Mylene Simon

Survey statistics of automated segmentations applied to optical imaging of mammalian cells

BackgroundThe goal of this survey paper is to overview cellular measurements using optical microscopy imaging followed by automated image segmentation. The cellular measurements of primary interest are taken from mammalian cells and their components. They are denoted as two- or three-dimensional (2D or 3D) image objects of biological interest. In our applications, such cellular measurements are important for understanding cell phenomena, such as cell counts, cell-scaffold interactions, cell colony growth rates, or cell pluripotency stability, as well as for establishing quality metrics for stem cell therapies. In this context, this survey paper is focused on automated segmentation as a software-based measurement leading to quantitative cellular measurements.MethodsWe define the scope of this survey and a classification schema first. Next, all found and manually filteredpublications are classified according to the main categories: (1) objects of interests (or objects to be segmented), (2) imaging modalities, (3) digital data axes, (4) segmentation algorithms, (5) segmentation evaluations, (6) computational hardware platforms used for segmentation acceleration, and (7) object (cellular) measurements. Finally, all classified papers are converted programmatically into a set of hyperlinked web pages with occurrence and co-occurrence statistics of assigned categories.ResultsThe survey paper presents to a reader: (a) the state-of-the-art overview of published papers about automated segmentation applied to optical microscopy imaging of mammalian cells, (b) a classification of segmentation aspects in the context of cell optical imaging, (c) histogram and co-occurrence summary statistics about cellular measurements, segmentations, segmented objects, segmentation evaluations, and the use of computational platforms for accelerating segmentation execution, and (d) open research problems to pursue.ConclusionsThe novel contributions of this survey paper are: (1) a new type of classification of cellular measurements and automated segmentation, (2) statistics about the published literature, and (3) a web hyperlinked interface to classification statistics of the surveyed papers at https://isg.nist.gov/deepzoomweb/resources/survey/index.html.

A method for the evaluation of thousands of automated 3D stem cell segmentations

There is no segmentation method that performs perfectly with any dataset in comparison to human segmentation. Evaluation procedures for segmentation algorithms become critical for their selection. The problems associated with segmentation performance evaluations and visual verification of segmentation results are exaggerated when dealing with thousands of three‐dimensional (3D) image volumes because of the amount of computation and manual inputs needed.

Interoperability Between Software and Hardware

While WIPP has been designed for big image data experiments, its execution speed depends on the underlying hardware, which can vary with each deployment. Readers may be interested in optimizing WIPP for their hardware or in rewriting existing algorithms to better utilize available RAM, CPU, and bandwidth. In this chapter, our goal is to assist the reader in: Choosing hardware for WIPP deployment Characterizing image data to estimate storage and processing requirements Measuring execution time Leveraging several known models for parallel execution of image processing algorithms

Functionality of Web Image Processing Pipeline

This chapter presents functionalities of the web image processing pipeline (WIPP). It guides a reader through deployment of WIPP and functionalities of three WIPP modules, such as image processing, image feature extraction, and statistical modeling modules. By providing functionality exercises, a reader can become familiar with the three modules of WIPP.

Components of Web Image Processing Pipeline

This chapter is for advanced readers who would like to learn about the building blocks of the web image processing pipeline (WIPP). By overviewing a WIPP generic usage scenario, we first map the WIPP functionality to information technologies that form a set of building blocks for client-server applications. Next, we describe the information technologies at a high level so that a reader gains some general understanding about designing web image processing pipelines.

Introduction to Big Data Microscopy Experiments

This chapter introduces the key topics of this book, such as web image processing pipeline and big data experiments, and then presents topics of interest to the multiple disciplines of big image data. We also highlight challenges of big data experiments and the role of a web image processing pipeline in the quest for reproducible science.

Example Use Cases

This chapter describes three use cases that leverage the functionalities in WIPP. The use cases focus on the following problems: 1. Cell count and single cell detection 2. Stem cell colony growth computation 3. Feature variability analysis

Image Processing Algorithms

In this chapter, we will focus on image processing algorithms implemented in WIPP. These algorithms include image correction, stitching, segmentation, tracking, feature extraction, intensity scaling, and image pyramid building. We will provide a high-level overview of each algorithm and its relevance to microscopy image processing. The purpose of the algorithmic overview is to make the reader aware of the assumptions and trade-offs embedded in each algorithm implementation. For more in-depth knowledge about algorithms, we will refer the reader to a collection of image processing books and journal papers that could expand reader’s knowledge beyond reading the material presented in this chapter.

Modeling, validation and verification of cell-scaffold contact measurements over terabyte-sized 3D image collection

This poster presents the problem of 3D contact measurements from two co-registered volumetric images (z-stacks). The 3D contact measurement consists of (a) segmenting an object of interest in each z-stack, (b) computing the relative spatial positions of the detected objects to detect contacts, (c) validating the accuracy of segmentation, and (d) visually verifying correct contact detection. The 3D measurement has to overcome challenges related to (1) intensity bleed-through across co-registered volumes, (2) insufficient knowledge about statistics and geometry of objects, (3) large RAM requirements (∼3GB just to load the input data) and data volume (>1TB), and (4) complexity of 3D visual inspection.

Do we trust image measurements? Variability, accuracy and traceability of image features

The paper addresses the problem of understanding quality of image measurements extracted using widely used software libraries from large images. Image measurements (features) are extracted using software packages that vary in terms of programming languages, theoretical formulas for the same image feature, algorithmic implementations, input parameters, units of measurements, and definitions of image regions of interest. Our motivation is to quantify numerical variability of image features across software packages and determine image accuracy with respect to reference images. In addition, our objective is to enable scientists to extract any image features of interest from heterogeneous software libraries and gain provenance of every extracted numerical feature value. The provenance information is critical to achieve traceability of computations in terascale imaging. We pursue this objective by designing a client-server system that integrates image feature extractions from open source libraries such as ImageJ/Fiji, Python (scikit-image), CellProfiler, and in-house Java software packages. The system becomes useful for evaluating quality of image measurements, leveraging distributed computational resources for feature computations over big image data, sharing resulting feature values, and reproducing the feature values based on provenance. As an application of the designed system, we report the quality evaluations of 319 image features extracted using ImageJ/Fiji, Python (scikit-image), CellProfiler and in-house Java software packages with 43 duplicate features across the four packages. Using the normalized difference as metric, we identified 6 out of the 43 common features to differ over 1% in value and discuss the sources of these numerical differences.