Analyzes, processes and visualizes multi-dimensional microscopy images. BioImageXD puts open-source computer science tools for three-dimensional visualization and analysis into the hands of all researchers, through a user-friendly graphical interface tuned to the needs of biologists. BioImageXD has no restrictive licenses or undisclosed algorithms and enables publication of precise, reproducible and modifiable workflows. It allows simple construction of processing pipelines and should enable biologists to perform challenging analyses of complex processes.
Consists of a cloud based deep learning solution for image segmentation of light, electron and X-ray microscopy. CDeep3M serves for image segmentation tasks of large and complex 2D and 3D microscopy datasets by taking advantage of the underlying architecture of a deep learning convolutional neural network (CNN) called DeepEM3D. This software is also available on the Amazon Web Services (AWS) platform.
Describes the reconstruction of biological molecules from the electron micrographs of single particles. SPIDER is an open source project that manages computation using the image-processing software and can also using a graphical user interface (GUI), termed the SPIDER Reconstruction Engine. These two approaches are described to obtain an initial reconstruction: random-conical tilt and common lines.
Allows users to perform image analysis for high-resolution connectomics. SegEM is a program dedicated to reconstruct neuronal circuits. It supplies a classifier selection procedure permitting researchers to study different types of nerve tissue. This tool can solve the exchange between semi-automated reconstruction performance and synapse detection in high-resolution connectomics.
Provides state of the art automatic segmentation accuracy for image segmentation. GALA utilises machine learning to obtain a merge priority function or policy, which dictates which pair of segments to merge next. It is able to mask volumes so that partial ground truth can be used. The tool outperforms previous agglomeration methods for automatic segmentation of an isotropic focused ion beam scanning electron microscope (FIBSEM) dataset of Drosophila larva neuropil.
Finds and quantifies synapses in electron microscopy (EM) images. This program consists of a machine-learning framework that is able to realize prediction in a high-throughput and fully-automated manner. It can decrease the search space of possible synapses, thereby significantly reducing false positives with a first-pass filtration step. The approach employs texture- and shape-based features extracted from small image patches to identify synapses.
Represents a deep learning method for segmenting 3D anisotropic brain electron microscopy images. DeepEM3D can efficiently build feature representation and incorporate sufficient multi-scale contextual information. DeepEM3D is able to produce highly accurate 3D neurite boundary probability maps, thereby requiring only a simple watershed method to do segmentation. This tool uses the power of inception and residual structures in the bottom-up path to integrate image information, and combines skip connection techniques with pyramid multi-scale contexture aggregation in the top-down path.
A deep neural network model architecture that is highly optimized for serial-section transmitted electron microscopy image segmentation. We trained a pixel classifier that operates on raw pixel intensities with no preprocessing to generate probability values for each pixel being a membrane or not. While the use of neural networks in image segmentation is not completely new, we developed novel insights and model architectures that allow us to achieve superior performance on EM image segmentation tasks. Our submission based on these insights to the 2D EM Image Segmentation Challenge achieved the best performance consistently across all the three evaluation metrics.
Permits to reconstruct complete connectome. NeuroProof is based on a threshold k determined heuristically. It is amenable to large-scale, crowd-sourcing efforts. The tool was used in the domain of electron microscopic (EM) reconstruction and can be applied to other domains. It provides routines for efficient agglomeration for an initial volume that is over-segmented.
A three-dimensional (3D) reconstruction and modeling software. Free-D allows to generate, process and analyze 3D point and surface models from stacks of 2D images. It is an integrated software tool, offering in a single graphical user interface all the functionalities required for 3D modeling.
Allows users to segment cryo-electron microscopy (cryo-EM) density maps. Segger is a module, based on watershed and scale space filtering, that can be run through the Chimera software. The application employs segmented regions to enable rigid-body docking of known structures within density maps. It also interprets the quality and reliability of the docking results by calculating atom inclusion, density occupancy, and clashes with symmetry-related copies.
Serves for montaging, alignment, analysis and visualization of serial sections. Reconstruct allows users to organize, transform and display different types of data. It is able to analyze series with large number of sections and images over a large range of magnifications. Moreover, this tool includes functionalities that simplify cropping, scaling and comparison of images.
