立體影像描繪法在醫學影像之應用

Abstract

立體數位化影像在許多領域中是非常重要的一項工具。在醫學的應用上， 醫師們也常用它用做診斷。在這篇論文中，我們建立了一套視覺化系統， 除了利用它來幫助醫師們處理立體數位化影像外，並能得到一些數量化的 資料，例如腦瘤的大小；這些數據對研究人員而言可能是非常重要的。在 這套系統中，我們將建立一新的資料結構，以便應用較多的處理技術；此 系統中使用了大量的形態學的技術來幫助我們做大致上的切割，並提供一 使用者界面以便醫師做更細部的處理。另外，在醫學的應用上，我們常不 能得到較高解析度的影像；這將使得所有的立體影像描繪法失效！故在這 篇論文中，我們發展出一新的技術來將原始影像做一先期處理，使其能適 用在傳統的立體影像描繪法。 Many area of science engineering rely on three-dimensional digital image data generated by a image device. In medical applications, surgeons make some diagnosis according to such 3-D data set, the CT or MRI. Here, we built a visulization system which can help surgeons to deal with such data. And we also expect to capture some numerical informations which may be lost by traditional manual method in this system. In this system, we develop a new data structure which enable us to apply more different techniques in the volume data and manipulate it easily. We aso use the morphological operations to help us segmenting the volumtric data as a preprocessing stage. Then we provide a interactive user-interface for helping users to segment data precisely. In our concentrated area, the medical data set, we usually can not get a high-resolution data set in clinical applications. It will cause all traditional algorithms of visualization can't work correctly. Thus we develop a new algorithm to modify the original data set, such that the modified data is suitable for traditional method.