Enhancing brain metastasis prediction in non-small cell lung cancer: a deep learning-based segmentation and CT radiomics-based ensemble learning model

Abstract

Background: Brain metastasis (BM) is most common in non-small cell lung cancer (NSCLC) patients. This study aims to enhance BM risk prediction within three years for advanced NSCLC patients by using a deep learning-based segmentation and computed tomography (CT) radiomics-based ensemble learning model.

Methods: This retrospective study included 602 stage IIIA-IVB NSCLC patients, 309 BM patients and 293 non-BM patients, from two centers. Patients were divided into a training cohort (N = 376), an internal validation cohort (N = 161) and an external validation cohort (N = 65). Lung tumors were first segmented by using a three-dimensional (3D) deep residual U-Net network. Then, a total of 1106 radiomics features were computed by using pretreatment lung CT images to decode the imaging phenotypes of primary lung cancer. To reduce the dimensionality of the radiomics features, recursive feature elimination configured with the least absolute shrinkage and selection operator (LASSO) regularization method was applied to select the optimal image features after removing the low-variance features. An ensemble learning algorithm of the extreme gradient boosting (XGBoost) classifier was used to train and build a prediction model by fusing radiomics features and clinical features. Finally, Kaplan‒Meier (KM) survival analysis was used to evaluate the prognostic value of the prediction score generated by the radiomics-clinical model.

Results: The fused model achieved area under the receiver operating characteristic curve values of 0.91 ± 0.01, 0.89 ± 0.02 and 0.85 ± 0.05 on the training and two validation cohorts, respectively. Through KM survival analysis, the risk score generated by our model achieved a significant prognostic value for BM-free survival (BMFS) and overall survival (OS) in the two cohorts (P < 0.05).

Conclusions: Our results demonstrated that (1) the fusion of radiomics and clinical features can improve the prediction performance in predicting BM risk, (2) the radiomics model generates higher performance than the clinical model, and (3) the radiomics-clinical fusion model has prognostic value in predicting the BMFS and OS of NSCLC patients.

Keywords: Brain Metastasis; CT radiomics; Deep learning; Ensemble learning; Non-small cell Lung cancer.

Fig. 1

Workflow of the proposed BM prediction model

Fig. 2

The architecture and segmentation result of the deep residual U-Net network. (a) The architecture of the proposed deep residual U-Net network; (b) the output heatmap and segmentation result generated by the U-Net model. The images from top to bottom depict the original tumor images, output probability heatmaps of the deep U-Net model, and segmentation results of the proposed model and the ground truth

Fig. 3

Comparison and visualization of the selected features in the prediction model. (a) Boxplots of the selected features; (b) feature importance of the XGBoost classifier-based fused feature model; (c) examples of the selected radiomics features for BM and non-BM patients by using a voxel-based radiomics feature visualization technique

Fig. 4

ROC curves of the different models for the training and validation cohorts. (a)-(b) ROC curves of four different classifiers; (c)-(d) ROC curves and the corresponding AUC values of the radiomics, clinical and fused feature models, respectively

Fig. 5

DCA curves of the three models to assess the net benefits

Fig. 6

BMFS and OS KM survival curves for the training and validation cohorts in terms of the prediction scores generated by the fused feature model. (a)-(b) BMFS KM curves for the training and validation cohorts; (c)-(d) OS KM curves for the training and validation cohorts