Captures the mass spectrometry (MS) lesion spatial distribution and identifies lesions regardless of their orientation, shape or size. RMNMS detects MS lesions using a training library containing T2-weighted (T2W) and FLAIR images along with manual T2W lesion masks. Moreover, it assists users in the detection of presence of lesions as lesion-wise measures.
Segments nearly elliptic objects via parametric active contour. E-Snake applies an active contour (named snake) segmentation method by using exponential splines as basis functions to represent the outline of the shape. This ImageJ plugin emulates elliptical and circular shapes and can produce an approximation of any closed curve in the plane. These proprieties can be useful to delineate cross sections of cylindrical-like conduits and to outline blob-like objects.
Allows users to segment cardiac ultrastructure from serial block-face scanning electron microscopy (SBF-SEM) data. Sbfsem-cardiac-call-segmenter is a software intended for performing the extraction and the analysis of ultrastructure content and organization without manual segmentation or prior knowledge of image processing techniques. In addition, the program was developed for being implemented in other software such as IMOD or Fiji.
Splits microscopy data dealing with axon and myelin. AxonDeepSeg provides a standalone software able to create models for new imaging procedures as well as to perform semantic segmentation of histological images. The application includes two models for both scanning electron microscopy (SEM) and transmission electron microscopy (TEM) samples segmentation. It can be applied to the study of the distribution and size of myelinated fibers in samples.
Determines stomatal density from epidermal micrographs. Stomata Counter consists of a network trained with a humanin-the-loop approach. It can be used to upload plant epidermal image datasets to pre-trained networks and make annotation of stomata on cuticle images. This tool builds a stomata likelihood map for each input image with a deep convolutional neural network (DCNN). It allows users to estimate the empirical error rate of the automatic counts.
A simple, user-friendly tool for interactive image classification, segmentation and analysis. It is built as a modular software framework, which currently has workflows for automated (supervised) pixel- and object-level classification, automated and semi-automated object tracking, semi-automated segmentation and object counting without detection. Most analysis operations are performed lazily, which enables targeted interactive processing of data subvolumes, followed by complete volume analysis in offline batch mode.
Allows manual annotation and segmentation of large volumetric (voxel) data sets. VAST is a software which enables to work with voxel data sets in the Terabyte or even Petabyte range at interactive speeds, to explore them visually and to label structures of interest by voxel painting. It can be used to generate volumetric training data sets, and can to some extent also be used for importing, proof-reading and correcting results of segmentation algorithms.
Assists in exploration of possible watershed segmentation. Interactive-H-Watershed is a plugin for the image analysis software ImageJ. It provides an interactive way to explore local minima (maxima) on the fly. This module is based on Watershed, a common tool to segment objects in an 2D and 3D images. The method permits to gradually flood the valleys starting from their lowest point.
Performs electron microscopy (EM) based segmentation of images in nifti or analyse format. NiftySeg contains a package of label fusion algorithms (MV, STAPLE, SBA) with different types of ranking strategies.
Performs image segmentation and registration. CMP-BIA is dedicated to ImageJ/Fiji and contains a plugin named jSLIC. This plugin is a segmentation method for clustering, in a given image, similar regions (as known as superpixels) which are usually used for other segmentation techniques, classification and registration. It gives reliable superpixels shapes, with no need of decreasing their size.
Allows users to visualize, manipulate, and understand data from imaging modalities such as computed tomography, microscopy or Magnetic resonance imaging (MRI). Amira 3D Software for Life Sciences provides features to import and process 2D and 3D images data, visualization techniques and tools for visual analysis. Users can also create and share presentations. The base product can be customized by adding functional extensions to fit special needs in different application areas.
Focuses on combinatorial optimization via graph cuts for digital image analysis. GC is a library aiming to finds solution to energy minimization based discrete labeling problems such as image segmentation. The software provides access to several algorithms, such as maximum flow algorithms, Metric approximation, Multi-label discrete energy optimization, and Image segmentation.
0 - 0 of 0