Machine learning applied to multi-sensor information to reduce false alarm rate in the ICU.

Studies reveal that the false alarm rate (FAR) demonstrated by intensive care unit (ICU) vital signs monitors ranges from 0.72 to 0.99. We applied machine learning (ML) to ICU multi-sensor information to imitate a medical specialist in diagnosing patient condition. We hypothesized that applying this data-driven approach to medical monitors will help reduce the FAR even when data from sensors are missing. An expert-based rules algorithm identified and tagged in our dataset seven clinical alarm scenarios. We compared a random forest (RF) ML model trained using the tagged data, where parameters (e.g., heart rate or blood pressure) were (deliberately) removed, in detecting ICU signals with the full expert-based rules (FER), our ground truth, and partial expert-based rules (PER), missing these parameters. When all alarm scenarios were examined, RF and FER were almost identical. However, in the absence of one to three parameters, RF maintained its values of the Youden index (0.94-0.97) and positive predictive value (PPV) (0.98-0.99), whereas PER lost its value (0.54-0.8 and 0.76-0.88, respectively). While the FAR for PER with missing parameters was 0.17-0.39, it was only 0.01-0.02 for RF. When scenarios were examined separately, RF showed clear superiority in almost all combinations of scenarios and numbers of missing parameters. When sensor data are missing, specialist performance worsens with the number of missing parameters, whereas the RF model attains high accuracy and low FAR due to its ability to fuse information from available sensors, compensating for missing parameters.

Deep convolutional networks do not classify based on global object shape.

Deep convolutional networks (DCNNs) are achieving previously unseen performance in object classification, raising questions about whether DCNNs operate similarly to human vision. In biological vision, shape is arguably the most important cue for recognition. We tested the role of shape information in DCNNs trained to recognize objects. In Experiment 1, we presented a trained DCNN with object silhouettes that preserved overall shape but were filled with surface texture taken from other objects. Shape cues appeared to play some role in the classification of artifacts, but little or none for animals. In Experiments 2-4, DCNNs showed no ability to classify glass figurines or outlines but correctly classified some silhouettes. Aspects of these results led us to hypothesize that DCNNs do not distinguish object's bounding contours from other edges, and that DCNNs access some local shape features, but not global shape. In Experiment 5, we tested this hypothesis with displays that preserved local features but disrupted global shape, and vice versa. With disrupted global shape, which reduced human accuracy to 28%, DCNNs gave the same classification labels as with ordinary shapes. Conversely, local contour changes eliminated accurate DCNN classification but caused no difficulty for human observers. These results provide evidence that DCNNs have access to some local shape information in the form of local edge relations, but they have no access to global object shapes.

Influence of Curvature, Growth, and Anisotropy on the Evolution of Turing Patterns on Growing Manifolds.

We study two-species reaction-diffusion systems on growing manifolds, including situations where the growth is anisotropic yet dilational in nature. In contrast to the literature on linear instabilities in such systems, we study how growth and anisotropy impact the qualitative properties of nonlinear patterned states which have formed before growth is initiated. We produce numerical solutions to numerous reaction-diffusion systems with varying reaction kinetics, manner of growth (both isotropic and anisotropic), and timescales of growth on both planar elliptical and curved ellipsoidal domains. We find that in some parameter regimes, some of these factors have a negligible effect on the long-time patterned state. On the other hand, we find that some of these factors play a role in determining the patterns formed on surfaces and that anisotropic growth can produce qualitatively different patterns to those formed under isotropic growth.

DAFi: A directed recursive data filtering and clustering approach for improving and interpreting data clustering identification of cell populations from polychromatic flow cytometry data.

