Segmentační model pro automatickou detekci a modelování jaterní oblasti z medicínských obrazů

Abstract

This thesis focuses on describing the clinical aspects of the liver area and presents the most commonly used imaging methods for this area. This work aims to propose and implement a segmentation model for automatically detecting the liver area from medical images. A study of professional literature on currently used automatic segmentation methods for the liver area is necessary to select a suitable segmentation method for automated liver detection. The proposed model for automatic detection is subsequently tested for variable imaging conditions. Testing is performed on real CT liver images. Subsequently, an evaluation of the robustness of the proposed model is carried out. The final objective of this work is to detect significant structures from automatically detected liver area images. The model is designed and implemented in MATLAB, and Microsoft Excel is used for result processing.