A semi-localized elastic net for surface reconstruction of objects from multislice images

The traveling salesman problem (TSP) is a prototypical problem of combinatorial optimization and, as such, it has received considerable attention from neural-network researchers seeking quick, heuristic solutions. An early stage in many computer vision tasks is the extraction of object shape from an image consisting of noisy candidate edge points. Since the desired shape will often be a closed contour, this problem can be viewed as a version of the TSP in which we wish to link only a subset of the points/cities (i.e. the "noisefree" ones). None of the extant neural techniques for solving the TSP can deal directly with this case. In this paper, we present a simple but effective modification to the (analog) elastic net of Durbin and Willshaw which shifts emphasis from global to local behavior during convergence, so allowing the net to ignore some image points. Unlike the original elastic net, this semi-localized version is shown to tolerate considerable amounts of noise. As an example practical application, we describe the extraction of "pseudo-3D" human lung outlines from multiple preprocessed magnetic resonance images of the torso. An effectiveness measure (ideally zero) quantifies the difference between the extracted shape and some idealized shape exemplar. Our method produces average effectiveness scores of 0.06 for lung shapes extracted from initial semi-automatic segmentations which define the noisefree case. This deteriorates to 0.1 when extraction is from a noisy edge-point image obtained fully-automatically using a feedforward neural network.     

Effect of segmentation errors on 3D-to-2D registration of implant models in X-ray images

In many biomedical applications, it is desirable to estimate the three-dimensional (3D) position and orientation (pose) of a metallic rigid object (such as a knee or hip implant) from its projection in a two-dimensional (2D) X-ray image. If the geometry of the object is known, as well as the details of the image formation process, then the pose of the object with respect to the sensor can be determined. A common method for 3D-to-2D registration is to first segment the silhouette contour from the X-ray image; that is, identify all points in the image that belong to the 2D silhouette and not to the background. This segmentation step is then followed by a search for the 3D pose that will best match the observed contour with a predicted contour. Although the silhouette of a metallic object is often clearly visible in an X-ray image, adjacent tissue and occlusions can make the exact location of the silhouette contour difficult to determine in places. Occlusion can occur when another object (such as another implant component) partially blocks the view of the object of interest. In this paper, we argue that common methods for segmentation can produce errors in the location of the 2D contour, and hence errors in the resulting 3D estimate of the pose. We show, on a typical fluoroscopy image of a knee implant component, that interactive and automatic methods for segmentation result in segmented contours that vary significantly. We show how the variability in the 2D contours (quantified by two different metrics) corresponds to variability in the 3D poses. Finally, we illustrate how traditional segmentation methods can fail completely in the (not uncommon) cases of images with occlusion.     

Learning paradigms for image interpretation

In this paper we discuss image understanding and object recognition as a class of processes which involve binding what is seen with what is known. It follows from this perspective that it is important to explicate how systems may learn about spatial information from images, how it is encoded, and, how, in turn, such knowledge is matched with new image data. As well as discussing very specific solutions to these problems, this paper questions the view that our scientific understanding of image understanding is sufficiently circumscribed by the study of how image features are extracted and matched, per se.     

Tracking facilitates 3-D motion estimation

The recently emerging paradigm of Active Vision advocates studying visual problems in form of modules that are directly related to a visual task for observers that are active. Along these lines, we are arguing that in many cases when an object is moving in an unrestricted manner (translation and rotation) in the 3D world, we are just interested in the motion's translational components. For a monocular observer, using only the normal flow — the spatio-temporal derivatives of the image intensity function — we solve the problem of computing the direction of translation and the time to collision. We do not use optical flow since its computation is an ill-posed problem, and in the general case it is not the same as the motion field — the projection of 3D motion on the image plane. The basic idea of our motion parameter estimation strategy lies in the employment of fixation and tracking. Fixation simplifies much of the computation by placing the object at the center of the visual field, and the main advantage of tracking is the accumulation of information over time. We show how tracking is accomplished using normal flow measurements and use it for two different tasks in the solution process. First it serves as a tool to compensate for the lack of existence of an optical flow field and thus to estimate the translation parallel to the image plane; and second it gathers information about the motion component perpendicular to the image plane.     

