Showing papers on "Pipeline (computing) published in 2017"
01 Jul 2017
TL;DR: This work proposes a simple yet powerful pipeline that yields fast and accurate text detection in natural scenes, and significantly outperforms state-of-the-art methods in terms of both accuracy and efficiency.
Abstract: Previous approaches for scene text detection have already achieved promising performances across various benchmarks. However, they usually fall short when dealing with challenging scenarios, even when equipped with deep neural network models, because the overall performance is determined by the interplay of multiple stages and components in the pipelines. In this work, we propose a simple yet powerful pipeline that yields fast and accurate text detection in natural scenes. The pipeline directly predicts words or text lines of arbitrary orientations and quadrilateral shapes in full images, eliminating unnecessary intermediate steps (e.g., candidate aggregation and word partitioning), with a single neural network. The simplicity of our pipeline allows concentrating efforts on designing loss functions and neural network architecture. Experiments on standard datasets including ICDAR 2015, COCO-Text and MSRA-TD500 demonstrate that the proposed algorithm significantly outperforms state-of-the-art methods in terms of both accuracy and efficiency. On the ICDAR 2015 dataset, the proposed algorithm achieves an F-score of 0.7820 at 13.2fps at 720p resolution.
1,161 citations
TL;DR: In this paper, a convolutional neural network is used to predict the coefficients of a locally affine model in bilateral space, which is then applied to the full-resolution image.
Abstract: Performance is a critical challenge in mobile image processing. Given a reference imaging pipeline, or even human-adjusted pairs of images, we seek to reproduce the enhancements and enable real-time evaluation. For this, we introduce a new neural network architecture inspired by bilateral grid processing and local affine color transforms. Using pairs of input/output images, we train a convolutional neural network to predict the coefficients of a locally-affine model in bilateral space. Our architecture learns to make local, global, and content-dependent decisions to approximate the desired image transformation. At runtime, the neural network consumes a low-resolution version of the input image, produces a set of affine transformations in bilateral space, upsamples those transformations in an edge-preserving fashion using a new slicing node, and then applies those upsampled transformations to the full-resolution image. Our algorithm processes high-resolution images on a smartphone in milliseconds, provides a real-time viewfinder at 1080p resolution, and matches the quality of state-of-the-art approximation techniques on a large class of image operators. Unlike previous work, our model is trained off-line from data and therefore does not require access to the original operator at runtime. This allows our model to learn complex, scene-dependent transformations for which no reference implementation is available, such as the photographic edits of a human retoucher.
510 citations
Posted Content•
TL;DR: This paper proposed a simple yet powerful pipeline that yields fast and accurate text detection in natural scenes, eliminating unnecessary intermediate steps (e.g., candidate aggregation and word partitioning) with a single neural network.
Abstract: Previous approaches for scene text detection have already achieved promising performances across various benchmarks. However, they usually fall short when dealing with challenging scenarios, even when equipped with deep neural network models, because the overall performance is determined by the interplay of multiple stages and components in the pipelines. In this work, we propose a simple yet powerful pipeline that yields fast and accurate text detection in natural scenes. The pipeline directly predicts words or text lines of arbitrary orientations and quadrilateral shapes in full images, eliminating unnecessary intermediate steps (e.g., candidate aggregation and word partitioning), with a single neural network. The simplicity of our pipeline allows concentrating efforts on designing loss functions and neural network architecture. Experiments on standard datasets including ICDAR 2015, COCO-Text and MSRA-TD500 demonstrate that the proposed algorithm significantly outperforms state-of-the-art methods in terms of both accuracy and efficiency. On the ICDAR 2015 dataset, the proposed algorithm achieves an F-score of 0.7820 at 13.2fps at 720p resolution.
409 citations
Pennsylvania State University1, California Institute of Technology2, University of Wisconsin–Milwaukee3, University of Tokyo4, University of Toronto5, Albert Einstein Institution6, The Chinese University of Hong Kong7, Northwestern University8, Goddard Space Flight Center9, University of Birmingham10, Kenyon College11
TL;DR: In this article, a stream-based analysis pipeline was proposed to detect gravitational waves from the merger of binary neutron stars, binary black holes, and neutron-star-black-hole binaries within ∼1 min of the arrival of the merger signal at Earth.
