Fig. 4From: Multi-modal image analysis for semi-automatic segmentation of the total liver and liver arterial perfusion territories for radioembolizationTumor segmentation steps for one of the tumors (patient ID BI01 from training set in our database) that is visible in the slice by using the method proposed in [78]. For the other tumor(s), the same algorithm was used a initial 18F-FDG PET and (liver segmentation: green), b18F-FDG initial threshold and finding initial tumor burden (liver segmentation, green ; tumor cores, red), c finding specific background for each tumor (liver segmentation, green; tumor cores, red; background, blue), and d final tumor thresholding (liver segmentation, green; initial tumor volume which is not included in the final tumor segmentation, red; final tumor segmentation, blue)Back to article page