Finding paths in a labyrinth based on reaction-diffusion media.

During the past few decades, many proposals were made on how to take an effective solution for finding a path in a labyrinth, one of the most well known problems of high computational complexity inherent in information processing by biomolecular and biological entities. In particular, attempts were made to use a technique attractive enough for solving this problem based on wave processes in reaction-diffusion media. Trigger waves in reaction diffusion systems spread simultaneously through all paths of the labyrinth in a highly parallel mode. Regretfully, the velocity of these waves is extremely low which gave no way for the practical implementation of this technique until now. An effective 'hardware' system was designed which was capable of finding a path in a labyrinth using fast phase waves. Three principal points were assumed as a basis for this design, i.e. (1) hybrid architecture that combined an information processing reaction-diffusion medium which performs operations of high computational complexity with a digital computer carrying out supplementary image processing operations; (2) light-sensitive information processing media of Belousov-Zhabotinsky type that enables the simulation of the labyrinth and spreading wave evolution by their images stored in the medium and reduces the problem to the image processing operations; (3) fast light-induced phase wave processes that spreads through the labyrinth in several seconds instead of hours which is typical for trigger waves inherent in reaction-diffusion media. These fundamentals along with the additional procedure of testing for labyrinth fragment connectness provided us with the opportunity to solve labyrinth problems.     

Convergence Time towards Periodic Orbits in Discrete Dynamical Systems

We investigate the convergence towards periodic orbits in discrete dynamical systems. We examine the probability that a randomly chosen point converges to a particular neighborhood of a periodic orbit in a fixed number of iterations, and we use linearized equations to examine the evolution near that neighborhood. The underlying idea is that points of stable periodic orbit are associated with intervals. We state and prove a theorem that details what regions of phase space are mapped into these intervals (once they are known) and how many iterations are required to get there. We also construct algorithms that allow our theoretical results to be implemented successfully in practice.     

The singularities of the visual mapping

In this article we treat purely metrical properties of the visual image, e.g. the time changes of the relative positions and orientations of image details. Self-induced movements of an observer relative to rigid bodies in his environment generate charactertistic motion parallax fields. The observer may regard those fields as proprioceptive and interprete the geometrical invariants of the fields as indicators of solid shape. In this way his perceptions become object-oriented, which is the normal case as the many constancy-phenomena show. Similar arguments apply to the disparity field of binocular vision. In this paper we treat the qualitative nature of such fields. [In this case the qualitative nature is basic. Compare the case of an equation with a single unknown. Often one is interested primarily in the qualitative solution (are there roots? How many?), and only slightly in the quantitative information (the numerical value of a root).] The qualitative nature of the fields is fixed if their singularities are known. It is shown that the singularities are of two types: isolated points (so-called specular points) and line-singularities (so-called folds, cusps and T-junctions). It is shown that for most vantage points that an observer can occupy, the topological structure of the set of singularities does not change if the observer performs small exploratory movements. That is most vantage points are stable. At an unstable vantage point the set of singularities changes and the observer experiences an event. Because certain properties of the set of singularities are shown to be preserved, only a few simple types of event are possible. A complete list is presented. The occurrence of an event is shown to be simply related to the solid shape of the objects of vision. Our geometrical theory enables us to understand the structure of the observer's internal models of external bodies.     

