影像物件分割技術可以將器官或組織的輪廓描述出來，可以針對這些區域做特徵的分析，提供醫師或研究人員做為他們診斷或研究時的依據，本研究的主要目標是使用退化性膝關節炎患者的核磁共振影像作為研究對象，然而，在實際運用於醫學影像上的分析，則會產生重大的缺陷與情形（如雜訊敏感度、過度分割、邊界模糊或非均質區域的影響），我們使用影像濾波器進行前置處理，並依據Chan-Vese模式的等階集合法做影像分割，透過區域限制與分層區域限制進行最佳化，將影像序列中關節軟骨予以分割並提供相關的量化資訊（厚薄度和體積資訊）提供給臨床醫師作為術前評估，另一方面透過三維重建的技術也提供病患觀看實際患部受損情形，增加醫師與病患之間雙向的溝通。 The image segmentation technology can describe the contours of the organ or the structures. It can make the analysis of the characteristic to these areas, and offer to a doctor or the researchers in order to diagnose or study basis. The main target of this research is the use of degenerative knee Magnetic resonance imaging of patients as research subjects. However, when applied to medical image analysis, it has important drawbacks (ex:sensitivity to noise, over-segmentation, fuzzy boundary or the impact of non-homogeneous regions). We use the image pre-processing filter, and Chan-Vese model based Level set method of image segmentation, through the regional limitations and restrictions on the best stratification of the region. Image sequence will be divided in the articular cartilage and provide relevant quantitative information (thickness and volume of information) available to clinicians as a preoperative evaluation. On the other hand, through the three-dimensional reconstruction of the technology also provides patients with damage to watch the actual affected area where increase between physicians and patients with two-way communication.
