Comparison and Evaluation of Methods for Liver Segmentation From CT Datasets

摘要: This paper presents a comparison study between 10 automatic and six interactive methods for liver segmentation from contrast-enhanced CT images. It is based on results the "MICCAI 2007 Grand Challenge" workshop, where 16 teams evaluated their algorithms common database. A collection of 20 clinical images with reference segmentations was provided to train tune in advance. Participants were also allowed use additional proprietary training data that purpose. All then had apply test datasets submit obtained results. Employed include statistical shape models, atlas registration, level-sets, graph-cuts rule-based systems. compared five error measures highlight different aspects accuracy. combined according specific scoring system relating values human expert variability. In general, reached higher average scores than approaches featured better consistency quality. However, best (mainly models some free deformation) could compete well majority The provides an insight performance under real-world conditions highlights achievements limitations current image analysis techniques.