Search NASA⌕ Search

Engineering topics

Kerekes, Ryan A

Publications and source records attributed to Kerekes, Ryan A.

Extraction and analysis of neuron firing signals from deep cortical video microscopy

We introduce a method for extracting and analyzing neuronal activity time signals from video of the cortex of a live animal. The signals correspond to the firing activity of individual cortical neurons. Activity signals are based on the changing fluorescence of calcium indicators in the cells over time. We propose a cell segmentation method that relies on a user-specified center point, from which the signal extraction method proceeds. A stabilization approach is used to reduce tissue motion in the video. The extracted signal is then processed to flatten the baseline and detect action potentials. We show results from applying the method to a cortical video of a live mouse.

Kerekes, Ryan A↗

Automated Tracing and Segmentation Tool for Migrating Neurons in 4D Confocal Imagery

Accurate tracing and segmentation of subcellular components of migrating neurons in confocal image sequences are prerequisite steps in many neurobiology studies to understand the biological machinery behind the movement of developing neurons. In this paper, we present an automated tracking, tracing, and segmentation tool for soma, leading, and trailing process of migrating neurons in time-lapse image stacks acquired with a confocal fluorescence microscope. In our approach, we first localize each neuron in the maximum intensity projection of the first frame using manual labeling of the soma and end points of the leading and trailing process. By using each soma position at the first frame, we automatically track the somas in rest of the frames. Then, leading and trailing process are traced in each frame from the soma center to the labeled end tip of the process by using fast marching algorithm. Finally, the soma, leading and trailing processes of each neuron are segmented by using the soma center and traces as seed points, and their boundaries are separated from each other. Based on qualitative results, we demonstrate the capability to automatically track, trace, and segment the soma, leading, and trailing processes of a migrating neuron with minimal user input.

Karakaya, Mahmut↗

Analysis of Tight Junction Formation and Integrity

In this paper, we study segmentation of tight junctions and analyze the formation and integrity of tight junctions in large-scale confocal image stacks, a challenging biological problem because of the low spatial resolution images and the presence of breaks in tight junction structure. We present an automated, three-step processing approach for tight junction analysis. In our approach, we first localize each individual nucleus in the image by using thresholding, morphological filters and active contours. By using each nucleus position as a seed point, we automatically segment the cell body based on the active contour. We then use an intensity-based skeletonization algorithm to generate the boundary regions for each cell, and features are extracted from tight junctions associated with each cell to assess tight junction continuity. Based on qualitative results and quantitative comparisons, we show that we are able to automatically segment tight junctions and compute relevant features that provide a quantitative measure of tight junction formation to which the permeability of the cell monolayer can ultimately be correlated.

Karakaya, Mahmut↗

Automated Tracing of Horizontal Neuron Processes During Retinal Development

In the developing mammalian retina, horizontal neurons undergo a dramatic reorganization oftheir processes shortly after they migrate to their appropriate laminar position. This is an importantprocess because it is now understood that the apical processes are important for establishing theregular mosaic of horizontal cells in the retina and proper reorganization during lamination isrequired for synaptogenesis with photoreceptors and bipolar neurons. However, this process isdifficult to study because the analysis of horizontal neuron anatomy is labor intensive and time-consuming. In this paper, we present a computational method for automatically tracing the three-dimensional (3-D) dendritic structure of horizontal retinal neurons in two-photon laser scanningmicroscope (TPLSM) imagery. Our method is based on 3-D skeletonization and is thus able topreserve the complex structure of the dendritic arbor of these cells. We demonstrate theeffectiveness of our approach by comparing our tracing results against two sets of semi-automatedtraces over a set of 10 horizontal neurons ranging in age from P1 to P5. We observe an averageagreement level of 81% between our automated trace and the manual traces. This automatedmethod will serve as an important starting point for further refinement and optimization.

59 BASIC BIOLOGICAL SCIENCES↗

Retinoblastoma (Rb) regulates laminar dendritic arbor reorganization in retinal horizontal neurons

Neuronal differentiation with respect to the acquisition of synaptic competence needs to be regulated precisely during neurogenesis to ensure proper formation of circuits at the right place and time in development.This regulation is particularly important for synaptic triads among photoreceptors, horizontal cells (HCs), and bipolar cells in the retina, because HCs are among the rst cell types produced during development, and bipolar cells are among thel ast.HCs undergo a dramatic transition from vertically oriented neurites that form columnar arbors to overlapping laminar dendritic arbors with differentiation.However, how this process is regulated and coordinated with differentiation of photoreceptors and bipolar cells remains unknown. Previous studies have suggested that there tino-blastoma(Rb) tumor suppressor gene may play a role in horizontal cell differentiation and synaptogenesis. By combining genetic mosaic analysis of individual synaptictriads with neuroanatomic analyses and multiphoton live imaging of developing HCs, we found that Rb plays a cell-autonomous role in there organization of horizontal cell neurites as they differentiate. Aberrant vertical processes in Rb-de cient HCs form ectopic synapses with rods in the outer nuclear layer but lack bipolar dendrites. Although previous reports indicate that photoreceptor abnormalities can trigger formation of ectopic synapses, our studies now demonstrate that defects in a post synaptic partner contribute to the formation of ectopic photoreceptor synapses in the mammalian retina.