Computational methods for identification of cell populations from polychromatic flow cytometry data are changing the paradigm of cytometry bioinformatics. Data clustering is the most common computational approach to unsupervised identification of cell populations from multidimensional cytometry data. However, interpretation of the identified data clusters is labor-intensive. Certain types of user-defined cell populations are also difficult to identify by fully automated data clustering analysis. Both are roadblocks before a cytometry lab can adopt the data clustering approach for cell population identification in routine use. We found that combining recursive data filtering and clustering with constraints converted from the user manual gating strategy can effectively address these two issues. We named this new approach DAFi: Directed Automated Filtering and Identification of cell populations. Design of DAFi preserves the data-driven characteristics of unsupervised clustering for identifying novel cell subsets, but also makes the results interpretable to experimental scientists through mapping and merging the multidimensional data clusters into the user-defined two-dimensional gating hierarchy. The recursive data filtering process in DAFi helped identify small data clusters which are otherwise difficult to resolve by a single run of the data clustering method due to the statistical interference of the irrelevant major clusters. Our experiment results showed that the proportions of the cell populations identified by DAFi, while being consistent with those by expert centralized manual gating, have smaller technical variances across samples than those from individual manual gating analysis and the nonrecursive data clustering analysis. Compared with manual gating segregation, DAFi-identified cell populations avoided the abrupt cut-offs on the boundaries. DAFi has been implemented to be used with multiple data clustering methods including K-means, FLOCK, FlowSOM, and the ClusterR package. For cell population identification, DAFi supports multiple options including clustering, bisecting, slope-based gating, and reversed filtering to meet various autogating needs from different scientific use cases. © 2018 International Society for Advancement of Cytometry.

Multiple objects detection in biological images using a marked point process framework.

The marked point process framework has been successfully developed in the field of image analysis to detect a configuration of predefined objects. The goal of this paper is to show how it can be particularly applied to biological imagery. We present a simple model that shows how some of the challenges specific to biological data are well addressed by the methodology. We further describe an extension to this first model to address other challenges due, for example, to the shape variability in biological material. We finally show results that illustrate the MPP framework using the "simcep" algorithm for simulating populations of cells.

TED: A Tolerant Edit Distance for segmentation evaluation.

In this paper, we present a novel error measure to compare a computer-generated segmentation of images or volumes against ground truth. This measure, which we call Tolerant Edit Distance (TED), is motivated by two observations that we usually encounter in biomedical image processing: (1) Some errors, like small boundary shifts, are tolerable in practice. Which errors are tolerable is application dependent and should be explicitly expressible in the measure. (2) Non-tolerable errors have to be corrected manually. The effort needed to do so should be reflected by the error measure. Our measure is the minimal weighted sum of split and merge operations to apply to one segmentation such that it resembles another segmentation within specified tolerance bounds. This is in contrast to other commonly used measures like Rand index or variation of information, which integrate small, but tolerable, differences. Additionally, the TED provides intuitive numbers and allows the localization and classification of errors in images or volumes. We demonstrate the applicability of the TED on 3D segmentations of neurons in electron microscopy images where topological correctness is arguable more important than exact boundary locations. Furthermore, we show that the TED is not just limited to evaluation tasks. We use it as the loss function in a max-margin learning framework to find parameters of an automatic neuron segmentation algorithm. We show that training to minimize the TED, i.e., to minimize crucial errors, leads to higher segmentation accuracy compared to other learning methods.

An Overview of data science uses in bioimage informatics.

This review aims at providing a practical overview of the use of statistical features and associated data science methods in bioimage informatics. To achieve a quantitative link between images and biological concepts, one typically replaces an object coming from an image (a segmented cell or intracellular object, a pattern of expression or localisation, even a whole image) by a vector of numbers. They range from carefully crafted biologically relevant measurements to features learnt through deep neural networks. This replacement allows for the use of practical algorithms for visualisation, comparison and inference, such as the ones from machine learning or multivariate statistics. While originating mainly, for biology, in high content screening, those methods are integral to the use of data science for the quantitative analysis of microscopy images to gain biological insight, and they are sure to gather more interest as the need to make sense of the increasing amount of acquired imaging data grows more pressing.

Ultrasound registration: A review.

