Region-based progressive localization of cell nuclei in microscopic images with data adaptive modeling

Abstract

BackgroundSegmenting cell nuclei in microscopic images has become one of the most important routines in modern biological applications. With the vast amount of data, automatic localization, i.e. detection and segmentation, of cell nuclei is highly desirable compared to time-consuming manual processes. However, automated segmentation is challenging due to large intensity inhomogeneities in the cell nuclei and the background.ResultsWe present a new method for automated progressive localization of cell nuclei using data-adaptive models that can better handle the inhomogeneity problem. We perform localization in a three-stage approach: first identify all interest regions with contrast-enhanced salient region detection, then process the clusters to identify true cell nuclei with probability estimation via feature-distance profiles of reference regions, and finally refine the contours of detected regions with regional contrast-based graphical model. The proposed region-based progressive localization (RPL) method is evaluated on three different datasets, with the first two containing grayscale images, and the third one comprising of color images with cytoplasm in addition to cell nuclei. We demonstrate performance improvement over the state-of-the-art. For example, compared to the second best approach, on the first dataset, our method achieves 2.8 and 3.7 reduction in Hausdorff distance and false negatives; on the second dataset that has larger intensity inhomogeneity, our method achieves 5% increase in Dice coefficient and Rand index; on the third dataset, our method achieves 4% increase in object-level accuracy.ConclusionsTo tackle the intensity inhomogeneities in cell nuclei and background, a region-based progressive localization method is proposed for cell nuclei localization in fluorescence microscopy images. The RPL method is demonstrated highly effective on three different public datasets, with on average 3.5% and 7% improvement of region- and contour-based segmentation performance over the state-of-the-art.

On Visual Analytics and Evaluation in Cell Physiology: A Case Study

“…Experts in this domain are using their domain knowledge in combination with both automatic and visual analysis together, but need to be guided by computer science experts to improve the choice of tools that are used. There are certain tasks still done manually that could be automated or at least semi-automated, using [20] and automatic localization of cell nuclei [21]. The case study supports the statement, proposed by VA, that automated analysis often speeds up analysis tasks.…”

Large Margin Local Estimate With Applications to Medical Image Classification

“…Popular choices used in medical imaging include the linear discriminant analysis (LDA) [21], [22] and support vector machine (SVM) [2], [3], [9], [13], [16], [17], [22], [23]. These classifiers can be very effective in generating clear feature space separation with highly descriptive and discriminative feature descriptors.…”

A fast segmentation-free fully automated approach to white matter injury detection in preterm infants

“…As described later, one of the steps in our proposed method involves the localization of ventricles as a collection of blobs detected using the Maximally Stable Extremal Regions (MSER) algorithm [22] and optimized using Genetic Algorithms (GAs) [35]. MSER or its modified forms have been used earlier to detect various retinopathy pathologies [23], segment ultrasound liver images [24], localize cell nuclei in microscopic images [25], isolate fetal brain tissues from maternal anatomy during fetal brain in-utero MR imaging [26] and for 3D segmentation of simulated brain MR images [27]. However, they have not been tested in preterm brain WMI detection.…”