Skip to Main Content
In this paper, we have proposed an image processing system for the acquisition and processing of three-dimensional images based on confocal scanning laser microscopy for the purpose of three-dimensional visualization and quantitative analysis of cell nuclei. The three-dimensional visualization methods can be divided into surface rendering and volume rendering. The way that surface rendering is used within this system is based on contour modeling. This method consists of several steps as follows. The first step is to preprocess the volume data obtained. Secondly, the extraction of the contours of each slice is carried out. Thirdly, smoothing algorithms are used to refine the contour data and remove wiggles. Since the surface rendering accounts only for the surface, the inside is not visible. Therefore, based on the basic volume rendering pipeline, we implemented the volume rendering. In the quantification step, in order to extract quantitative features, we made a three-dimensional labeling method based on slice information. Compared to the conventional algorithms, this method has advantages due to the use of memory is highly efficient and it is possible to combine a variety of two-dimensional labeling algorithms to find an appropriate labeling to its application. After applying the labeling algorithm, we extracted the measurements for the three-dimensional quantitative analysis of cell nuclei: nuclear volume, surface area and spherical shape factor. This could become a way to improve the accuracy and reproducibility of quantifying cell nuclei. We believe that our method will become a useful diagnostic tool for the medical image analysis.