The role of presynaptic dynamics in processing of natural spike trains in hippocampal synapses.

Short-term plasticity (STP) represents a key neuronal mechanism of information processing. In excitatory hippocampal synapses, STP serves as a high-pass filter optimized for the transmission of information-carrying place-field discharges. This STP filter enables synapses to perform a highly nonlinear, switch-like operation permitting the passage and amplification of signals with place-field-like characteristics. Because of the complexity of interactions among STP processes, the synaptic mechanisms underlying this filtering paradigm remain poorly understood. Here, we describe a simple mechanistic model of STP, derived in large part from basic principles of synaptic function, that reproduces this highly nonlinear synaptic behavior. The model, formulated in terms of release probability, considers the interactions between calcium-dependent forms of presynaptic enhancement and their impact on vesicle pool dynamics, which is described using a two-pool model of vesicle recruitment. By considering the interdependency between release probability and various forms of STP, the model attempts to provide a realistic coupling among major presynaptic processes. The model parameters are first determined using synaptic dynamics during constant-frequency stimulation. The model then accurately reproduces all major characteristics of the synaptic filtering paradigm during natural stimulus patterns without free parameters. An elimination approach is then used to identify the contribution of each STP component to synaptic dynamics. Based on this analysis, the model predicts strong calcium dependence of synaptic filtering properties, which is verified experimentally in rat hippocampal slices. This simple model may thus offer a useful framework to further investigate the role of STP in neural computations.

Automated condition-invariable neurite segmentation and synapse classification using textural analysis-based machine-learning algorithms.

High-resolution live-cell imaging studies of neuronal structure and function are characterized by large variability in image acquisition conditions due to background and sample variations as well as low signal-to-noise ratio. The lack of automated image analysis tools that can be generalized for varying image acquisition conditions represents one of the main challenges in the field of biomedical image analysis. Specifically, segmentation of the axonal/dendritic arborizations in brightfield or fluorescence imaging studies is extremely labor-intensive and still performed mostly manually. Here we describe a fully automated machine-learning approach based on textural analysis algorithms for segmenting neuronal arborizations in high-resolution brightfield images of live cultured neurons. We compare performance of our algorithm to manual segmentation and show that it combines 90% accuracy, with similarly high levels of specificity and sensitivity. Moreover, the algorithm maintains high performance levels under a wide range of image acquisition conditions indicating that it is largely condition-invariable. We further describe an application of this algorithm to fully automated synapse localization and classification in fluorescence imaging studies based on synaptic activity. Textural analysis-based machine-learning approach thus offers a high performance condition-invariable tool for automated neurite segmentation.

Robust color texture features under varying illumination conditions.

Under varying illumination, both the statistical and structural contents of color texture are modified, leading to changes in the observed texture surface. We model the effect of illumination as a perturbation on an ideal color texture and show that the spectra of the ambient light have a significant impact on the observed texture patterns in the individual color channels. Motivated by studies in human color constancy, we propose a correlation-based transformation that minimizes the effect of illumination variation in color texture analysis. Experimental results are included, which validate the performance of the proposed minvariance model in the analysis of color texture.

Comparison of texture analysis schemes under nonideal conditions.

Several recent advancements in the field of texture analysis prompt some fundamental questions. For instance, what is the true impact of these novel advancements under real-world environments? When do these novel advancements fail to perform? Which methods perform better and under what conditions? In this work, we investigate these and other issues under nonideal image acquisition environments, specifically, environments with changing conditions due to illumination variations and those caused by both affine and nonaffine transformations. We study the performance of nine popular texture analysis algorithms using three different datasets, with varying levels of difficulty. Experiments are performed on nonideal texture datasets under five different setups. We find that most state-of-the-art techniques do not perform well under these conditions. To a large extent, their performance under nonideal conditions depends critically on the nature of the textural surface. Moreover, most techniques fail to perform reliably when the number of classes in the dataset is increased significantly, over the regular-size datasets used in previous work. Multiscale features performed reasonably well against variations caused by illumination and rotation but are prone to fail under changes in scale. Surprisingly, the performance for most of the algorithms is generally stable on structured or periodic textures, even with variations in illumination or affine transformations.

Texton-based segmentation of retinal vessels.

With improvements in fundus imaging technology and the increasing use of digital images in screening and diagnosis, the issue of automated analysis of retinal images is gaining more serious attention. We consider the problem of retinal vessel segmentation, a key issue in automated analysis of digital fundus images. We propose a texture-based vessel segmentation algorithm based on the notion of textons. Using a weak statistical learning approach, we construct textons for retinal vasculature by designing filters that are specifically tuned to the structural and photometric properties of retinal vessels. We evaluate the performance of the proposed approach using a standard database of retinal images. On the DRIVE data set, the proposed method produced an average performance of 0.9568 specificity at 0.7346 sensitivity. This compares well with the best-published results on the data set 0.9773 specificity at 0.7194 sensitivity [Proc. SPIE5370, 648 (2004)].