Preoperative prediction of lymph node metastasis in intrahepatic cholangiocarcinoma: an integrative approach combining ultrasound-based radiomics and inflammation-related markers

Abstract

Objectives: To develop ultrasound-based radiomics models and a clinical model associated with inflammatory markers for predicting intrahepatic cholangiocarcinoma (ICC) lymph node (LN) metastasis. Both are integrated for enhanced preoperative prediction.

Methods: This study retrospectively enrolled 156 surgically diagnosed ICC patients. A region of interest (ROI) was manually identified on the ultrasound image of the tumor to extract radiomics features. In the training cohort, we performed a Wilcoxon test to screen for differentially expressed features, and then we used 12 machine learning algorithms to develop 107 models within the cross-validation framework and determine the optimal radiomics model through receiver operating characteristic (ROC) curve analysis. Multivariable logistic regression analysis was used to identify independent risk factors to construct a clinical model. The combined model was established by combining ultrasound-based radiomics and clinical parameters. The Delong test and decision curve analysis (DCA) were used to compare the diagnostic efficacy and clinical utility of different models.

Results: A total of 1239 radiomics features were extracted from the ROIs of tumors. Among the 107 prediction models, the model (Stepglm + LASSO) utilizing 10 radiomics features ultimately yielded the highest average area under the receiver operating characteristic curve (AUC) of 0.872, with an AUC of 0.916 in the training cohort and 0.827 in the validation cohort. The combined model, which incorporates the optimal radiomics score, clinical N stage, and platelet-to-lymphocyte ratio (PLR), achieved an AUC of 0.882 in the validation cohort, significantly outperforming the clinical model with an AUC of 0.687 (P = 0.009). According to the DCA analysis, the combined model also showed better clinical benefits.

Conclusions: The combined model incorporating ultrasound-based radiomics features and the PLR marker offers an effective, noninvasive intelligence-assisted tool for preoperative LN metastasis prediction in ICC patients.

Clinical trial number: Not applicable.

Keywords: Inflammation-related marker; Intrahepatic cholangiocarcinoma; Lymph node metastasis; Radiomics; Ultrasound.

Fig. 1

Patient recruitment and grouping for the study. (a) According to the inclusion and exclusion criteria of the study, 82 patients with LN metastasis and 74 patients without LN metastasis were ultimately included. (b) Grouping of the patients

Fig. 2

Workflow of the radiomics analysis

Fig. 3

Cluster heatmap of radiomics features based on ultrasound medicine images Z-score normalization was applied to scale the quantitative expression values of 1239 extracted radiomics characteristics, and heatmap visualization was utilized to demonstrate the clustering patterns among these radiomics characteristics

Fig. 4

Predictive models were developed and validated by screening radiomics features with machine learning algorithms. A total of 107 predictive models were constructed, and the AUC values of each model in the training and validation cohorts and their average AUC values were calculated and ranked in descending order of average AUC values. The performance of each model is presented through three key indicators: Training cohort AUC, Validation cohort AUC, and their average. Average AUC= (training cohort AUC + validation cohort AUC) /2

Fig. 5

The use of radiomics features in predictive models. (a) Statistics of the number of times radiomics features used in the 107 predictive models, identifying the important features for this study. (b) Radiomics features of the top 10 models. The vertical coordinate values represent the total number of features used to construct the model, the horizontal coordinates represent the model’s name, and the bar graph colors represent the features used to construct the model

Fig. 6

Nomogram of the combined model for predicting LN metastasis in ICC

Fig. 7

Evaluation of the diagnostic performance of different models ROC curves for predicting LN metastasis were compared among the clinical, radiomics, and combined models in the training (a) and validation (b) cohorts. The DCA results for the three models are shown for the training (c) and validation (d) cohorts

Fig. 8

Confusion matrices for different models. Clinical model training cohort (a) and validation cohort (b). Radiomics model training cohort (c) and validation cohort (d). Combined model training cohort (e) and validation cohort (f). The horizontal coordinate represents the predicted label, and the vertical coordinate represents the real label. label 0: no lymph node metastasis, label 1: lymph node metastasis