This article is a review of registration algorithms for use between ultrasound images (monomodal image-based ultrasound registration). Ultrasound is safe, inexpensive, and real-time, providing many advantages for clinical and scientific use on both humans and animals, but ultrasound images are also notoriously noisy and subject to several unique artifacts/distortions. This paper introduces the topic and unique aspects of ultrasound-to-ultrasound image registration, providing a broad introduction and summary of the literature and the field. Both theoretical and practical aspects are introduced. The first half of the paper is theoretical, organized according to the basic components of a registration framework, namely preprocessing, image-similarity metrics, optimizers, etc. It further subdivides these methods between those suitable for elastic (non-rigid) vs. inelastic (matrix) transforms. The second half of the paper is organized by anatomy and is practical in nature, presenting and discussing the complete published systems that have been validated for registration in specific anatomic regions.

Analysis of a model for banded vegetation patterns in semi-arid environments with nonlocal dispersal.

Vegetation patterns are a characteristic feature of semi-arid regions. On hillsides these patterns occur as stripes running parallel to the contours. The Klausmeier model, a coupled reaction-advection-diffusion system, is a deliberately simple model describing the phenomenon. In this paper, we replace the diffusion term describing plant dispersal by a more realistic nonlocal convolution integral to account for the possibility of long-range dispersal of seeds. Our analysis focuses on the rainfall level at which there is a transition between uniform vegetation and pattern formation. We obtain results, valid to leading order in the large parameter comparing the rate of water flow downhill to the rate of plant dispersal, for a negative exponential dispersal kernel. Our results indicate that both a wider dispersal of seeds and an increase in dispersal rate inhibit the formation of patterns. Assuming an evolutionary trade-off between these two quantities, mathematically motivated by the limiting behaviour of the convolution term, allows us to make comparisons to existing results for the original reaction-advection-diffusion system. These comparisons show that the nonlocal model always predicts a larger parameter region supporting pattern formation. We then numerically extend the results to other dispersal kernels, showing that the tendency to form patterns depends on the type of decay of the kernel.

A primer to frequent itemset mining for bioinformatics.

Over the past two decades, pattern mining techniques have become an integral part of many bioinformatics solutions. Frequent itemset mining is a popular group of pattern mining techniques designed to identify elements that frequently co-occur. An archetypical example is the identification of products that often end up together in the same shopping basket in supermarket transactions. A number of algorithms have been developed to address variations of this computationally non-trivial problem. Frequent itemset mining techniques are able to efficiently capture the characteristics of (complex) data and succinctly summarize it. Owing to these and other interesting properties, these techniques have proven their value in biological data analysis. Nevertheless, information about the bioinformatics applications of these techniques remains scattered. In this primer, we introduce frequent itemset mining and their derived association rules for life scientists. We give an overview of various algorithms, and illustrate how they can be used in several real-life bioinformatics application domains. We end with a discussion of the future potential and open challenges for frequent itemset mining in the life sciences.

Graphical aids for visualizing and interpreting patterns in departures from agreement in ordinal categorical observer agreement data.

When studying the agreement between two observers rating the same n units into the same k discrete ordinal categories, Bangdiwala (1985) proposed using the "agreement chart" to visually assess agreement. This article proposes that often it is more interesting to focus on the patterns of disagreement and visually understanding the departures from perfect agreement. The article reviews the use of graphical techniques for descriptively assessing agreement and disagreements, and also reviews some of the available summary statistics that quantify such relationships.

Evolutionary Visual Exploration: Evaluation of an IEC Framework for Guided Visual Search.

We evaluate and analyse a framework for evolutionary visual exploration (EVE) that guides users in exploring large search spaces. EVE uses an interactive evolutionary algorithm to steer the exploration of multidimensional data sets toward two-dimensional projections that are interesting to the analyst. Our method smoothly combines automatically calculated metrics and user input in order to propose pertinent views to the user. In this article, we revisit this framework and a prototype application that was developed as a demonstrator, and summarise our previous study with domain experts and its main findings. We then report on results from a new user study with a clearly predefined task, which examines how users leverage the system and how the system evolves to match their needs. While we previously showed that using EVE, domain experts were able to formulate interesting hypotheses and reach new insights when exploring freely, our new findings indicate that users, guided by the interactive evolutionary algorithm, are able to converge quickly to an interesting view of their data when a clear task is specified. We provide a detailed analysis of how users interact with an evolutionary algorithm and how the system responds to their exploration strategies and evaluation patterns. Our work aims at building a bridge between the domains of visual analytics and interactive evolution. The benefits are numerous, in particular for evaluating interactive evolutionary computation (IEC) techniques based on user study methodologies.

