Novel deep learning radiomics nomogram-based multiparametric MRI for predicting the lymph node metastasis in rectal cancer: A dual-center study

Abstract

Purpose: To develop and evaluate a nomogram that integrates clinical parameters with deep learning radiomics (DLR) extracted from Magnetic Resonance Imaging (MRI) data to enhance the predictive accuracy for preoperative lymph node (LN) metastasis in rectal cancer.

Methods: A retrospective analysis was conducted on 356 patients diagnosed with rectal cancer. Of these, 286 patients were allocated to the training set, and 70 patients comprised the external validation cohort. Preprocessed T2-weighted and diffusion-weighted imaging performed preoperatively facilitated the extraction of DLR features. Five machine learning algorithms-k-nearest neighbor, light gradient boosting machine, logistic regression, random forest, and support vector machine-were utilized to develop DLR models. The most effective algorithm was identified and used to establish a clinical DLR (CDLR) nomogram specifically designed to predict LN metastasis in rectal cancer. The performance of the nomogram was evaluated using receiver operating characteristic curve analysis.

Results: The logistic regression classifier demonstrated significant predictive accuracy using the DLR signature, achieving an Area Under the Curve (AUC) of 0.919 in the training cohort and 0.778 in the external validation cohort. The integrated CDLR nomogram exhibited robust predictive performance across both datasets, with AUC values of 0.921 in the training cohort and 0.818 in the external validation cohort. Notably, it outperformed both the clinical model, which had AUC values of 0.770 and 0.723 in the training and external validation cohorts, respectively, and the stand-alone DLR model.

Conclusion: The nomogram derived from multiparametric MRI data, referred to as the CDLR model, demonstrates strong predictive efficacy in forecasting LN metastasis in rectal cancer.

Keywords: Deep learning; Lymph node metastasis; Nomogram; Radiomics; Rectal cancer.

Fig. 1

Patient recruitment workflow

Fig. 2

Flow chart of this study

Fig. 3

Utilization of the LASSO algorithm for filtering radiomics and DL features. Cross-validation plot (a), coefficient profile plot (b), and feature weight histogram (c). LASSO: least absolute shrinkage and selection operator

Fig. 4

Nomogram, decision curve analysis, and ROC curves. (a) A composite nomogram that integrates predictive clinical factors and DLR signatures to comprehensively assess LN metastasis in rectal cancer. Decision curve analyses for various models in the training cohort (b) and the external validation cohort (c) highlight the LR model’s efficacy based on the clinical signature, DLR signature, and the combined clinical-DLR signature in both the training cohort (d) and external validation cohort (e). This multifaceted analysis evaluates the comparative merits of these models, providing valuable insights into their clinical relevance and potential application. ROC: receiver operating characteristic curve; LN: lymph node; DLR: deep learning radiomics; LR: logistic regression

Fig. 5

Visualization of attention regions within the DL model for image analysis in rectal cancer patients, distinguishing between those with and without LN metastasis. The highlighted red areas indicate regions with higher weights, deemed more critical for predictive purposes. LN: lymph node