59 BASIC BIOLOGICAL SCIENCES↗

Automatic Detection, Segmentation and Classification of Retinal Horizontal Neurons in Large-scale 3D Confocal Imagery

Automatic analysis of neuronal structure from wide-field-of-view 3D image stacks of retinal neurons is essential for statistically characterizing neuronal abnormalities that may be causally related to neural malfunctions or may be early indicators for a variety of neuropathies. In this paper, we study classification of neuron fields in large-scale 3D confocal image stacks, a challenging neurobiological problem because of the low spatial resolution imagery and presence of intertwined dendrites from different neurons. We present a fully automated, four-step processing approach for neuron classification with respect to the morphological structure of their dendrites. In our approach, we first localize each individual soma in the image by using morphological operators and active contours. By using each soma position as a seed point, we automatically determine an appropriate threshold to segment dendrites of each neuron. We then use skeletonization and network analysis to generate the morphological structures of segmented dendrites, and shape-based features are extracted from network representations of each neuron to characterize the neuron. Based on qualitative results and quantitative comparisons, we show that we are able to automatically compute relevant features that clearly distinguish between normal and abnormal cases for postnatal day 6 (P6) horizontal neurons.

99 GENERAL AND MISCELLANEOUS↗

Fully automated segmentation and characterization of the dendritic trees of retinal horizontal neurons

We introduce a new fully automated method for segmenting and characterizing the dendritic tree of neurons in confocal image stacks. Our method is aimed at wide-field-of-view, low-resolution imagery of retinal neurons in which dendrites can be intertwined and difficult to follow. The approach is based on 3-D skeletonization and includes a method for automatically determining an appropriate global threshold as well as a soma detection algorithm. We provide the details of the algorithm and a qualitative performance comparison against a commercially available neurite tracing software package, showing that a segmentation produced by our method more closely matches the ground-truth segmentation.

59 BASIC BIOLOGICAL SCIENCES↗

Automated method for tracing leading and trailing processes of migrating neurons in confocal image sequences

Segmentation, tracking, and tracing of neurons in video imagery are important steps in many neuronal migration studies and can be in accurate and time-consuming when performed manually. In this paper, we present an automated method for tracing the leading and trailing processes of migrating neurons in time-lapse image stacks acquired with a confocal fluorescence microscope. In our approach, we first locate and track the soma of the cell of interest by smoothing each frame and tracking the local maxima through the sequence. We then trace the leading process in each frame by starting at the center of the soma and stepping repeatedly in the most likely direction of the leading process. This direction is found at each step by examining second derivatives of fluorescent intensity along curves of constant radius around the current point. Tracing terminates after a fixed number of steps or when fluorescent intensity drops below a fixed threshold. We evolve the resulting trace to form an improved trace that more closely follows the approximate centerline of the leading process. We apply a similar algorithm to the trailing process of the cell by starting the trace in the opposite direction. We demonstrate our algorithm on two time-lapse confocal video sequences of migrating cerebellar granule neurons(CGNs). We show that the automated traces closely approximate ground truth traces to within 1 or 2 pixels on average. Additionally, we compute line intensity profiles of fluorescence along the automated traces and quantitatively demonstrate their similarity to manually generated profiles in terms of fluorescence peak locations.

99 GENERAL AND MISCELLANEOUS↗

Design of Dual-Road Transportable Portal Monitoring System for Visible Light and Gamma-Ray Imaging

The use of radiation sensors as portal monitors is increasing due to heightened concerns over the smuggling of fissile material. Transportable systems that can detect significant quantities of fissile material that might be present in vehicular traffic are of particular interest, especially if they can be rapidly deployed to different locations. To serve this application, we have constructed a rapid-deployment portal monitor that uses visible-light and gamma-ray imaging to allow simultaneous monitoring of multiple lanes of traffic from the side of a roadway. The system operation uses machine vision methods on the visible-light images to detect vehicles as they enter and exit the field of view and to measure their position in each frame. The visible-light and gamma-ray cameras are synchronized which allows the gamma-ray imager to harvest gamma-ray data specific to each vehicle, integrating its radiation signature for the entire time that it is in the field of view. Thus our system creates vehicle-specific radiation signatures and avoids source confusion problems that plague non-imaging approaches to the same problem. Our current prototype instrument was designed for measurement of upto five lanes of freeway traffic with a pair of instruments, one on either side of the roadway. Stereoscopic cameras are used with a third alignment camera for motion compensation and are mounted on a 50 deployable mast. In this paper we discuss the design considerations for the machine-vision system, the algorithms used for vehicle detection and position estimates, and the overall architecture of the system. We also discuss system calibration for rapid deployment. We conclude with notes on preliminary performance and deployment.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Myosin II Motors and F-Actin Dynamics Drive the Coordinated Movement of the Centrosome and Soma during CNS Glial-Guided Neuronal Migration

