Özet:

Due to changes such as shape, border and density that occur in the slices of computed tomography (CT) images, liver segmentation remains a difficult process. Compared to other segmentation methods, more successful segmentation results with deep learning models are general phenomenon. In this study, a method accelerated with a multi-GPU is proposed for computer-aided automatic segmentation of the liver on CT scans obtained from the abdominal region using Mask Regional-Convolutional Neural Networks (Mask R-CNN). Experimental studies are conducted on liver CT image datasets to specific for this study with both single and double GPU hardware structure. The results obtained using the proposed method and the segmentation results realized by the specialist physician compared with parameters such as Dice similarity coefficient (DSC), Jaccard similarity coefficient (JSC), volumetric overlap error (VOE), average symmetric surface distance (ASD) and relative volume difference (RVD) metrics. In experimental studies carried out on the test images with the proposed approach, DSC, JSC, VOE, ASD and RVD segmentation performance metrics are gained as 97.32, 94.79, 5.21, 0.390, -1.008, respectively. With these results, it is seen that the proposed method in this study can be used as a secondary tool in the decision making processes of physicians for the segmentation of the liver.

Anahtar Kelimeler: