A Wegner's Granulomatosis Risk Prediction Model Based on Machine Learning Algorithms - Development and Evaluation

Aug 3, 2026

A machine learning–based approach was developed to predict Wegener’s granulomatosis relapse using clinical data from 189 hospitalized patients followed for about two years. Candidate models including Support Vector Machines, Random Forest, Gradient Boosting, and XGBoost were assessed by accuracy, precision, recall, and F1-measure, with performance further evaluated across different relapse rates and major relapse criteria based on the Birmingham Vasculitis Activity Score (BVAS). The XGBoost model achieved the highest accuracy (82% for more than one relapse rate; 92% for more than twice). SHAP analysis identified key predictive features, supporting earlier, more informed clinical decision-making.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Methods

Results

Conclusion

Journal of Biostatistics and Epidemiology J BiostatEpidemiol. 2023;9(2):189-200 Original Article A Wegner's Granulomatosis Risk Prediction Model Based on Machine Learning Algorithms Jaleh Shoshtarian Malak1, Samira Alsaeidi2*, Fatemeh Haji Ali Asgari1, Fahimeh Khedmatkon1 1Department of Electronic Health, Virtual School, Tehran University of Medical Sciences, Tehran, Iran. 2Rheumatology Research Center, Tehran University of Medical Sciences, Tehran, Iran. | ARTICLE INFO | ABSTRACT | | --- | --- | Received 27.12.2022 Revised 28.01.2023 Accepted 14.02.2023 Published 15.06.2023 Keywords: Wegener's granulomatosis relapse; Relapse prediction; Machine learning; Clinical decision-making; Xgboost algorithm; Birmingham vasculitis activity score; Predictive modeling; Healthcare analytics; Autoimmune diseases; Precision medicine Introduction: Prediction of Wegener's granulomatosis diagnosis and relapse is a complex process. In this study, we applied machine learning algorithms to predict Wegener's granulomatosis relapse. Methods: In this research, 189 patients admitted to Amiralam Hospital were studied and followed for approximately 2 years. Patient features included demographics, organ involvement, symptoms, and other clinical data. Different popular machine learning algorithms were applied for predicting Wegener's granulomatosis relapse, including Support Vector Machines, Random Forest, Gradient Boosting, and XGBoost algorithms. The prediction model performance was measured for the different candidate prediction algorithms using accuracy, precision, recall, and F1-measure. The selected prediction model performance was calculated based on different relapse rates and major relapse occurrence according to Birmingham Vasculitis Activity Score (BVAS) fields. Results: Applying different machine learning algorithms, the XGBoost algorithm performed the best. The results indicated that the prediction model's performance increased when calculating higher relapse rate possibilities. The XGBoost model had 82% accuracy while predicting more than one relapse rate and 92% accuracy in predicting more than twice the relapse rate. We also calculated the SHAP value for the prediction model. The results indicated that Cr, BVAS, lymphocyte percentage, vitamin D, nose involvement, alkaline phosphatase, diagnosis age, white blood cell count, erythrocyte sedimentation rate, and initial nose presentation are the 10 most important features according to SHAP value. Conclusion: In this study, we have developed Wegener's granulomatosis relapse prediction model using machine learning algorithms. We achieved reasonable precision and recall for early prediction and decisionmaking regarding Wegener's granulomatosis relapse Introduction Granulomatosis with Polyangiitis (Wegener's) disease, also known as GPA, is an autoimmune disease of unknown etiology. It causes symptoms in the sinuses, lungs, and kidneys, as well as other organs. GPA is diagnosed based on a combination of clinical manifestations * .Corresponding Author: [s-alesaeidi@sina.tums.ac.ir](s-alesaeidi@sina.tums.ac.ir), [salesaeidi@gmail.com](salesaeidi@gmail.com) Copyright © 2023 Tehran University of Medical Sciences. Published by Tehran University of Medical Sciences. This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International license ([https://creativecommons.org/licenses/by-nc/4.0/](https://creativecommons.org/licenses/by-nc/4.0/)) . Noncommercial uses of the work are permitted, provided the original work is properly cited. A Wegner's Granulomatosis Risk Prediction Model Based on ... including blood tests, X-rays, CT scans, and physical exams. Tissue biopsies of the affected organ are most often necessary to confirm the diagnosis. Antineutrophil cytoplasmic antibodies directed against proteinase 3 (PR3-ANCA) highly specify GPA. 1,2 GPA can be categorized as mild, moderate, or severe depending on the extent of organ involveme

FAQs

How was the Wegener’s granulomatosis relapse prediction model evaluated?

Model performance was measured using accuracy, precision, recall, and F1-measure, and then compared across different relapse rates and major relapse occurrence based on BVAS fields.

Which machine learning algorithm performed best in predicting relapse?

The XGBoost algorithm showed the best results, with 82% accuracy for predicting more than one relapse rate and 92% accuracy for predicting more than twice the relapse rate.

What were the most important features according to SHAP analysis?

SHAP identified Cr, BVAS, lymphocyte percentage, vitamin D, nose involvement, alkaline phosphatase, diagnosis age, white blood cell count, erythrocyte sedimentation rate, and initial nose presentation as the top features.

Aug 3, 2026

Upload by Emma Wilson