EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Development of a venous thromboembolism risk prediction model for patients with primary membranous nephropathy based on machine learning

Entwicklung eines venösen-Thromboembolie-Risikovorhersagemodells für Patient:innen mit primärer membranöser Nephropathie auf Basis von Machine Learning

Lian Li, Liuyun Wu, Yin Wang, Hulin Wang, Xingyue Zheng, Lizhu Han, Qinan Yin, Xingwei Wu, Yuan Bian

Frontiers in Pharmacology · 06.11.2025 · 16

DOI: 10.3389/fphar.2025.1683708 ↗

Auf einen Blick

Bei 643 Patient:innen mit primärer membranöser Nephropathie wurde ein Machine-Learning-Modell zur Vorhersage des VTE-Risikos entwickelt. Das beste Modell erreichte eine AUC von 0,911.

Fragestellung

Lässt sich bei Patient:innen mit primärer membranöser Nephropathie das Risiko für venöse Thromboembolien mit Machine Learning vorhersagen?

Methode

Retrospektive Modellentwicklungsstudie mit Real-World-Daten; Daten von 1. Januar 2018 bis 30. September 2024; 8:2-Teilung in Trainings- und Testset; 14 Machine-Learning-Algorithmen, SHAP-Analyse.

Ergebnisse

643 Patient:innen, davon 93 mit VTE. Aus 504 Modellen wurde NGBoost mit KNN-Imputation, Borderline-SMOTE und Frequency-based Selection als bestes Modell gewählt; AUC 0,911. Das Modell umfasste 10 Merkmale.

Schlussfolgerung der Autor:innen

Die Studie entwickelte ein personalisiertes Risikovorhersagemodell für VTE bei PMN-Patient:innen und eine webbasierte Anwendung dazu. Das Modell zeigte eine starke Vorhersageleistung und kann die klinische Entscheidungsfindung unterstützen.

Kernaussagen

  • 643 Patient:innen mit primärer membranöser Nephropathie
  • 93 entwickelten eine VTE
  • Bestes Modell: NGBoost, AUC 0,911

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Pharmacology, online erschienen am 06.11.2025. Verfasst von Lian Li und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmacy, Personalized Drug Research and Therapy Key Laboratory of Sichuan Province, Sichuan Provincial People's Hospital, School of Me…. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

This study utilizes real-world data from primary membranous nephropathy (PMN) patients to preliminarily develop a venous thromboembolism (VTE) risk prediction model with machine learning. The aim is to improve the rational use of prophylactic anticoagulant therapy by predicting VTE risk in these patients.

Methods

We collected diagnostic and treatment data for PMN patients hospitalized at Sichuan Provincial People’s Hospital from 1 January 2018, to 30 September 2024. The data was divided into training and test sets at an 8:2 ratio, followed by processed using combinations of three imputation methods, three sampling methods, and three feature selection methods. After preprocessing, fourteen machine learning algorithms were employed to develop a predictive model for VTE risk in PMN patients. The SHapley Additive exPlanation (SHAP) method was used to interpret the contribution of outcome features. Finally, a VTE risk prediction tool for PMN patients was constructed using Streamlit.

Results

A total of 643 patients with PMN were included in the study, of whom 93 developed VTE. Among the 504 models constructed, the NGBoost model, which incorporated imputation by K-Nearest Neighbor, sampling by Borderline-SMOTE, and feature selection by Frequency-based Selection, was identified as the optimal model, achieving an area under the curve (AUC) of 0.911. The optimal model included ten features: D-dimer (DD), Fibrin Degradation Products (FDP)>5 mg/L, international normalized ratio (INR) of prothrombin, Recurrent nephrotic syndrome (RNS), cholinesterase (CHE), Urinary Microalbumin to Creatinine Ratio (umALB/Ucr), statins, antithrombin III (AT III) activity, albumin, and anti-phospholipase A2 receptor antibody (aPLA2Rab). Finally, an online predictive tool based on the optimal model was developed to provide real-time individualized VTE risk predictions for PMN patients.

Conclusion

This study developed a personalized risk prediction model for VTE in PMN patients using machine learning techniques. Additionally, a web-based tool for this predictive model was created. The model demonstrates strong predictive performance and can assist in clinical decision-making for the prevention and treatment of VTE in PMN patients.

Schlagwörter

Stichwörter der Autor:innen

machine learningprediction modelprimary membranous nephropathyriskfactorsvenous thromboembolism

Autor:innen und Institutionen

  • Department of Pharmacy, Personalized Drug Research and Therapy Key Laboratory of Sichuan Province, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chen…
  • The Fourth People's Hospital of Chengdu, Chengdu, China.
  • Department of Pharmacy, Chengdu Women's and Children's Central Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.

Zitieren

APA

Li, L., Wu, L., Wang, Y., Wang, H., Zheng, X., Han, L., Yin, Q., Wu, X., & Bian, Y. (2025). Development of a venous thromboembolism risk prediction model for patients with primary membranous nephropathy based on machine learning. Frontiers in Pharmacology, 16. https://doi.org/10.3389/fphar.2025.1683708

Vancouver

Li L, Wu L, Wang Y, Wang H, Zheng X, Han L, et al. Development of a venous thromboembolism risk prediction model for patients with primary membranous nephropathy based on machine learning. Frontiers in Pharmacology. 2025;16. doi:10.3389/fphar.2025.1683708

BibTeX
@article{Li2025,
  author = {Li, Lian and Wu, Liuyun and Wang, Yin and Wang, Hulin and Zheng, Xingyue and Han, Lizhu and Yin, Qinan and Wu, Xingwei and Bian, Yuan},
  title = {Development of a venous thromboembolism risk prediction model for patients with primary membranous nephropathy based on machine learning},
  journal = {Frontiers in Pharmacology},
  year = {2025},
  volume = {16},
  doi = {10.3389/fphar.2025.1683708}
}
Kostenlos starten →