Handcrafted and Deep Learning-Based Radiomic Models Can Distinguish GBM from Brain Metastasis

Abstract

Objective: The purpose of this study was to investigate the feasibility of applying handcrafted radiomics (HCR) and deep learning-based radiomics (DLR) for the accurate preoperative classification of glioblastoma (GBM) and solitary brain metastasis (BM).

Methods: A retrospective analysis of the magnetic resonance imaging (MRI) data of 140 patients (110 in the training dataset and 30 in the test dataset) with GBM and 128 patients (98 in the training dataset and 30 in the test dataset) with BM confirmed by surgical pathology was performed. The regions of interest (ROIs) on T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1WI (T1CE) were drawn manually, and then, HCR and DLR analyses were performed. On this basis, different machine learning algorithms were implemented and compared to find the optimal modeling method. The final classifiers were identified and validated for different MRI modalities using HCR features and HCR + DLR features. By analyzing the receiver operating characteristic (ROC) curve, the area under the curve (AUC), accuracy, sensitivity, and specificity were calculated to evaluate the predictive efficacy of different methods.

Results: In multiclassifier modeling, random forest modeling showed the best distinguishing performance among all MRI modalities. HCR models already showed good results for distinguishing between the two types of brain tumors in the test dataset (T1WI, AUC = 0.86; T2WI, AUC = 0.76; T1CE, AUC = 0.93). By adding DLR features, all AUCs showed significant improvement (T1WI, AUC = 0.87; T2WI, AUC = 0.80; T1CE, AUC = 0.97; p < 0.05). The T1CE-based radiomic model showed the best classification performance (AUC = 0.99 in the training dataset and AUC = 0.97 in the test dataset), surpassing the other MRI modalities (p < 0.05). The multimodality radiomic model also showed robust performance (AUC = 1 in the training dataset and AUC = 0.84 in the test dataset).

Conclusion: Machine learning models using MRI radiomic features can help distinguish GBM from BM effectively, especially the combination of HCR and DLR features.

Figure 1

Study workflow overview.

Figure 2

The Boruta selection of HCR + DLR features of T1CE data. Green and yellow indicate high-importance features, and blue and red indicate low-importance features.

Figure 3

Performance of different machine learning modeling methods in 6 feature groups. (a)–(c) ROC curves of the HCR models based on T1CE, T1WI, and T2WI data, respectively. (d)–(f) ROC curves of the HCR + DLR models based on T1CE, T1WI, and T2WI data, respectively.

Figure 4

Diagnostic performance of 6 feature groups. (a)-(b) ROC curves of the HCR model based on T1CE data in the training dataset and test dataset, respectively. (c)-(d) HCR + DLR models of T1CE data in the training and test datasets. (e)–(h) HCR and HCR + DLR models of T1WI data in the training dataset and test dataset. (i)–(l) Details of T2WI data.

Figure 5

ROC curves of the multimodality radiomic model. (a) Results of 10-fold cross-validation in the training dataset. (b) Comparison of different machine learning models. (c) Final multimodality model in the training dataset. (d) Final multimodality model in the test dataset.