目前CT等序列式醫學斷層影像普遍受各大醫院使用，因此醫學影像處理及二維輪廓點資料之三角網格模型重建有其研究之重要性，以利後續如人工植入物客製化、頭顱修補及牙模重建等應用。本研究目的便是從醫學影像取得之腿骨二維輪廓點資料重建其三角網格模型。首先進行點資料前處理以統一所有二維輪廓之格式，接著判斷各層輪廓組合的相對應關係，藉此對應關係研究其適用之網格連結方法，最後重建各輪廓組合之網格，合併輸出完整的腿骨三角網格模型。在重建結果的驗證上，與現有的影像處理軟體重建同筆腿骨二維輪廓點資料，並比較兩者之重建結果與原始輪廓比較之誤差量及網格品質之差異，以證明本研究所發展之三角網格模型重建方法之可行性。 Computer tomography (CT) has extensively been used as a tool for diseases diagnosis. Traditionally, a series of CT images are displayed sequentially in a screen for visualization. But, the CT images can provide additional information for the use in biomedical engineering. One of the techniques received extensive interest is the construction of the 3D model form 2D CT images. A series of section points corresponding to the hardware tissue to be studies should be obtained from the CT images first. The section points can then be converted into triangular meshes, extensively used in various fields to represent a 3D model. The purpose of this study is to develop a technique to construct triangular meshes from a series of section points. First, a pre-processing of the section points is necessary, including rearranging the points, removal of redundant points and scan line smoothing. Second, the relation between every two neighboring arrays of section points is investigated and the method is proposed to construct triangular meshes. Additional points may be added for some empty regions. Appropriate algorithm is proposed to ensure that the region with points inserted is smooth globally. Finally, the triangular meshes are output as a STL file format, which is a universal data format and can be used in many CAD/CAM systems. The proposed method is primarily used for the reconstruction of leg bone models. Therefore, extensive experiments and simulations are performed to verify the feasibility of the proposed method for the leg model construction. The results of the proposed method are also compared with those from AMIRA (commercial software) to verify the accuracy of the proposed method.