Image Registration Algorithm Using Mexican Hat Function-Based Operator and Grouped Feature Matching Strategy

Feature detection and matching are crucial for robust and reliable image registration. Although many methods have been developed, they commonly focus on only one class of image features. The methods that combine two or more classes of features are still novel and significant. In this work, methods for feature detection and matching are proposed. A Mexican hat function-based operator is used for image feature detection, including the local area detection and the feature point detection. For the local area detection, we use the Mexican hat operator for image filtering, and then the zero-crossing points are extracted and merged into the area borders. For the feature point detection, the Mexican hat operator is performed in scale space to get the key points. After the feature detection, an image registration is achieved by using the two classes of image features. The feature points are grouped according to a standardized region that contains correspondence to the local area, precise registration is achieved eventually by the grouped points. An image transformation matrix is estimated by the feature points in a region and then the best one is chosen through competition of a set of the transformation matrices. This strategy has been named the Grouped Sample Consensus (GCS). The GCS has also ability for removing the outliers effectively. The experimental results show that the proposed algorithm has high registration accuracy and small computational volume.     

Localizing Genes to Cerebellar Layers by Classifying ISH Images

Gene expression controls how the brain develops and functions. Understanding control processes in the brain is particularly hard since they involve numerous types of neurons and glia, and very little is known about which genes are expressed in which cells and brain layers. Here we describe an approach to detect genes whose expression is primarily localized to a specific brain layer and apply it to the mouse cerebellum. We learn typical spatial patterns of expression from a few markers that are known to be localized to specific layers, and use these patterns to predict localization for new genes. We analyze images of in-situ hybridization (ISH) experiments, which we represent using histograms of local binary patterns (LBP) and train image classifiers and gene classifiers for four layers of the cerebellum: the Purkinje, granular, molecular and white matter layer. On held-out data, the layer classifiers achieve accuracy above 94% (AUC) by representing each image at multiple scales and by combining multiple image scores into a single gene-level decision. When applied to the full mouse genome, the classifiers predict specific layer localization for hundreds of new genes in the Purkinje and granular layers. Many genes localized to the Purkinje layer are likely to be expressed in astrocytes, and many others are involved in lipid metabolism, possibly due to the unusual size of Purkinje cells.     

Visuomotor transformation in the fly gaze stabilization system

For sensory signals to control an animal's behavior, they must first be transformed into a format appropriate for use by its motor systems. This fundamental problem is faced by all animals, including humans. Beyond simple reflexes, little is known about how such sensorimotor transformations take place. Here we describe how the outputs of a well-characterized population of fly visual interneurons, lobula plate tangential cells (LPTCs), are used by the animal's gaze-stabilizing neck motor system. The LPTCs respond to visual input arising from both self-rotations and translations of the fly. The neck motor system however is involved in gaze stabilization and thus mainly controls compensatory head rotations. We investigated how the neck motor system is able to selectively extract rotation information from the mixed responses of the LPTCs. We recorded extracellularly from fly neck motor neurons (NMNs) and mapped the directional preferences across their extended visual receptive fields. Our results suggest that-like the tangential cells-NMNs are tuned to panoramic retinal image shifts, or optic flow fields, which occur when the fly rotates about particular body axes. In many cases, tangential cells and motor neurons appear to be tuned to similar axes of rotation, resulting in a correlation between the coordinate systems the two neural populations employ. However, in contrast to the primarily monocular receptive fields of the tangential cells, most NMNs are sensitive to visual motion presented to either eye. This results in the NMNs being more selective for rotation than the LPTCs. Thus, the neck motor system increases its rotation selectivity by a comparatively simple mechanism: the integration of binocular visual motion information.     

Discussion of errors and measuring strategies in morphometry using analysis of variance

Statistical techniques known as the analysis of variance make it possible for the morphologist to plan work in such a way as to get quantitative data with the greatest possible economy of effort. This paper explains how to decide how many measurements to make per micrograph, how many micrographs per tissue block or organ, and how many organs or individuals are necessary for getting an exactness of sufficient quality of the results. The examples furnished have been taken from measuring volume densities of mitochondria in heart muscle cells and from cell counting in lymph nodes. Finally we show, how to determine sample sizes, if we are interested in demonstration of significant differences between mean values.     