Spatially regularized mixture model for lesion segmentation with application to stroke patients.

In medical imaging, lesion segmentation (differentiation between lesioned and non-lesioned tissue) is a crucial and difficult task. Automated segmentation algorithms based on intensity analysis have been already proposed and recent developments have shown that integrating spatial information enhances automatic image segmentation. However, spatial modeling is often limited to short-range spatial interactions that deal only with noise or small artifacts. Previous tissue alterations (e.g. white matter disease (WMD)) similar in intensity with the lesion of interest require a broader-scale approach to be corrected. On the other hand, imaging techniques offer now a multiparametric voxel characterization that may help differentiating lesioned from non-lesioned voxels. We developed an unsupervised multivariate segmentation algorithm based on finite mixture modeling that incorporates spatial information. We extended the usual spatial Potts model to the regional scale using a 'multi-order' neighborhood potential, with internal adjustment of the regional scale according to the lesion size. We validate the ability of this new algorithm to deal with noise and artifacts (linear and spherical) using artificial data. We then assess its performance on real magnetic resonance imaging brain scans of stroke patients with history of WMD and show that regional regularization was able to remove large-scale WMD artifacts.

A bioinformatician's guide to the forefront of suffix array construction algorithms.

The suffix array and its variants are text-indexing data structures that have become indispensable in the field of bioinformatics. With the uninitiated in mind, we provide an accessible exposition of the SA-IS algorithm, which is the state of the art in suffix array construction. We also describe DisLex, a technique that allows standard suffix array construction algorithms to create modified suffix arrays designed to enable a simple form of inexact matching needed to support 'spaced seeds' and 'subset seeds' used in many biological applications.

Transforms and Operators for Directional Bioimage Analysis: A Survey.

We give a methodology-oriented perspective on directional image analysis and rotation-invariant processing. We review the state of the art in the field and make connections with recent mathematical developments in functional analysis and wavelet theory. We unify our perspective within a common framework using operators. The intent is to provide image-processing methods that can be deployed in algorithms that analyze biomedical images with improved rotation invariance and high directional sensitivity. We start our survey with classical methods such as directional-gradient and the structure tensor. Then, we discuss how these methods can be improved with respect to robustness, invariance to geometric transformations (with a particular interest in scaling), and computation cost. To address robustness against noise, we move forward to higher degrees of directional selectivity and discuss Hessian-based detection schemes. To present multiscale approaches, we explain the differences between Fourier filters, directional wavelets, curvelets, and shearlets. To reduce the computational cost, we address the problem of matching directional patterns by proposing steerable filters, where one might perform arbitrary rotations and optimizations without discretizing the orientation. We define the property of steerability and give an introduction to the design of steerable filters. We cover the spectrum from simple steerable filters through pyramid schemes up to steerable wavelets. We also present illustrations on the design of steerable wavelets and their application to pattern recognition.

Segmenting and Tracking Multiple Dividing Targets Using ilastik.

Tracking crowded cells or other targets in biology is often a challenging task due to poor signal-to-noise ratio, mutual occlusion, large displacements, little discernibility, and the ability of cells to divide. We here present an open source implementation of conservation tracking (Schiegg et al., IEEE international conference on computer vision (ICCV). IEEE, New York, pp 2928-2935, 2013) in the ilastik software framework. This robust tracking-by-assignment algorithm explicitly makes allowance for false positive detections, undersegmentation, and cell division. We give an overview over the underlying algorithm and parameters, and explain the use for a light sheet microscopy sequence of a Drosophila embryo. Equipped with this knowledge, users will be able to track targets of interest in their own data.

