It is still a challenge to make confirming diagnosis of malignant pleural mesothelioma (MPM), especially differentiating from metastatic pleural disease (MPD). The aim of this study was to develop a model to distinguish MPM with MPD based on primary CT signs. We retrospectively recruited 150 MPM patients and 147 MPD patients from two centers and assigned them to training (115 MPM patients and 113 MPD patients) and testing (35 MPM patients and 34 MPD patients) cohorts. The images were analyzed for pleural thickening, hydrothorax, lymphadenopathy, thoracic volume and calcified pleural plaque (CPP). The selected clinical characteristics and primary CT signs comprised the model by multivariate logistic regression in the training cohort. Then the model was tested on the external testing cohort. ROC curve and F1 score were used to validate the capability of the model in both two cohorts. There were significant differences between two groups: (1) carcinoembryonic antigen (CEA); (2) nodular and mass pleural thickening; (3) the enhancement of pleura; (4) focal, diffuse and circumferential pleural thickening; (5) the thickest pleura; (6) thickening of diaphragmatic pleura; (7) multiple nodules and effusion of interlobar pleura; (8) hilar LN and ring enhancement of LN; (9) punctate and stipe CPP. The AUC and F1 score of the model were 0.970 and 0.857 in the training cohort, 0.955 and 0.818 in the testing cohort. The model holds promise for use as a diagnostic tool to distinguish MPM from MPD.

Background: To establish a predictive model based on multisequence magnetic resonance imaging (MRI) using deep learning to identify wild-type (WT) epidermal growth factor receptor (EGFR), EGFR exon 19 deletion (19Del), and EGFR exon 21-point mutation (21L858R) simultaneously. Methods: A total of 399 patients with proven brain metastases of non-small cell lung cancer (NSCLC) were retrospectively enrolled and divided into training (n = 306) and testing (n = 93) cohorts separately based on two timepoints. All patients underwent 3.0-T brain MRI including T2-weighted, T2-weighted fluid-attenuated inversion recovery, diffusion-weighted imaging, and contrast-enhanced T1-weighted sequences. Radiomics features were extracted from each lesion based on four sequences. An algorithm combining radiomics approach with graph convolutional networks architecture (Radio-GCN) was designed for the prediction of EGFR mutation status and subtype. The area under the curve (AUC) at receiver operating characteristic analysis was used to evaluate the predication capabilities of each model. Results: We extracted 1,290 radiomics features from each MRI sequence. The AUCs of the Radio-GCN model for identifying EGFR 19Del, 21L858R, and WT for the lesion-wise analysis were 0.996 ± 0.004, 0.971 ± 0.013, and 1.000 ± 0.000 on the independent testing cohort separately. It also yielded AUCs of 1.000 ± 0.000, 0.991 ± 0.009, and 1.000 ± 0.000 for predicting EGFR mutations respectively for the patient-wise analysis. The κ coefficients were 0.735 and 0.812, respectively. Conclusions: The constructed Radio-GCN model is a new potential tool to predict the EGFR mutation status and subtype in NSCLC patients with brain metastases. Relevance statement: The study demonstrated that a deep learning approach based on multisequence MRI can help to predict the EGFR mutation status in NSCLC patients with brain metastases, which is beneficial to guide a personalized treatment. Key points: • This is the first study to predict the EGFR mutation subtype simultaneously. • The Radio-GCN model holds the potential to be used as a diagnostic tool. • This study provides an imaging surrogate for identifying the EGFR mutation subtype. Graphical