Interactive 3‐dimensional segmentation method based on region growing method

Abstract

AbstractIt has become easy to obtain three‐dimensional (3D) data of the inside of the human body because of developments in X‐ray computed tomography (CT) and magnetic resonance imaging (MRI). The 3D images of internal organs constructed from these data are useful in understanding the 3D shapes and positional relationships of the organs and are expected to be the diagnostic images of the next generation. However, many problems remain to be solved in making them practical, such as, improving the display speed of 3D images and implementing a 3D interface. For MRI, in particular, the difficulty in segmenting the display image is becoming a major impediment to the practical use of 3D MRI images. In this paper, we propose a practical segmentation method in which interactive corrective operations for 3D images are included in 3D region growing. To implement this method, we developed a 3D monitoring function for the segmentation process and an automatic function for removing the leaks generated during region growing. This method was applied to several 3D MRI data of the head and its effectiveness was confirmed.