Connectivity graphs: a method for proving deadlock freedom based on separation logic

We introduce the notion of a connectivity graph —an abstract representation of the topology of concurrently interacting entities, which allows us to encapsulate generic principles of reasoning about deadlock freedom . Connectivity graphs are parametric in their vertices (representing entities like threads and channels) and their edges (representing references between entities) with labels (representing interaction protocols). We prove deadlock and memory leak freedom in the style of progress and preservation and use separation logic as a meta theoretic tool to treat connectivity graph edges and labels substructurally. To prove preservation locally, we distill generic separation logic rules for local graph transformations that preserve acyclicity of the connectivity graph. To prove global progress locally, we introduce a waiting induction principle for acyclic connectivity graphs. We mechanize our results in Coq, and instantiate our method with a higher-order binary session-typed language to obtain the first mechanized proof of deadlock and leak freedom.

Novel Semi-Supervised Hyperspectral Image Classification Based on a Superpixel Graph and Discrete Potential Method

Hyperspectral image (HSI) classification plays an important role in the automatic interpretation of the remotely sensed data. However, it is a non-trivial task to classify HSI accurately and rapidly due to its characteristics of having a large amount of data and massive noise points. To address this problem, in this work, a novel, semi-supervised, superpixel-level classification method for an HSI was proposed based on a graph and discrete potential (SSC-GDP). The key idea of the proposed scheme is the construction of the weighted connectivity graph and the division of the weighted graph. Based on the superpixel segmentation, a weighted connectivity graph is constructed usingthe weighted connection between a superpixel and its spatial neighbors. The generated graph is then divided into different communities/sub-graphs by using a discrete potential and the improved semi-supervised Wu–Huberman (ISWH) algorithm. Each community in the weighted connectivity graph represents a class in the HSI. The local connection strategy, together with the linear complexity of the ISWH algorithm, ensures the fast implementation of the suggested SSC-GDP method. To prove the effectiveness of the proposed spectral–spatial method, two public benchmarks, Indian Pines and Salinas, were utilized to test the performance of our proposal. The comparative test results confirmed that the proposed method was superior to several other state-of-the-art methods.

Automatic Detection of Synaptic Partners in a Whole-Brain Drosophila EM Dataset

The study of neural circuits requires the reconstruction of neurons and the identification of synaptic connections between them. To scale the reconstruction to the size of whole-brain datasets, semi-automatic methods are needed to solve those tasks. Here, we present an automatic method for synaptic partner identification in insect brains, which uses convolutional neural networks to identify post-synaptic sites and their pre-synaptic partners. The networks can be trained from human generated point annotations alone and require only simple post-processing to obtain final predictions. We used our method to extract 244 million putative synaptic partners in the fifty-teravoxel full adult fly brain (FAFB) electron microscopy (EM) dataset and evaluated its accuracy on 146,643 synapses from 702 neurons with a total cable length of 312 mm in four different brain regions. The predicted synaptic connections can be used together with a neuron segmentation to infer a connectivity graph with high accuracy: between 92% and 96% of edges linking connected neurons are correctly classified as weakly connected (less than five synapses) and strongly connected (at least five synapses). Our synaptic partner predictions for the FAFB dataset are publicly available, together with a query library allowing automatic retrieval of up- and downstream neurons.

Identifying Brain Region Connectivity using Steiner Minimal Tree Approximation and a Genetic Algorithm

