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.

Though deep learning has shown successful performance in classifying the label and severity stage of certain disease, most of them give few evidence on how to make prediction. Here, we propose to exploit the interpretability of deep learning application in medical diagnosis. Inspired by Koch’s Postulates, a well-known strategy in medical research to identify the property of pathogen, we define a pathological descriptor that can be extracted from the activated neurons of a diabetic retinopathy detector. To visualize the symptom and feature encoded in this descriptor, we propose a GAN based method to synthesize pathological retinal image given the descriptor and a binary vessel segmentation. Besides, with this descriptor, we can arbitrarily manipulate the position and quantity of lesions. As verified by a panel of 5 licensed ophthalmologists, our synthesized images carry the symptoms that are directly related to diabetic retinopathy diagnosis. The panel survey also shows that our generated images is both qualitatively and quantitatively superior to existing methods.

This study aimed to establish a new clinical-radiomics nomogram based on ultrasound (US) for cervical lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC). We collected 211 patients with PTC between June 2018 and April 2020, then we randomly divided these patients into the training set (n = 148) and the validation set (n = 63). 837 radiomics features were extracted from B-mode ultrasound (BMUS) images and contrast-enhanced ultrasound (CEUS) images. The maximum relevance minimum redundancy (mRMR) algorithm, least absolute shrinkage and selection operator (LASSO) algorithm, and backward stepwise logistic regression (LR) were applied to select key features and establish a radiomics score (Radscore), including BMUS Radscore and CEUS Radscore. The clinical model and clinical-radiomics model were established using the univariate analysis and multivariate backward stepwise LR. The clinical-radiomics model was finally presented as a clinical-radiomics nomogram, the performance of which was evaluated by the receiver operating characteristic curves, Hosmer–Lemeshow test, calibration curves, and decision curve analysis (DCA). The results show that the clinical-radiomics nomogram was constructed by four predictors, including gender, age, US-reported LNM, and CEUS Radscore. The clinical-radiomics nomogram performed well in both the training set (AUC = 0.820) and the validation set (AUC = 0.814). The Hosmer–Lemeshow test and the calibration curves demonstrated good calibration. The DCA showed that the clinical-radiomics nomogram had satisfactory clinical utility. The clinical-radiomics nomogram constructed by CEUS Radscore and key clinical features can be used as an effective tool for individualized prediction of cervical LNM in PTC.