Fig. 6From: Feasibility and effectiveness of automatic deep learning network and radiomics models for differentiating tumor stroma ratio in pancreatic ductal adenocarcinomaThe confusion matrix of all models and original images and the corresponding gradient weighted class activation mapping (Grad-CAM) generated by ResNet18 of the representative patients. a The figure shows the number of patients in the test set who were correctly and incorrectly classified. b A 63-year-old man was diagnosed with pancreatic ductal adenocarcinoma (PDAC) with a high tumor stroma ratio (TSR). The tumor was in the pancreatic head. b, c Sixty-two-year-old man was diagnosed with PDAC with a low TSR. The tumor was in the pancreatic tailBack to article page