The Origin of Patterning Systems in Bilateria——Insights from the Hox and ParaHox Genes in Acoelomorpha

Hox and ParaHox genes constitute two families of developmental regulators that pattern the Anterior-Posterior body axis in all bilaterians.The members of these two groups of genes are usually arranged in genomic clusters and work in a coordinated fashion,both in space and in time. While the mechanistic aspects of their action are relatively well known,it is still unclear how these systems evolved. For instance,we still need a proper model of how the Hox and ParaHox clusters were assembled over time.This problem is due to the shortage of information on gene complements for many taxa (mainly basal metazoans) and the lack of a consensus phylogenetic model of animal relationships to which we can relate our new findings.Recently, several studies have shown that the Acoelomorpha most probably represent the first offshoot of the Bilateria. This finding has prompted us,and others, to study the Hox and ParaHox complements in these animals,as well as their activity during development.In this review,we analyze how the current knowledge of Hox and ParaHox genes in the Acoelomorpha is shaping our view of bilaterian evolution.     

The cell cycle and Toxoplasma gondii cell division: tightly knit or loosely stitched?

The flexibility displayed by apicomplexan parasites to vary their mode of replication has intrigued biologists since their discovery by electron microscopy in the 1960s and 1970s. Starting in the 1990s we began to understand the cell biology of the cytoskeleton elements driving cytokinesis. By contrast, the molecular mechanisms that regulate the various division modes and how they translate into the budding process that uniquely characterizes this parasite family are much less understood. Although growth mechanisms are a neglected area of study, it is an important pathogenic parameter as fast division rounds are associated with fulminant infection whereas slower growth attenuates virulence, as is exploited in some vaccine strains. In this review we summarize a recent body of cell biological experiments that are rapidly leading to an understanding of the events that yield successful mitosis and cytokinesis in Toxoplasma. We place these observations within a cell cycle context with comments on how these events may be regulated by known eukaryotic checkpoints active in fission and budding yeasts as well as mammalian cells. The presence of cell cycle control mechanisms in the Apicomplexa is supported by our findings that identify several cell cycle checkpoints in Toxoplasma. The progress of the cell cycle is ultimately controlled by cyclin-Cdk pair activities, which are present throughout the Apicomplexa. Although many of the known controllers of cyclin-Cdk activity are present, several key controls cannot readily be identified, suggesting that apicomplexan parasites deviate at these points from the higher eukaryotic models. Altogether, new insights in Toxoplasma replication are reciprocally applied to hypothesize how other division modes in the Toxoplasma life cycle and in other Apicomplexa species could be controlled in terms of cell cycle checkpoint regulation.     

Visual ecology and functional morphology of cubozoa (cnidaria)

Jellyfish belong to one of the oldest extant animal phyla, the Cnidaria. The first Cnidaria appear in the fossil record 600 million years ago, preceeding the Cambrian explosion. They are an extremely successful group present in all marine environments and some freshwater environments. In contrast to many animal phyla in which vision is a primary sense Cnidarians do not, generally, employ image forming eyes. One small class stands alone: the Cubozoa. Cubomedusae are commonly known as box jellyfish. They possess image forming eyes (Coates et al., 2001) which certainly evolved independently from other metazoans. Cubomedusae therefore offer a unique perspective on the evolution of image forming eyes. This literature review collects, into one place, what is known about: the multiple eye types of box jellyfish, cubomedusan life history and ecology, and the sensory and neural systems of box jellyfish. Here I discuss how these features set cubomedusae apart from scyphomedusae and hydromedusae. Knowledge in these areas is sparse; the work done to date inspires increased efforts.     

