Biodata Mining & Applied Neuroinformatics Group, Bielefeld University, Universitaetsstrasse 25, 33615 Bielefeld, Germany.
Comput Med Imaging Graph. 2010 Sep;34(6):446-52. doi: 10.1016/j.compmedimag.2009.10.004. Epub 2009 Dec 6.
The challenging problem of computational bioimage analysis receives growing attention from life sciences. Fluorescence microscopy is capable of simultaneously visualizing multiple molecules by staining with different fluorescent dyes. In the analysis of the result multichannel images, segmentation of ROIs resembles only a first step which must be followed by a second step towards the analysis of the ROI's signals in the different channels. In this paper we present a system that combines image segmentation and information visualization principles for an integrated analysis of fluorescence micrographs of tissue samples. The analysis aims at the detection and annotation of cells of the Islets of Langerhans and the whole pancreas, which is of great importance in diabetes studies and in the search for new anti-diabetes treatments. The system operates with two modules. The automatic annotation module applies supervised machine learning for cell detection and segmentation. The second information visualization module can be used for an interactive classification and visualization of cell types following the link-and-brush principle for filtering. We can compare the results obtained with our system with results obtained manually by an expert, who evaluated a set of example images three times to account for his intra-observer variance. The comparison shows that using our system the images can be evaluated with high accuracy which allows a considerable speed up of the time-consuming evaluation process.
计算生物图像分析的难题受到生命科学领域越来越多的关注。荧光显微镜能够通过用不同的荧光染料染色来同时可视化多个分子。在多通道图像的结果分析中，ROI 的分割类似于第一步，之后必须进行第二步，以分析不同通道中 ROI 的信号。在本文中，我们提出了一个系统，该系统结合了图像分割和信息可视化原理，用于组织样本荧光显微照片的综合分析。该分析旨在检测和注释胰岛和整个胰腺的细胞，这在糖尿病研究和寻找新的抗糖尿病治疗方法中非常重要。该系统有两个模块。自动注释模块应用有监督的机器学习进行细胞检测和分割。第二个信息可视化模块可以用于根据链接和刷选原理进行交互式分类和可视化细胞类型，以过滤信息。我们可以将我们的系统获得的结果与专家手动获得的结果进行比较，专家为了考虑他的观察者内差异，对一组示例图像进行了三次评估。比较表明，使用我们的系统可以非常准确地评估图像，这允许大大加快耗时的评估过程。
