As for this invention, the segmentation unit 6,16,26 which segments the image object in the data set in regard to the image processing device which processes the multidimensional data set and the graphics data of medical volume of especially 3 dimensions which consist of the graphics data and the image processing method of corresponding, is included. Furthermore this invention regards the computer program which actualizes image processing method with the computer. Rationality to achieve with processing of brief being possible the result of segmentation being robustness with the vulnerability viewpoint for image arch fact, in order to make the segmentation which is accurate with the viewpoint which reflects the details with respect to structure possible, as for the image processing device, in order the model based segmentation expedient 61,161,261 which does the segmentation 32 which is based on the model of the data set in order to acquire the multidimensional model which displays the first approximation of image object, to acquire the multidimensional subset of the multidimensional data set which displays the second approximation of image object, the segmentation 33 which is driven by the data of the multidimensional data setThe model which was adjusted the data driven segmentation expedient 62,162,262 which is done, the model and/or the optimization expedient as a segmentation of 63,163,263 which adjusts subset, and image object in order to optimize the value of similarity with first approximation and the aforementioned second approximation, or it possesses the selective expedient 64,164,264 which selects the subset which was adjusted.本発明は、画像データからなる多次元のデータセット、特に3次元の医療ボリュームの画像データを処理する画像処理装置及び対応する画像処理方法に関し、データセットにおけるイメージオブジェクトのセグメント化を行うセグメント化ユニット6,16,26を含む。更に本発明は、コンピュータで画像処理方法を実現するコンピュータプログラムに関する。セグメント化の結果を合理的な短時間の処理で達成することができる、イメージアーチファクトに対する脆弱性の観点でロバストであって、構造上の詳細を反映する観点で正確であるセグメント化を可能にするため、画像処理装置は、イメージオブジェクトの第一の近似を表す多次元モデルを取得するためにデータセットのモデルに基づいたセグメント化32を行うモデルベースのセ