Lamination of cortical regions of the vertebrate brain depends on glial-guided neuronal migration. The conserved polarity protein Par6{alpha} localizes to the centrosome and coordinates forward movement of the centrosome and soma in migrating neurons. The cytoskeletal components that produce this unique form of cell polarity and their relationship to polarity signaling cascades are unknown. We show that F-actin and Myosin II motors are enriched in the neuronal leading process and that Myosin II activity is necessary for leading process actin dynamics. Inhibition of Myosin II decreased the speed of centrosome and somal movement, whereas Myosin II activation increased coordinated movement. Ectopic expression or silencing of Par6{alpha} inhibited Myosin II motors by decreasing Myosin light-chain phosphorylation. These findings suggest leading-process Myosin II may function to 'pull' the centrosome and soma forward during glial-guided migration by a mechanism involving the conserved polarity protein Par6{alpha}.

59 BASIC BIOLOGICAL SCIENCES↗

Enhanced video-based target detection using multi-frame correlation filtering

Most existing video-based target detection systems employ state-space models to keep track of an explicit number of individual targets. We introduce a framework for enhancing target detection in video by applying probabilistic models to the soft information in correlation outputs before thresholding. We show how to efficiently compute arrays of posterior target probabilities for every position in the scene conditioned on all current and past frames of a video sequence. These arrays can then be thresholded in the typical manner to yield more reliable target detections. Because the framework avoids the formation of explicit tracks, it is well suited for handling scenes with unknown numbers of targets at unknown positions. Simulation results on forward-looking infrared (FLIR) video sequences show that our proposed framework can significantly reduce the false-alarm rate of a bank of correlation filters while requiring only a marginal increase in computation.

42 ENGINEERING↗

Decision-level Information Fusion to Assess Threat Likelihood in Shipped Containers

A significant challenge in the distribution of goods is assessing the potential threat that an individual shipping container poses. Due to the high volume of shipped goods, a primary concern is balancing accuracy and container scan time. The application of information fusion to the problem enables automated threat determination and the presentation of relevant data to an operator, in a decision support capacity, in order to maintain a sufficient level of processing. This paper outlines an approach to container threat assessment that combines data from multiple sources in order to reliably score the likelihood that a given container holds a threat. Fused data is also leveraged as a tool to optimize the routing of containers through a scanning system comprised of multiple data acquisition stations and providing data in multiple modes. Furthermore, we propose methods for the consolidated presentation of fused data to an operator in order to both minimize the time expended in container evaluation and maximize the accuracy of the assessment.

97 MATHEMATICS AND COMPUTING↗

An Information Fusion Framework for Threat Assessment

Modern enterprises are becoming increasingly sensitive to the potential destructive power of small groups or individuals with malicious intent. In response, significant investments are being made in developing a means to assess the likelihood of certain threats to their enterprises. Threat assessment needs are typically focused in very specific application areas where current processes rely heavily on human analysis to both combine any available data and draw conclusions about the probability of a threat. A generic approach to threat assessment is proposed, including a threat taxonomy and decision-level information fusion framework, that provides a computational means for merging multi-modal data for the purpose of assessing the presence of a threat. The framework is designed for flexibility, and intentionally accounts for the accuracy of each data source, given the environmental conditions, in order to manage the uncertainty associated with any acquired data. The taxonomy and information fusion framework is described, and discussed in the context of real-world applications such as shipping container security and cyber security.

99 GENERAL AND MISCELLANEOUS↗

Automated 3-D Tracking of Centrosomes in Sequences of Confocal Image Stacks

In order to facilitate the study of neuron migration, we propose a method for 3-D detection and tracking of centrosomes in time-lapse confocal image stacks of live neuron cells. We combine Laplacian-based blob detection, adaptive thresholding, and the extraction of scale and roundness features to find centrosome-like objects in each frame. We link these detections using the joint probabilistic data association filter (JPDAF) tracking algorithm with a Newtonian state-space model tailored to the motion characteristics of centrosomes in live neurons. We apply our algorithm to image sequences containing multiple cells, some of which had been treated with motion-inhibiting drugs. We provide qualitative results and quantitative comparisons to manual segmentation and tracking results showing that our motion estimates closely agree with those generated by neurobiology experts.

60 APPLIED LIFE SCIENCES↗