Abstract: We describe a stream-based analysis pipeline to detect gravitational waves from the merger of binary neutron stars, binary black holes, and neutron-star–black-hole binaries within ∼1 min of the arrival of the merger signal at Earth. Such low-latency detection is crucial for the prompt response by electromagnetic facilities in order to observe any fading electromagnetic counterparts that might be produced by mergers involving at least one neutron star. Even for systems expected not to produce counterparts, low-latency analysis of the data is useful for deciding when not to point telescopes, and as feedback to observatory operations. Analysts using this pipeline were the first to identify GW151226, the second gravitational-wave event ever detected. The pipeline also operates in an offline mode, in which it incorporates more refined information about data quality and employs acausal methods that are inapplicable to the online mode. The pipeline’s offline mode was used in the detection of the first two gravitational-wave events, GW150914 and GW151226, as well as the identification of a third candidate, LVT151012.
313 citations
11 Jan 2017
TL;DR: A fruit counting pipeline based on deep learning that accurately counts fruit in unstructured environments and is able to perform well even on highly occluded fruits that are challenging for human labelers to annotate is described.
Abstract: This paper describes a fruit counting pipeline based on deep learning that accurately counts fruit in unstructured environments. Obtaining reliable fruit counts is challenging because of variations in appearance due to illumination changes and occlusions from foliage and neighboring fruits. We propose a novel approach that uses deep learning to map from input images to total fruit counts. The pipeline utilizes a custom crowdsourcing platform to quickly label large data sets. A blob detector based on a fully convolutional network extracts candidate regions in the images. A counting algorithm based on a second convolutional network then estimates the number of fruits in each region. Finally, a linear regression model maps that fruit count estimate to a final fruit count. We analyze the performance of the pipeline on two distinct data sets of oranges in daylight, and green apples at night, utilizing human generated labels as ground truth. We also show that the pipeline has a short training time and performs well with a limited data set size. Our method generalizes across both data sets and is able to perform well even on highly occluded fruits that are challenging for human labelers to annotate.
307 citations
TL;DR: A hierarchical distributed Fog Computing architecture is introduced to support the integration of massive number of infrastructure components and services in future smart cities and demonstrates the feasibility of the system's city-wide implementation in the future.
Abstract: Data intensive analysis is the major challenge in smart cities because of the ubiquitous deployment of various kinds of sensors. The natural characteristic of geodistribution requires a new computing paradigm to offer location-awareness and latency-sensitive monitoring and intelligent control. Fog Computing that extends the computing to the edge of network, fits this need. In this paper, we introduce a hierarchical distributed Fog Computing architecture to support the integration of massive number of infrastructure components and services in future smart cities. To secure future communities, it is necessary to integrate intelligence in our Fog Computing architecture, e.g., to perform data representation and feature extraction, to identify anomalous and hazardous events, and to offer optimal responses and controls. We analyze case studies using a smart pipeline monitoring system based on fiber optic sensors and sequential learning algorithms to detect events threatening pipeline safety. A working prototype was constructed to experimentally evaluate event detection performance of the recognition of 12 distinct events. These experimental results demonstrate the feasibility of the system's city-wide implementation in the future.
284 citations
01 Apr 2017
TL;DR: This paper proposes a novel architecture for implementing Winograd algorithm on FPGAs and proposes an analytical model to predict the resource usage and reason about the performance, and uses the model to guide a fast design space exploration.
Abstract: In recent years, Convolutional Neural Networks (CNNs) have become widely adopted for computer vision tasks. FPGAs have been adequately explored as a promising hardware accelerator for CNNs due to its high performance, energy efficiency, and reconfigurability. However, prior FPGA solutions based on the conventional convolutional algorithm is often bounded by the computational capability of FPGAs (e.g., the number of DSPs). In this paper, we demonstrate that fast Winograd algorithm can dramatically reduce the arithmetic complexity, and improve the performance of CNNs on FPGAs. We first propose a novel architecture for implementing Winograd algorithm on FPGAs. Our design employs line buffer structure to effectively reuse the feature map data among different tiles. We also effectively pipeline the Winograd PE engine and initiate multiple PEs through parallelization. Meanwhile, there exists a complex design space to explore. We propose an analytical model to predict the resource usage and reason about the performance. Then, we use the model to guide a fast design space exploration. Experiments using the state-of-the-art CNNs demonstrate the best performance and energy efficiency on FPGAs. We achieve an average 1006.4 GOP/s for the convolutional layers and 854.6 GOP/s for the overall AlexNet and an average 3044.7 GOP/s for the convolutional layers and 2940.7 GOP/s for the overall VGG16 on Xilinx ZCU102 platform.
