  • Manchot Foundation: Research Group “Decision-making with the help of Artificial Intelligence”

Zitieren

APA

Spohr, P., Fröhlich, R. C., Scharf, S., Rommerskirchen, A., Bobak, J., Schweier, S. S. M., Jäger, P., Kobbe, G., Dietrich, S., Dilthey, A. T., Henrich, B., Pfeffer, K., Haas, R., & Klau, G. W. W. (2025). Dynamic Prediction of Mortality Risk Following Allogeneic Hematopoietic Stem Cell Transplantation. Machine Learning: Health. https://doi.org/10.1088/3049-477x/adf74e

Vancouver

Spohr P, Fröhlich RC, Scharf S, Rommerskirchen A, Bobak J, Schweier SSM, et al. Dynamic Prediction of Mortality Risk Following Allogeneic Hematopoietic Stem Cell Transplantation. Machine Learning: Health. 2025. doi:10.1088/3049-477x/adf74e

BibTeX
@article{Spohr2025,
  author = {Spohr, Philipp and Fröhlich, Rebecca C. and Scharf, Sebastian and Rommerskirchen, Anna and Bobak, Jonathan and Schweier, Sarah Selina Mathilde and Jäger, Paul and Kobbe, Guido and Dietrich, Sascha and Dilthey, Alexander T. and Henrich, Birgit and Pfeffer, Klaus and Haas, Rainer and Klau, Gunnar W. W.},
  title = {Dynamic Prediction of Mortality Risk Following Allogeneic Hematopoietic Stem Cell Transplantation},
  journal = {Machine Learning: Health},
  year = {2025},
  doi = {10.1088/3049-477x/adf74e}
}
Kostenlos starten →