Yuyang Yuan, Fuchun Zheng, Jiming Yao, Kun Zhou, Jiaqing Yang, Xiaoqiang Liu, Hao Wan, Luyao Chen, Jieping Hu, Lizhi Zhou, Bin Fu
{"title":"Personalized prediction for recurrence of cystitis glandularis: insights from SHAP and machine learning models.","authors":"Yuyang Yuan, Fuchun Zheng, Jiming Yao, Kun Zhou, Jiaqing Yang, Xiaoqiang Liu, Hao Wan, Luyao Chen, Jieping Hu, Lizhi Zhou, Bin Fu","doi":"10.21037/tau-2024-665","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Cystitis glandularis (CG) is a rare urological condition characterized by glandular metaplasia of the bladder mucosa. Recurrence following transurethral resection (TUR) is a significant clinical challenge. Traditional predictive models often fail to capture the complexity of the data, resulting in insufficient accuracy. In contrast, machine learning (ML) has demonstrated substantial potential in medical prediction by identifying and analyzing complex patterns that are undetectable by conventional methods. This study aims to develop and evaluate an interpretable ML model to predict recurrence after TUR for CG, thereby improving clinical decision-making and patient outcomes.</p><p><strong>Methods: </strong>We analyzed predictors of recurrence using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression. We developed and tested seven ML-based models: Cox proportional hazards model (CoxPH), LASSO regression, decision tree (rpart), random survival forest (RSF), gradient boosting machine (GBM), support vector machine (SVM), and extreme gradient boosting (XGBoost). Participants were diagnosed with CG by pathology following TUR and treated from 2012 to 2018. Model discrimination was assessed using the receiver operating characteristic (ROC) curve and area under the ROC curve (AUC), while model preference was evaluated through the Brier score (BS). Decision curve analysis (DCA) was used for model comparison. The SHapley Additive exPlanations (SHAP) method was employed for interpretation, providing insights into recurrence prediction and prevention strategies. Finally, user-friendly platform was developed, allowing users to predict CG recurrence by entering feature values into designated text boxes on the webpage.</p><p><strong>Results: </strong>The RSF model demonstrated the best performance in predicting recurrence, as indicated by superior ROC, DCA, and BS metrics. In SHAP, postoperative regular instillation (PRI) contributed the most to model construction.</p><p><strong>Conclusions: </strong>The RSF model effectively predicts CG recurrence, offering a framework for individualized treatment strategies. PRI was identified as the most significant risk factor influencing recurrence.</p>","PeriodicalId":23270,"journal":{"name":"Translational andrology and urology","volume":"14 3","pages":"808-819"},"PeriodicalIF":1.9000,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11986474/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Translational andrology and urology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.21037/tau-2024-665","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/3/26 0:00:00","PubModel":"Epub","JCR":"Q4","JCRName":"ANDROLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Cystitis glandularis (CG) is a rare urological condition characterized by glandular metaplasia of the bladder mucosa. Recurrence following transurethral resection (TUR) is a significant clinical challenge. Traditional predictive models often fail to capture the complexity of the data, resulting in insufficient accuracy. In contrast, machine learning (ML) has demonstrated substantial potential in medical prediction by identifying and analyzing complex patterns that are undetectable by conventional methods. This study aims to develop and evaluate an interpretable ML model to predict recurrence after TUR for CG, thereby improving clinical decision-making and patient outcomes.
Methods: We analyzed predictors of recurrence using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression. We developed and tested seven ML-based models: Cox proportional hazards model (CoxPH), LASSO regression, decision tree (rpart), random survival forest (RSF), gradient boosting machine (GBM), support vector machine (SVM), and extreme gradient boosting (XGBoost). Participants were diagnosed with CG by pathology following TUR and treated from 2012 to 2018. Model discrimination was assessed using the receiver operating characteristic (ROC) curve and area under the ROC curve (AUC), while model preference was evaluated through the Brier score (BS). Decision curve analysis (DCA) was used for model comparison. The SHapley Additive exPlanations (SHAP) method was employed for interpretation, providing insights into recurrence prediction and prevention strategies. Finally, user-friendly platform was developed, allowing users to predict CG recurrence by entering feature values into designated text boxes on the webpage.
Results: The RSF model demonstrated the best performance in predicting recurrence, as indicated by superior ROC, DCA, and BS metrics. In SHAP, postoperative regular instillation (PRI) contributed the most to model construction.
Conclusions: The RSF model effectively predicts CG recurrence, offering a framework for individualized treatment strategies. PRI was identified as the most significant risk factor influencing recurrence.
期刊介绍:
ranslational Andrology and Urology (Print ISSN 2223-4683; Online ISSN 2223-4691; Transl Androl Urol; TAU) is an open access, peer-reviewed, bi-monthly journal (quarterly published from Mar.2012 - Dec. 2014). The main focus of the journal is to describe new findings in the field of translational research of Andrology and Urology, provides current and practical information on basic research and clinical investigations of Andrology and Urology. Specific areas of interest include, but not limited to, molecular study, pathology, biology and technical advances related to andrology and urology. Topics cover range from evaluation, prevention, diagnosis, therapy, prognosis, rehabilitation and future challenges to urology and andrology. Contributions pertinent to urology and andrology are also included from related fields such as public health, basic sciences, education, sociology, and nursing.