233 citations
TL;DR: In this article, a convolutional neural network is used to predict the coefficients of a locally affine model in bilateral space, which is then applied to the full-resolution image.
Abstract: Performance is a critical challenge in mobile image processing. Given a reference imaging pipeline, or even human-adjusted pairs of images, we seek to reproduce the enhancements and enable real-time evaluation. For this, we introduce a new neural network architecture inspired by bilateral grid processing and local affine color transforms. Using pairs of input/output images, we train a convolutional neural network to predict the coefficients of a locally-affine model in bilateral space. Our architecture learns to make local, global, and content-dependent decisions to approximate the desired image transformation. At runtime, the neural network consumes a low-resolution version of the input image, produces a set of affine transformations in bilateral space, upsamples those transformations in an edge-preserving fashion using a new slicing node, and then applies those upsampled transformations to the full-resolution image. Our algorithm processes high-resolution images on a smartphone in milliseconds, provides a real-time viewfinder at 1080p resolution, and matches the quality of state-of-the-art approximation techniques on a large class of image operators. Unlike previous work, our model is trained off-line from data and therefore does not require access to the original operator at runtime. This allows our model to learn complex, scene-dependent transformations for which no reference implementation is available, such as the photographic edits of a human retoucher.
163 citations
Posted Content•
TL;DR: In this paper, a simple yet powerful pipeline for medical image segmentation that combines Fully Convolutional Networks (FCNs) with Fully convolutional residual networks (FC-ResNets) is presented.
Abstract: In this paper, we introduce a simple, yet powerful pipeline for medical image segmentation that combines Fully Convolutional Networks (FCNs) with Fully Convolutional Residual Networks (FC-ResNets). We propose and examine a design that takes particular advantage of recent advances in the understanding of both Convolutional Neural Networks as well as ResNets. Our approach focuses upon the importance of a trainable pre-processing when using FC-ResNets and we show that a low-capacity FCN model can serve as a pre-processor to normalize medical input data. In our image segmentation pipeline, we use FCNs to obtain normalized images, which are then iteratively refined by means of a FC-ResNet to generate a segmentation prediction. As in other fully convolutional approaches, our pipeline can be used off-the-shelf on different image modalities. We show that using this pipeline, we exhibit state-of-the-art performance on the challenging Electron Microscopy benchmark, when compared to other 2D methods. We improve segmentation results on CT images of liver lesions, when contrasting with standard FCN methods. Moreover, when applying our 2D pipeline on a challenging 3D MRI prostate segmentation challenge we reach results that are competitive even when compared to 3D methods. The obtained results illustrate the strong potential and versatility of the pipeline by achieving highly accurate results on multi-modality images from different anatomical regions and organs.
160 citations
01 Jul 2017
TL;DR: The proposed pipeline achieves state of the art accuracy on the largest and most competitive stereo benchmarks, and the learned confidence is shown to outperform all existing alternatives.
Abstract: We present an improved three-step pipeline for the stereo matching problem and introduce multiple novelties at each stage. We propose a new highway network architecture for computing the matching cost at each possible disparity, based on multilevel weighted residual shortcuts, trained with a hybrid loss that supports multilevel comparison of image patches. A novel post-processing step is then introduced, which employs a second deep convolutional neural network for pooling global information from multiple disparities. This network outputs both the image disparity map, which replaces the conventional winner takes all strategy, and a confidence in the prediction. The confidence score is achieved by training the network with a new technique that we call the reflective loss. Lastly, the learned confidence is employed in order to better detect outliers in the refinement step. The proposed pipeline achieves state of the art accuracy on the largest and most competitive stereo benchmarks, and the learned confidence is shown to outperform all existing alternatives.
