Related Experiment Video
Updated: Jun 24, 2025
09:28
Patient-derived Orthotopic Xenograft Models for Human Urothelial Cell Carcinoma and Colorectal Cancer Tumor Growth and Spontaneous Metastasis
Published on: May 12, 2019
10.1K
Using machine learning to develop preoperative model for lymph node metastasis in patients with bladder urothelial
Junjie Ji1, Tianwei Zhang1, Ling Zhu2
1Department of Urology, The Affiliated Hospital of Qingdao University, Qingdao, China.
BMC Cancer
|June 13, 2024
Summary
Machine learning models accurately predict lymph node metastasis in bladder urothelial carcinoma patients before surgery. The support vector machine model showed the best performance, aiding clinical decisions for bladder cancer treatment.
Area of Science:
- Urology
- Oncology
- Medical Informatics
Background:
- Lymph node metastasis (LNM) significantly worsens prognosis for bladder urothelial carcinoma (BUC) patients.
- Accurate preoperative prediction of LNM is crucial for treatment planning in BUC.
Purpose of the Study:
- To develop and validate machine learning (ML) models for the preoperative prediction of LNM in BUC patients undergoing radical cystectomy (RC).
Main Methods:
- Retrospective collection of demographic, pathological, imaging, and laboratory data from BUC patients who underwent RC.
- Development of prediction models using five ML algorithms, including support vector machine (SVM).
- Performance evaluation using area under the receiver operating characteristic curve (AUC) and accuracy on training and testing sets.
Main Results:
- The SVM model demonstrated superior predictive performance with an AUC of 0.934 and accuracy of 0.916 in the training set.
- The SVM model achieved an AUC of 0.855 and accuracy of 0.809 in the independent testing set.
- Positive lymph node findings on imaging were the most significant predictor of LNM in the SVM model.
Conclusions:
- Validated ML models, particularly the SVM model with 14 variables, can reliably predict LNM preoperatively in BUC patients.
- The developed SVM model offers high clinical applicability for improving surgical and treatment strategies in BUC.
