António R. C. Paiva

Detection of neuron membranes in electron microscopy images using a serial neural network architecture

Study of nervous systems via the connectome, the map of connectivities of all neurons in that system, is a challenging problem in neuroscience. Towards this goal, neurobiologists are acquiring large electron microscopy datasets. However, the shear volume of these datasets renders manual analysis infeasible. Hence, automated image analysis methods are required for reconstructing the connectome from these very large image collections. Segmentation of neurons in these images, an essential step of the reconstruction pipeline, is challenging because of noise, anisotropic shapes and brightness, and the presence of confounding structures. The method described in this paper uses a series of artificial neural networks (ANNs) in a framework combined with a feature vector that is composed of image intensities sampled over a stencil neighborhood. Several ANNs are applied in series allowing each ANN to use the classification context provided by the previous network to improve detection accuracy. We develop the method of serial ANNs and show that the learned context does improve detection over traditional ANNs. We also demonstrate advantages over previous membrane detection methods. The results are a significant step towards an automated system for the reconstruction of the connectome.

A Reproducing Kernel Hilbert Space Framework for Information-Theoretic Learning

This paper provides a functional analysis perspective of information-theoretic learning (ITL) by defining bottom-up a reproducing kernel Hilbert space (RKHS) uniquely determined by the symmetric nonnegative definite kernel function known as the cross-information potential (CIP). The CIP as an integral of the product of two probability density functions characterizes similarity between two stochastic functions. We prove the existence of a one-to-one congruence mapping between the ITL RKHS and the Hilbert space spanned by square integrable probability density functions. Therefore, all the statistical descriptors in the original information-theoretic learning formulation can be rewritten as algebraic computations on deterministic functional vectors in the ITL RKHS, instead of limiting the functional view to the estimators as is commonly done in kernel methods. A connection between the ITL RKHS and kernel approaches interested in quantifying the statistics of the projected data is also established.

Automatic markup of neural cell membranes using boosted decision stumps

To better understand the central nervous system, neurobiologists need to reconstruct the underlying neural circuitry from electron microscopy images. One of the necessary tasks is to segment the individual neurons. For this purpose, we propose a supervised learning approach to detect the cell membranes. The classifier was trained using AdaBoost, on local and context features. The features were selected to highlight the line characteristics of cell membranes. It is shown that using features from context positions allows for more information to be utilized in the classification. Together with the nonlinear discrimination ability of the AdaBoost classifier, this results in clearly noticeable improvements over previously used methods.

Kernel Methods on Spike Train Space for Neuroscience: A Tutorial

Over the last decade, several positive-definite kernels have been proposed to treat spike trains as objects in Hilbert space. However, for the most part, such attempts still remain a mere curiosity for both computational neuroscientists and signal processing experts. This tutorial illustrates why kernel methods can, and have already started to, change the way spike trains are analyzed and processed. The presentation incorporates simple mathematical analogies and convincing practical examples in an attempt to show the yet unexplored potential of positive definite functions to quantify point processes. It also provides a detailed overview of the current state of the art and future challenges with the hope of engaging the readers in active participation.

Nonlinear Component Analysis Based on Correntropy

In this paper, we propose a new nonlinear principal component analysis based on a generalized correlation function which we call correntropy. The data is nonlinearly transformed to a feature space, and the principal directions are found by eigen-decomposition of the correntropy matrix, which has the same dimension as the standard covariance matrix for the original input data. The correntropy matrix characterizes the nonlinear correlations between the data. With the correntropy function, one can efficiently compute the principal components in the feature space by projecting the transformed data onto those principal directions. We give the derivation of the new method and present simulation results.

Fast AdaBoost training using weighted novelty selection

In this paper, a new AdaBoost learning framework, called WNS-AdaBoost, is proposed for training discriminative models. The proposed approach significantly speeds up the learning process of adaptive boosting (AdaBoost) by reducing the number of data points. For this purpose, we introduce the weighted novelty selection (WNS) sampling strategy and combine it with AdaBoost to obtain an efficient and fast learning algorithm. WNS selects a representative subset of data thereby reducing the number of data points onto which AdaBoost is applied. In addition, WNS associates a weight with each selected data point such that the weighted subset approximates the distribution of all the training data. This ensures that AdaBoost can trained efficiently and with minimal loss of accuracy. The performance of WNS-AdaBoost is first demonstrated in a classification task. Then, WNS is employed in a probabilistic boosting-tree (PBT) structure for image segmentation. Results in these two applications show that the training time using WNS-AdaBoost is greatly reduced at the cost of only a few percent in accuracy.

Semi-Automated Neuron Boundary Detection and Nonbranching Process Segmentation in Electron Microscopy Images

Neuroscientists are developing new imaging techniques and generating large volumes of data in an effort to understand the complex structure of the nervous system. The complexity and size of this data makes human interpretation a labor-intensive task. To aid in the analysis, new segmentation techniques for identifying neurons in these feature rich datasets are required. This paper presents a method for neuron boundary detection and nonbranching process segmentation in electron microscopy images and visualizing them in three dimensions. It combines both automated segmentation techniques with a graphical user interface for correction of mistakes in the automated process. The automated process first uses machine learning and image processing techniques to identify neuron membranes that deliniate the cells in each two-dimensional section. To segment nonbranching processes, the cell regions in each two-dimensional section are connected in 3D using correlation of regions between sections. The combination of this method with a graphical user interface specially designed for this purpose, enables users to quickly segment cellular processes in large volumes.

On the use of standards for microarray lossless image compression

The interest in methods that are able to efficiently compress microarray images is relatively new. This is not surprising, since the appearance and fast growth of the technology responsible for producing these images is also quite recent. In this paper, we present a set of compression results obtained with 49 publicly available images, using three image coding standards: lossless JPEG2000, JBIG, and JPEG-LS. We concluded that the compression technology behind JBIG seems to be the one that offers the best combination of compression efficiency and flexibility for microarray image compression.

Fast semi-supervised image segmentation by novelty selection

The goal of semi-supervised image segmentation is to obtain the segmentation from a partially labeled image. By utilizing the image manifold structure in labeled and unlabeled pixels, semi-supervised methods propagate the user labeling to the unlabeled data, thus minimizing the need for user labeling. Several semi-supervised learning methods have been proposed in the literature. Although results have been promising, these methods are very computationally intensive. In this paper, we propose novelty selection as a pre-processing step to reduce the number of data points while retaining the fundamental structure of the data. Since the computational complexity is a power of the number of points, it is possible to significantly reduce the overall computation requirements. Results in several images show that the computation time is greatly reduced without sacrifice in segmentation accuracy.

Self-organizing maps with dynamic learning for signal reconstruction

Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.