155 citations
TL;DR: This work built a comprehensive spike-sorting pipeline that performs reliably under noise and probe drift by incorporating covariance-based features and unsupervised clustering based on fast density-peak finding and validated performance using multiple ground-truth datasets that recently became available.
Abstract: Electrical recordings from a large array of electrodes give us access to neural population activity with single-cell, single-spike resolution. These recordings contain extracellular spikes which must be correctly detected and assigned to individual neurons. Despite numerous spike-sorting techniques developed in the past, a lack of high-quality ground-truth datasets hinders the validation of spike-sorting approaches. Furthermore, existing approaches requiring manual corrections are not scalable for hours of recordings exceeding 100 channels. To address these issues, we built a comprehensive spike-sorting pipeline that performs reliably under noise and probe drift by incorporating covariance-based features and unsupervised clustering based on fast density-peak finding. We validated performance of our workflow using multiple ground-truth datasets that recently became available. Our software scales linearly and processes up to 1000-channel recording in real-time using a single workstation. Accurate, real-time spike sorting from large recording arrays will enable more precise control of closed-loop feedback experiments and brain-computer interfaces.
16 Jun 2017
TL;DR: A novel inference software pipeline that targets the local execution of multiple deep vision models (specifically, CNNs) by interleaving the execution of computation-heavy convolutional layers with the loading of memory-heavy fully-connected layers.
Abstract: Wearable devices with built-in cameras present interesting opportunities for users to capture various aspects of their daily life and are potentially also useful in supporting users with low vision in their everyday tasks. However, state-of-the-art image wearables available in the market are limited to capturing images periodically and do not provide any real-time analysis of the data that might be useful for the wearers. In this paper, we present DeepEye - a match-box sized wearable camera that is capable of running multiple cloud-scale deep learn- ing models locally on the device, thereby enabling rich analysis of the captured images in near real-time without offloading them to the cloud. DeepEye is powered by a commodity wearable processor (Snapdragon 410) which ensures its wearable form factor. The software architecture for DeepEye addresses a key limitation with executing multiple deep learning models on constrained hardware, that is their limited runtime memory. We propose a novel inference software pipeline that targets the local execution of multiple deep vision models (specifically, CNNs) by interleaving the execution of computation-heavy convolutional layers with the loading of memory-heavy fully-connected layers. Beyond this core idea, the execution framework incorporates: a memory caching scheme and a selective use of model compression techniques that further minimizes memory bottlenecks. Through a series of experiments, we show that our execution framework outperforms the baseline approaches significantly in terms of inference latency, memory requirements and energy consumption.
07 Sep 2017
TL;DR: A novel, accurate tightly-coupled visual-inertial odometry pipeline for event cameras that leverages their outstanding properties to estimate the camera ego-motion in challenging conditions, such as high-speed motion or high dynamic range scenes.
Abstract: Event cameras are bio-inspired vision sensors that output pixel-level brightness changes instead of standard intensity frames. They offer significant advantages over standard cameras, namely a very high dynamic range, no motion blur, and a latency in the order of microseconds. We propose a novel, accurate tightly-coupled visual-inertial odom- etry pipeline for such cameras that leverages their outstanding properties to estimate the camera ego-motion in challenging conditions, such as high-speed motion or high dynamic range scenes. The method tracks a set of features (extracted on the image plane) through time. To achieve that, we consider events in overlapping spatio-temporal windows and align them using the current camera motion and scene structure, yielding motion-compensated event frames. We then combine these feature tracks in a keyframe- based, visual-inertial odometry algorithm based on nonlinear optimization to estimate the camera’s 6-DOF pose, velocity, and IMU biases. The proposed method is evaluated quantitatively on the public Event Camera Dataset [19] and significantly outperforms the state-of-the-art [28], while being computationally much more efficient: our pipeline can run much faster than real-time on a laptop and even on a smartphone processor. Fur- thermore, we demonstrate qualitatively the accuracy and robustness of our pipeline on a large-scale dataset, and an extremely high-speed dataset recorded by spinning an event camera on a leash at 850 deg/s.