Bayesian inference for stationary points in Gaussian process regression models for event-related potentials analysis

Stationary points embedded in the derivatives are often critical for a model to be interpretable and may be considered as key features of interest in many applications. We propose a semiparametric Bayesian model to efficiently infer the locations of stationary points of a nonparametric function, which also produces an estimate of the function. We use Gaussian processes as a flexible prior for the underlying function and impose derivative constraints to control the function's shape via conditioning. We develop an inferential strategy that intentionally restricts estimation to the case of at least one stationary point, bypassing possible mis-specifications in the number of stationary points and avoiding the varying dimension problem that often brings in computational complexity. We illustrate the proposed methods using simulations and then apply the method to the estimation of event-related potentials derived from electroencephalography (EEG) signals. We show how the proposed method automatically identifies characteristic components and their latencies at the individual level, which avoids the excessive averaging across subjects that is routinely done in the field to obtain smooth curves. By applying this approach to EEG data collected from younger and older adults during a speech perception task, we are able to demonstrate how the time course of speech perception processes changes with age.     

Statistical Confusion Among Graduate Students: Sickness or Symptom?

Abstract: Statistics is one of the most important yet difficult subjects for many ecology and wildlife graduate students to learn. Insufficient knowledge about how to conduct quality science and the ongoing debate about the relative value of competing statistical ideologies contribute to uncertainties among graduate students regarding which statistical tests are most appropriate. Herein, we argue that increased education of the available statistical tests alone is unlikely to ameliorate the problem. Instead, we suggest that statistical uncertainties among graduate students are a secondary symptom of a larger problem. We believe the root cause lies in the lack of education on how to conduct science as an integrated process from hypothesis creation through statistical analysis. We argue that if students are taught to think about how each step of the process will affect all other steps, many statistical uncertainties will be avoided.     

On age distribution in population growth

The statistical problem of age distribution in population growth involves the well known difficulties in mathematical probability of defining the distribution of a discrete number of random points (individuals) in a continuous parametric space (age). The assumption of the possibility of multiple births makes the problem more complicated, and we are constrained to introduce the concept of multiple points. This leads to an extension of the method of product densities devised earlier for the treatment of population problems. The paper deals with such an extension and as an example of the application of this method the population problem with twins is discussed. This paper was written while on leave of absence from there.     

Credit Assignment during Movement Reinforcement Learning

We often need to learn how to move based on a single performance measure that reflects the overall success of our movements. However, movements have many properties, such as their trajectories, speeds and timing of end-points, thus the brain needs to decide which properties of movements should be improved; it needs to solve the credit assignment problem. Currently, little is known about how humans solve credit assignment problems in the context of reinforcement learning. Here we tested how human participants solve such problems during a trajectory-learning task. Without an explicitly-defined target movement, participants made hand reaches and received monetary rewards as feedback on a trial-by-trial basis. The curvature and direction of the attempted reach trajectories determined the monetary rewards received in a manner that can be manipulated experimentally. Based on the history of action-reward pairs, participants quickly solved the credit assignment problem and learned the implicit payoff function. A Bayesian credit-assignment model with built-in forgetting accurately predicts their trial-by-trial learning.     

The fitting of dynamic models to observed data

The question of how to fit a general cubic model of a multicomponent, interactive growth system to observed data is addressed. A multidimensional-polynomial type of regression analysis is used, with a least-squares criterion. By testing the scheme on a problem with known solution, the way in which the accuracy of the results varies with the number of datum points used is investigated in an heuristic manner.     

A protocol for imaging alternative splicing regulation in vivo using fluorescence reporters in transgenic mice

Imaging technologies are influencing the way we study regulatory processes in vivo. Several recent reports use fluorescence minigenes to image alternative splicing events in living cells and animals. This type of reporter is being used to generate transgenic mice to visualize splicing regulation in diverse tissues and cell types. In this protocol, we describe how to develop animals that report on alternative splicing and how to assess reporter expression in excised organs and tissue sections. The entire procedure, from making the reporters to imaging organs and tissues in adult transgenic mice, should take approximately 1.5 years. Fluorescence reporters can be used to image many splicing decisions in normal tissues and organs and can be extended to the study of disease states.