A Machine Learning Model Based on Multi-Phase Contrast-enhanced CT for the Preoperative Prediction of the Muscle-Invasive Status of Bladder Cancer

Sep 23, 2026

Muscle infiltration of bladder cancer is a key prognostic indicator, yet preoperative image-based assessment remains challenging. This retrospective study developed and validated a machine learning prediction model using multi-phase contrast-enhanced CT (MCECT) radiomics from lesions with pathologic confirmation. Patients were split 8:2 into training and testing sets, with an independent external test for verification. The pipeline used dual-sample t-test feature screening, LASSO regression for signatures, and ROC, calibration, and decision-curve analyses. The model achieved promising AUC and diagnostic performance for predicting muscle-invasive status.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Background

Objective

Methods

Results

Conclusion

Introduction

[1573-4056/25](1573-4056/25 Send Orders for Reprints to reprints@benthamscience.net)[ Send Orders for Reprints to](1573-4056/25 Send Orders for Reprints to reprints@benthamscience.net)[ ](1573-4056/25 Send Orders for Reprints to reprints@benthamscience.net)[reprints@benthamscience.net](1573-4056/25 Send Orders for Reprints to reprints@benthamscience.net) 1 Current Medical Imaging Content list available at: [https://benthamscience.com/journals/cmir](https://benthamscience.com/journals/cmir) RESEARCH ARTICLE A Machine Learning Model Based on Multi-Phase Contrast-enhanced CT for the Preoperative Prediction of the Muscle-Invasive Status of Bladder Cancer Xucheng He1,\#, Yuqing Chen1,\#, Shanshan Zhou1,\#, Guisheng Wang1, Rongrong Hua1, Caihong Li2, Yang Wang1, Xiaoxia Chen1,* and Ju Ye1,* 1Department of Radiology, the Third Medical Center, Chinese PLA General Hospital, Beijing100039, China 2Department of Radiology, Dongfang Hospital, Beijing University of Chinese Medicine, Beijing100039, China Abstract: Background: Muscle infiltration of bladder cancer has become the most important index to evaluate its prognosis. Machine learning is expected to accurately identify its muscle infiltration status on images. Objective: This study aimed to establish and validate a machine learning prediction model based on multi-phase contrast-enhanced CT (MCECT) for preoperatively evaluating the muscle-invasive status of bladder cancer. Methods: A retrospective study was conducted on bladder cancer patients who underwent surgical resection and pathological confirmation by MCECT scans. They were randomly divided into training and testing groups at a ratio of 8:2; we used an independent external testing set for verification. The radiomics features of lesions were extracted from MCECT images and radiomics signatures were established by dual sample T-test and least absolute shrinkage selection operator (LASSO) regression. Afterward, four machine learning classifier models were established. The receiver operating characteristic (ROC) curve, calibration, and decision curve analysis were employed to evaluate the efficiency of the model and analyze diagnostic performance using accuracy, precision, sensitivity, specificity, and F1-score. Results: The best predictive model was found to have logic regression as the classifier. The AUC value was 0.89 (5-fold cross-validation range 0.83-0.96) in the training group, 0.80 in the test group, and 0.87 in the external testing group. In the testing and external testing group, the diagnostic accuracy, precision, sensitivity, specificity, and F1-score were 0.759, 0.826, 0.863, 0.729, 0.785, and 0.794, 0.755, 0.953, 0.720, and 0.809, respectively. Conclusion: The machine learning model showed good accuracy in predicting the muscle infiltration status of bladder cancer before surgery. Keywords: Machine learning (ML), Bladder cancer, Contrast-enhanced computer tomography (CECT), Muscle-invasive, Receiver operating characteristic (ROC), Least absolute shrinkage selection operator (LASSO), Uroepithelial carcinoma. | Article History | Received: December 26, 2024 | Revised: January 20, 2025 | Accepted: February 11, 2025 | | --- | --- | --- | --- | 1. INTRODUCTION Bladder cancer is the second most common urological malignancy worldwide, whose incidence also ranks ninth among all malignancies globally [1] . Uroepithelial carcinoma accounts for approximately 90% of bladder cancer cases and * Address correspondence to these authors at the Department of Radiology, the Third Medical Center, Chinese PLA General Hospital, Beijing 100039, China; E-mails: [chenxiaoxia@301hospital-3.mil.cn](chenxiaoxia@301hospital-3.mil.cn), [yjhjx@163.com](yjhjx@163.com) \#These authors contributed equally. typically presents as multifocal and recurrent; other subtypes are squamous cell carcinoma and adenocarcinoma [2] . The latest research shows that it is more important to judge whether bladder cance

FAQs

What clinical problem does the study address?

The best-performing model used logistic regression, with AUC values reported across training, testing, and external testing sets (0.89, 0.80, and 0.87, respectively).

How were the datasets organized for model development and validation?

Patients were randomly divided into training and testing groups at an 8:2 ratio, and an independent external testing set was used to verify generalization.

Which evaluation indicators were used to measure diagnostic performance?

Performance was assessed using ROC curve analysis plus calibration and decision curve analysis, reporting accuracy, precision, sensitivity, specificity, and F1-score.

Sep 23, 2026

Uploaded by MrHarris58