TL;DR: PipeCraft is introduced, a flexible and handy bioinformatics pipeline with a user‐friendly graphical interface that links several public tools for analysing amplicon sequencing data and is able to process large data sets within 24 hr.
Abstract: High-throughput sequencing methods have become a routine analysis tool in environmental sciences as well as in public and private sector. These methods provide vast amount of data, which need to be ...
15 Dec 2017
TL;DR: This paper proposed a unified neural network that jointly performs domain, intent, and slot predictions for multi-task learning and achieved significant improvements in all three tasks across all domains over strong baselines.
Abstract: In practice, most spoken language understanding systems process user input in a pipelined manner; first domain is predicted, then intent and semantic slots are inferred according to the semantic frames of the predicted domain. The pipeline approach, however, has some disadvantages: error propagation and lack of information sharing. To address these issues, we present a unified neural network that jointly performs domain, intent, and slot predictions. Our approach adopts a principled architecture for multitask learning to fold in the state-of-the-art models for each task. With a few more ingredients, e.g. orthography-sensitive input encoding and curriculum training, our model delivered significant improvements in all three tasks across all domains over strong baselines, including one using oracle prediction for domain detection, on real user data of a commercial personal assistant.
TL;DR: In this paper, a finite element method was used to predict burst pressure for corroded pipeline by using Monte Carlo Simulation (MCS) and the sensitivity analysis of parameters and model revealed that the corrosion depth and the pipeline operation pressure have the most influence on the pipeline failure probability.
TL;DR: In this article, the authors analyzed H2 compression and pipeline transportation processes with safety issues related to water electrolysis and H2 production for different values of the hydrogen mass flow rate: 02, 05, 10, 20, and 28 kg/s.
29 Aug 2017
TL;DR: Kami is introduced, a Coq library that enables similar expressive and modular reasoning for hardware designs expressed in the style of the Bluespec language, and can specify, implement, and verify realistic designs entirely within Coq, ending with automatic extraction into a pipeline that bottoms out in FPGAs.
Abstract: It has become fairly standard in the programming-languages research world to verify functional programs in proof assistants using induction, algebraic simplification, and rewriting. In this paper, we introduce Kami, a Coq library that enables similar expressive and modular reasoning for hardware designs expressed in the style of the Bluespec language. We can specify, implement, and verify realistic designs entirely within Coq, ending with automatic extraction into a pipeline that bottoms out in FPGAs. Our methodology, using labeled transition systems, has been evaluated in a case study verifying an infinite family of multicore systems, with cache-coherent shared memory and pipelined cores implementing (the base integer subset of) the RISC-V instruction set.
TL;DR: In this article, the optimal scales of time and spatial steps of a newly proposed implicit upwind model and the characteristic line model were studied for fast and accurate prediction of the pipeline outlet temperature within ± 0.5°C.
TL;DR: In this paper, a mixed-integer nonlinear programming model (MINLP) for products pipeline with single source and multiple pump stations has been proposed to schedule delivery and injection of numerous kinds of products.
TL;DR: A multi-objective optimization model in optimizing the operation of natural gas pipeline networks is presented and a set of Pareto optimal points from which a decision maker can select a specific preferred solution is obtained.
TL;DR: In this article, a chipless RFID-based sensor for pipeline integrity monitoring in real-time fashion is presented, which monitors the coating lift-off from the pipeline, which is the initial step in external corrosion of a metal pipe.
Abstract: This work presents a chipless RFID-based sensor for pipeline integrity monitoring in real-time fashion. The sensor monitors the coating lift-off from the pipeline, which is the initial step in external corrosion of a metal pipe. The sensor has a readout coil and an LC resonator on the passive tag with an interdigitated capacitor. The resonant frequency of the sensor demonstrates a strong relation to the gap between the coating and the metal pipe. The tag is built on a flexible substrate for wrapping around the pipe and to represent the pipe coating. The sensor is conformal, battery-free, and low cost which makes it suitable for pipeline monitoring in harsh environments. The resonator was tuned to 105 MHz with a Q factor of ∼115. The sensor demonstrates the maximum resonant frequency change of 11.7%, when 2 Standard Cubic Centimeter per Minute (SCCM) of air lifts the coating; and 7.46% when 4 mL of water ingress happens in between the coating and the pipe. This sensor has advantages of inexpensiveness, simplicity, and long lifetime with potential capability of prediction prior to failure in pipeline systems.