BackgroundComputation and visualization of connectivity among the brain regions is vital for tasks such as disease identification and drug discovery. An effective visualization can aid clinicians and biologists to perform these tasks addressing a genuine research and industrial need. In this paper, we present SMT-Neurophysiology, a web-based tool in a form of an approximation to the Steiner Minimal Tree (SMT) algorithm to search neurophysiology partonomy and connectivity graph in order to find biomedically-meaningful paths that could explain, to neurologists and neuroscientists, the mechanistic relationship, for example, among specific neurophysiological examinations. We also present SMT-Genetic, a web-based tool in a form of a Genetic Algorithm (GA) to find better paths than SMT-Neurophysiology.ResultsWe introduce an approximation to the SMT algorithm to identify the most parsimonious connectivity among the brain regions of interest. We have implemented our algorithm as a highly interactive web application called SMT-Neurophysiology that enables such computation and visualization. It operates on brain region connectivity dataset curated from the Neuroscience Information Framework (NIF) for four species – human, monkey, rat and bird. We present two case studies on finding the most biomedically-meaningful solutions that identifies connections among a set of brain regions over a specific route. The case studies demonstrate that SMT-Neurophysiology is able to find connection among brain regions of interest. Furthermore, SMT-Neurophysiology is modular and generic in nature allowing the underlying connectivity graph to model any data on which the tool can operate. In order to find better connections among a set of brain regions than SMT-Neurophysiology, we have implemented a web-based tool in a form of a GA called SMT-Genetic. We present further three case studies where SMT-Genetic finds better connections among a set of brain regions than SMT-Neurophysiology. SMT-Genetic gives better connections because SMT-Genetic finds global optimum whereas SMT-Neurophysiology finds local optimum although execution time of SMT-Genetic is higher than SMT-Neurophysiology.ConclusionOur analysis would provide key insights to clinical investigators about potential mechanisms underlying a particular neurological disease. The web-based tools and the underlying data are useful to clinicians and scientists to understand neurological disease mechanisms; discover pharmacological or surgical targets; and design diagnostic or therapeutic clinical trials. The source codes and links to the live tools are available at https://github.com/dewancse/connected-brain-regions and https://github.com/dewancse/SMT-Genetic.

Graph Signal Processing of Low and High-Order Dynamic Functional Connectivity Networks Using EEG Resting-State for Schizophrenia: A Whole Brain Breakdown

Conventional static or dynamic functional connectivity graph (FCG/DFCG) referred to as low-order FCG focusing on temporal correlation estimates of the resting-state electroencephalography (rs-EEG) time series between any potential pair of brain areas. A DFCG is first constructed from multichannel recordings by adopting the methodology of sliding-window and a proper functional connectivity estimator. However, low-order FC ignores the high-level inter-relationship of brain areas. Recently, a high-order version of FCG has emerged by estimating the correlations of the time series that describe the fluctuations of the functional strength of every pair of ROIs across experimental time.In the present study, a dynamic functional connectivity graph (DFCG) has been estimated using the imaginary part of phase lag value (iPLV). We analyzed DFCG profiles of electroencephalographic resting state (eyes-closed) recordings of healthy controls subjects (n=39) and subjects with symptoms of schizophrenia (n=45) in basic frequency bands {δ,θ,α1,α2,β1,β2,γ}. In our analysis, we incorporated both intra and cross-frequency coupling modes. Adopting our recent Dominant Intrinsic Coupling Mode (DICM) model leads to the construction of an integrated DFCG (iDFCG) that encapsulates both the functional strength but also the DICM of every pair of brain areas. Based on the LO - IDFCG, we constructed the HO- IDFCG by adopting the cosine similarity between the time-series derived from the LO-DIFCG. At a second level, we estimated the laplacian transformations of both LO and HO-IDFCG and by calculating the temporal evolution of Synchronizability (Syn), four network metric time series (NMTSSyn) were produced. Following, a machine learning approach based on multi-kernel SVM with the four NMTSSynused as potential features and appropriate kernels, we succeeded a superior classification accuracy (∼98%). DICM and flexibility index (FI) achieved a classification with absolute performance (100 %)Schizophrenic subjects demonstrated a hypo-synchronization compared to healthy control group which can be interpreted as a low global synchronization of co-fluctuate functional patterns. Our analytic pathway could be helpful both for the design of reliable biomarkers and also for evaluating non-intervention treatments tailored to schizophrenia. EEG offers a low-cost environment for applied neuroscience and the transfer of research knowledge from neuroimaging labs to daily clinical practice.