Machine Learning for Prediction of High‐Risk Hospitalizations in Lymphoma Patients: A Danish Population‐Based Study
European Journal of Haematology · 2025
DOI: 10.1111/ejh.70012 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Journal of Haematology, 2025. Verfasst von Alexander Djupnes Fuglkjaer und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Materials and Production, Aalborg University, Aalborg, Denmark. Hauptthemen laut MEDLINE: Machine Learning, Hospitalization, Lymphoma. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Sygeforsikringen "Danmark", Danish Data Science Academy, Novo Nordisk Foundation.
Abstract (Originalsprache)
Objective
Infections are a leading cause of hospitalization in patients treated for lymphoma and can be life‐threatening. This study developed a machine learning (ML)‐based risk stratification method to classify infection‐related hospitalizations (IRH) into serious‐IRH (S‐IRH) and non‐serious‐IRH (NS‐IRH).
Methods
S‐IRH was defined based on death, positive blood culture, blood stream infection, and sepsis during hospitalization. Clinical data from health records and registries were used to construct the feature matrix for an XGBoost model. The study included 727 adult lymphoma patients diagnosed 2013–2023 and treated with first‐line therapies including CHOP (or CHOP‐like), ABVD, BEACOPP, CVP, and Bendamustine.
Results
A total of 591 IRHs were identified, of which 119 were categorized as S‐IRH. The developed model achieved a ROC‐AUC of 71.0% for predicting S‐IRH. At a prediction threshold of 0.3, the average sensitivity and specificity were 63.0% and 63.0%, respectively. The negative and positive predictive values were 87.5% and 30.3%, respectively.
Conclusion
The developed model demonstrated acceptable clinical performance in predicting S‐IRH. However, despite the wealth of data points entered in the model, performance was not sufficient for a stand‐alone decision tool, but it should rather be seen as a decision support tool for clinicians.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Materials and Production, Aalborg University, Aalborg, Denmark.
- Department of Oncology, Clinical Cancer Research Center, Aalborg University Hospital, Aalborg, Denmark.
- Department of Clinical Medicine, Aalborg University, Aalborg, Denmark.
- Department of Hematology, Clinical Cancer Research Center, Aalborg University Hospital, Aalborg, Denmark.
- Department of Mathematical Sciences, Aalborg University, Aalborg, Denmark.
- Copenhagen University Hospital, Rigshospitalet, Department of Hematology, Copenhagen, Denmark.
ORCID-Profile: Alexander Djupnes Fuglkjaer, Deniz Kenan Kilic, Paw Jensen
Förderung
- Sygeforsikringen "Danmark"
- Danish Data Science Academy
- Novo Nordisk Foundation
- VILLUM FONDEN
- Novo Nordisk Fonden
Zitieren
APA
Fuglkjaer, A. D., Kilic, D. K., Eskesen, M. H., Simonsen, M. R., Poulsen, L. Ø., Niemann, C. U., Jensen, P., Søgaard, K. K., Werling, M., Christensen, F., Nielsen, I. E., & El-Galaly, T. C. (2025). Machine Learning for Prediction of High‐Risk Hospitalizations in Lymphoma Patients: A Danish Population‐Based Study. European Journal of Haematology. https://doi.org/10.1111/ejh.70012
Vancouver
Fuglkjaer AD, Kilic DK, Eskesen MH, Simonsen MR, Poulsen LØ, Niemann CU, et al. Machine Learning for Prediction of High‐Risk Hospitalizations in Lymphoma Patients: A Danish Population‐Based Study. European Journal of Haematology. 2025. doi:10.1111/ejh.70012
BibTeX
@article{Fuglkjaer2025,
author = {Fuglkjaer, Alexander Djupnes and Kilic, Deniz Kenan and Eskesen, Mathias Holmsgaard and Simonsen, Mikkel Runason and Poulsen, Laurids Østergaard and Niemann, Carsten Utoft and Jensen, Paw and Søgaard, Kirstine Kobberøe and Werling, Mikkel and Christensen, Frederik and Nielsen, Izabela Ewa and El-Galaly, Tarec Christoffer},
title = {Machine Learning for Prediction of High‐Risk Hospitalizations in Lymphoma Patients: A Danish Population‐Based Study},
journal = {European Journal of Haematology},
year = {2025},
doi = {10.1111/ejh.70012}
}
EUVIMED