Challenges and Benchmarks in Bioimage Analysis.

Similar to the medical imaging community, the bioimaging community has recently realized the need to benchmark various image analysis methods to compare their performance and assess their suitability for specific applications. Challenges sponsored by prestigious conferences have proven to be an effective means of encouraging benchmarking and new algorithm development for a particular type of image data. Bioimage analysis challenges have recently complemented medical image analysis challenges, especially in the case of the International Symposium on Biomedical Imaging (ISBI). This review summarizes recent progress in this respect and describes the general process of designing a bioimage analysis benchmark or challenge, including the proper selection of datasets and evaluation metrics. It also presents examples of specific target applications and biological research tasks that have benefited from these challenges with respect to the performance of automatic image analysis methods that are crucial for the given task. Finally, available benchmarks and challenges in terms of common features, possible classification and implications drawn from the results are analysed.

Bioimage Informatics for Big Data.

Bioimage informatics is a field wherein high-throughput image informatics methods are used to solve challenging scientific problems related to biology and medicine. When the image datasets become larger and more complicated, many conventional image analysis approaches are no longer applicable. Here, we discuss two critical challenges of large-scale bioimage informatics applications, namely, data accessibility and adaptive data analysis. We highlight case studies to show that these challenges can be tackled based on distributed image computing as well as machine learning of image examples in a multidimensional environment.

Automated Robust Image Segmentation: Level Set Method Using Nonnegative Matrix Factorization with Application to Brain MRI.

We address the problem of fully automated region discovery and robust image segmentation by devising a new deformable model based on the level set method (LSM) and the probabilistic nonnegative matrix factorization (NMF). We describe the use of NMF to calculate the number of distinct regions in the image and to derive the local distribution of the regions, which is incorporated into the energy functional of the LSM. The results demonstrate that our NMF-LSM method is superior to other approaches when applied to synthetic binary and gray-scale images and to clinical magnetic resonance images (MRI) of the human brain with and without a malignant brain tumor, glioblastoma multiforme. In particular, the NMF-LSM method is fully automated, highly accurate, less sensitive to the initial selection of the contour(s) or initial conditions, more robust to noise and model parameters, and able to detect as small distinct regions as desired. These advantages stem from the fact that the proposed method relies on histogram information instead of intensity values and does not introduce nuisance model parameters. These properties provide a general approach for automated robust region discovery and segmentation in heterogeneous images. Compared with the retrospective radiological diagnoses of two patients with non-enhancing grade 2 and 3 oligodendroglioma, the NMF-LSM detects earlier progression times and appears suitable for monitoring tumor response. The NMF-LSM method fills an important need of automated segmentation of clinical MRI.

MixGF: spectral probabilities for mixture spectra from more than one peptide.

In large-scale proteomic experiments, multiple peptide precursors are often cofragmented simultaneously in the same mixture tandem mass (MS/MS) spectrum. These spectra tend to elude current computational tools because of the ubiquitous assumption that each spectrum is generated from only one peptide. Therefore, tools that consider multiple peptide matches to each MS/MS spectrum can potentially improve the relatively low spectrum identification rate often observed in proteomics experiments. More importantly, data independent acquisition protocols promoting the cofragmentation of multiple precursors are emerging as alternative methods that can greatly improve the throughput of peptide identifications but their success also depends on the availability of algorithms to identify multiple peptides from each MS/MS spectrum. Here we address a fundamental question in the identification of mixture MS/MS spectra: determining the statistical significance of multiple peptides matched to a given MS/MS spectrum. We propose the MixGF generating function model to rigorously compute the statistical significance of peptide identifications for mixture spectra and show that this approach improves the sensitivity of current mixture spectra database search tools by a ≈30-390%. Analysis of multiple data sets with MixGF reveals that in complex biological samples the number of identified mixture spectra can be as high as 20% of all the identified spectra and the number of unique peptides identified only in mixture spectra can be up to 35.4% of those identified in single-peptide spectra.