KEY TAKEAWAYS
- The study aimed to investigate ML models for predicting LNM in patients with BUC preoperatively.
- Researchers noticed that the SVM model with 14 variables was the most effective for predicting LNM.
Lymph node metastasis (LNM) is associated with worse prognosis in patients with bladder urothelial carcinoma (BUC).
Junjie Ji and the team aimed to develop and validate machine learning (ML) models to preoperatively predict LNM in patients with BUC treated with radical cystectomy (RC).
They performed an inclusive analysis by retrospectively collecting demographic, pathological, imaging, and laboratory information of patients with BUC who underwent RC and bilateral lymphadenectomy. Patients were randomly categorized into training and testing sets. Five ML algorithms were utilized to establish prediction models.
The performance of each model was assessed by the area under the receiver operating characteristic curve (AUC) and accuracy. Finally, the corresponding variable coefficients were calculated based on the optimal model to reveal the contribution of each variable to LNM.
About 524 and 131 patients with BUC were finally enrolled into the training set and testing set, respectively. They identified that the support vector machine (SVM) model had the best prediction ability with an AUC of 0.934 (95% confidence interval [CI]: 0.903-0.964) and accuracy of 0.916 in the training set, and an AUC of 0.855 (95% CI: 0.777-0.933) and accuracy of 0.809 in the testing set. The SVM model contained 14 predictors, and positive lymph node imaging contributed the most to the prediction of LNM in patients with BUC.
The study concluded that the developed and validated ML models effectively preoperatively predict LNM in patients with BUC treated with RC, with the SVM model containing 14 variables demonstrating the best performance and high clinical applicability.
This study was partly funded by the Natural Science Foundation of Shandong Province (ZR2021MH354), Medical and health research program of Qingdao (2021-WJZD170).
Source: https://pubmed.ncbi.nlm.nih.gov/38872141/
Ji J, Zhang T, Zhu L, et al. (2024). “Using machine learning to develop preoperative model for lymph node metastasis in patients with bladder urothelial carcinoma.” BMC Cancer. 2024 Jun 13;24(1):725. doi: 10.1186/s12885-024-12467-4. PMID: 38872141; PMCID: PMC11170799.
