Sparse Coding Driven Deep Decision Tree Ensembles for Nucleus Segmentation in Digital Pathology Images

Automating generalized nucleus segmentation has proven to be non-trivial and challenging in digital pathology. Most existing techniques in the field rely either on deep neural networks or on shallow learning-based cascading models. The former lacks theoretical understanding and tends to degrade performance when only limited amounts of training data are available while the latter often suffers from limitations for generalization. To address these issues, we propose sparse coding driven deep decision tree ensembles (ScD²TE), an easily trained yet powerful representation learning approach with performance highly competitive to deep neural networks in the generalized nucleus segmentation task. We explore the possibility of stacking several layers based on fast convolutional sparse coding-decision tree ensemble pairwise modules and generate a layer-wise encoder-decoder architecture with intra-decoder and inter-encoder dense connectivity patterns. Under this architecture, all the encoders share the same assumption across the different layers to represent images and interact with their decoders to give fast convergence. Compared with deep neural networks, our proposed ScD²TE does not require back-propagation computation and depends on less hyper-parameters. ScD²TE is able to achieve a fast end-to-end pixel-wise training in a layer-wise manner. We demonstrated the superiority of our segmentation method by evaluating it on the multi-disease state and multi-organ dataset where consistently higher performances were obtained for comparison against other state-of-the-art deep learning techniques and cascading methods with various connectivity patterns.

Automatic three-dimensional segmentation of mouse embryonic stem cell nuclei by utilising multiple channels of confocal fluorescence images

Time-lapse confocal fluorescence microscopy images from mouse embryonic stem cells (ESCs) carrying reporter genes, histone H2B-mCherry and Mvh-Venus, have been used to monitor dynamic changes in cellular/differentiation characteristics of live ESCs. Accurate cell nucleus segmentation is required to analyse the ESC dynamics and differentiation at a single cell resolution. Several methods used concavities on nucleus contours to segment overlapping cell nuclei. Our proposed method evaluates not only the concavities but also the size and shape of every 2D nucleus region to determine if any of the strait, extrusion, convexity and large diameter criteria is satisfied to segment overlapping nuclei inside the region. We then use a 3D segmentation method to reconstruct simple, convex, and reasonably sized 3D nuclei along the image stacking direction using the radius and centre of every segmented region in respective microscopy images. To avoid false concavities on nucleus boundaries, fluorescence images of the H2B-mCherry reporter are used for localisation of cell nuclei and Venus fluorescence images are used for determining the cell colony ranges. We use a series of image preprocessing procedures to remove noise outside and inside cell colonies, and in respective nuclei, and to smooth nucleus boundaries based on the colony ranges.  We propose dynamic data structures to record every segmented nucleus region and solid in sets (volumes) of 3D confocal images. The experimental results show that the proposed image preprocessing method preserves the areas of mouse ESC nuclei on microscopy images and that the segmentation method effectively segment out every nucleus with a reasonable size and shape. All 3D nuclei in a set (volume) of confocal microscopy images can be accessed by the dynamic data structures for 3D reconstruction. The 3D nuclei in time-lapse confocal microscopy images can be tracked to calculate cell movement and proliferation in consecutive volumes for understanding the dynamics of the differentiation characteristics about ESCs. LAY DESCRIPTION: Embryonic stem cells (ESCs) are considered as an ideal source for basic cell biology study and producing medically useful cells in vitro. This study uses time-lapse confocal fluorescence microscopy images from mouse ESCs carrying reporter gene to monitor dynamic changes in cellular/differentiation characteristics of live ESCs. To automate analyses of ESC differentiation behaviours, accurate cell nucleus segmentation to distinguish respective cells are required. A series of image preprocessing procedures are implemented to remove noise in live-cell fluorescence images but yield overlapping cell nuclei. A segmentation method that evaluates boundary concavities and the size and shape of every nucleus is then used to determine if any of the strait, extrusion, convexity, large and local minimum diameter criteria satisfied to segment overlapping nuclei. We propose a dynamic data structure to record every newly segmented nucleus. The experimental results show that the proposed image preprocessing method preserves the areas of mouse ESC nuclei and that the segmentation method effectively detects overlapping nuclei. All segmented nuclei in confocal images can be accessed using the dynamic data structures to be visualised and manipulated for quantitative analyses of the ESC differentiation behaviours. The manipulation can be tracking of segmented 3D cell nuclei in time-lapse images to calculate their dynamics of differentiation characteristics.

A segmentation method combining probability map and boundary based on multiple fully convolutional networks and repetitive training

Cell nuclei image segmentation technology can help researchers observe each cell's stress response to drug treatment. However, it is still a challenge to accurately segment the adherent cell nuclei. At present, image segmentation based on a fully convolutional network (FCN) is attracting researchers' attention. We propose a multiple FCN architecture and repetitive training (M-FCN-RT) method to learn features of cell nucleus images. In M-FCN-RT, the multiple FCN (M-FCN) architecture is composed of several single FCNs (S-FCNs) with the same structure, and each FCN is used to learn the specific features of image datasets. In this paper, the M-FCN contains three FCNs; FCN_1-2, FCN₃ and FCN_B. FCN_1-2 and FCN₃ are respectively used to learn the spatial features of cell nuclei for generating probability maps to indicate nucleus regions of an image; FCN_B (boundary FCN) is used to learn the edge features of cell nuclei for generating the nucleus boundary. For the training of each FCN, we propose a repetitive training (RT) method to improve the classification accuracy of the model. To segment cell nuclei, we finally propose an algorithm combining the probability map and boundary (PMB) to segment the adherent nuclei. This paper uses a public opening nucleus image dataset to train, verify and evaluate the proposed M-FCN-RT and PMB methods. Our M-FCN-RT method achieves a high Dice similarity coefficient (DSC) of 92.11%, 95.64% and 87.99% on the three types of sub-datasets respectively for probability maps. In addition, segmentation experimental results show the PMB method is more effective and efficient compared with other methods.

Contour-Seed Pairs Learning-Based Framework for Simultaneously Detecting and Segmenting Various Overlapping Cells/Nuclei in Microscopy Images

In this paper, we propose a novel contour-seed pairs learning-based framework for robust and automated cell/nucleus segmentation. Automated granular object segmentation in microscopy images has significant clinical importance for pathology grading of the cell carcinoma and gene expression. The focus of the past literature is dominated by either segmenting a certain type of cells/nuclei or simply splitting the clustered objects without contours inference of them. Our method addresses these issues by formulating the detection and segmentation tasks in terms of a unified regression problem, where a cascade sparse regression chain model is trained and then applied to return object locations and entire boundaries of clustered objects. In particular, we first learn a set of online convolutional features in each layer. Then, in the proposed cascade sparse regression chain, with the input from the learned features, we iteratively update the locations and clustered object boundaries until convergence. In this way, the boundary evidences of each individual object can be easily delineated and be further fed to a complete contour inference procedure optimized by the minimum description length principle. For any probe image, our method enables to analyze free-lying and overlapping cells with complex shapes. Experimental results show that the proposed method is very generic and performs well on contour inferences of various cell/nucleus types. Compared with the current segmentation techniques, our approach achieves state-of-the-art performances on four challenging datasets, i.e., the kidney renal cell carcinoma histopathology dataset, Drosophila Kc167 cellular dataset, differential interference contrast red blood cell dataset, and cervical cytology dataset.