01 Oct 2017
TL;DR: The first photometric registration pipeline for Mixed Reality based on high quality illumination estimation using convolutional neural networks (CNNs) is presented, and experimental results show that the proposed method yields highly accurate estimates for photo-realistic augmentations.
Abstract: This paper presents the first photometric registration pipeline for Mixed Reality based on high quality illumination estimation using convolutional neural networks (CNNs). For easy adaptation and deployment of the system, we train the CNNs using purely synthetic images and apply them to real image data. To keep the pipeline accurate and efficient, we propose to fuse the light estimation results from multiple CNN instances and show an approach for caching estimates over time. For optimal performance, we furthermore explore multiple strategies for the CNN training. Experimental results show that the proposed method yields highly accurate estimates for photo-realistic augmentations.
TL;DR: This study developed a modified electromechanical impedance technique for crack detection that involves fusing information from multiple sensors and derived a new damage-sensitive feature factor based on a pipeline EMI model that considers the influence of the bonding layer between the EMI sensors and pipeline.
Abstract: An extensive network of pipeline systems is used to transport and distribute national energy resources that heavily influence a nation's economy. Therefore, the structural integrity of these pipeline systems must be monitored and maintained. However, structural damage detection remains a challenge in pipeline engineering. To this end, this study developed a modified electromechanical impedance (EMI) technique for crack detection that involves fusing information from multiple sensors. We derived a new damage-sensitive feature factor based on a pipeline EMI model that considers the influence of the bonding layer between the EMI sensors and pipeline. We experimentally validated the effectiveness of the proposed method. Finally, we used a damage index—root mean square deviation—to examine the degree and position of crack damage in a pipeline.
TL;DR: Seqping provides researchers a seamless pipeline to train species-specific HMMs and predict genes in newly sequenced or less-studied genomes, and is able to generate better gene predictions compared to three HMM-based programs using their respective available HMMs.
Abstract: Gene prediction is one of the most important steps in the genome annotation process. A large number of software tools and pipelines developed by various computing techniques are available for gene prediction. However, these systems have yet to accurately predict all or even most of the protein-coding regions. Furthermore, none of the currently available gene-finders has a universal Hidden Markov Model (HMM) that can perform gene prediction for all organisms equally well in an automatic fashion. We present an automated gene prediction pipeline, Seqping that uses self-training HMM models and transcriptomic data. The pipeline processes the genome and transcriptome sequences of the target species using GlimmerHMM, SNAP, and AUGUSTUS pipelines, followed by MAKER2 program to combine predictions from the three tools in association with the transcriptomic evidence. Seqping generates species-specific HMMs that are able to offer unbiased gene predictions. The pipeline was evaluated using the Oryza sativa and Arabidopsis thaliana genomes. Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis showed that the pipeline was able to identify at least 95% of BUSCO’s plantae dataset. Our evaluation shows that Seqping was able to generate better gene predictions compared to three HMM-based programs (MAKER2, GlimmerHMM and AUGUSTUS) using their respective available HMMs. Seqping had the highest accuracy in rice (0.5648 for CDS, 0.4468 for exon, and 0.6695 nucleotide structure) and A. thaliana (0.5808 for CDS, 0.5955 for exon, and 0.8839 nucleotide structure). Seqping provides researchers a seamless pipeline to train species-specific HMMs and predict genes in newly sequenced or less-studied genomes. We conclude that the Seqping pipeline predictions are more accurate than gene predictions using the other three approaches with the default or available HMMs.
TL;DR: This work reimplemented the Chemogenomics pipeline using Apache Spark, which enabled it to lift the existing programs to a multi-node cluster without making changes to the predictors.
TL;DR: In this pursuit, it is found that the multi-core and multiple processor architectures are promising candidates for multi-GNSS implementation, especially the underlying baseband and software realization platform.
Abstract: Receiver design challenges arising from new GNSS signals include required intermediate frequency, sampling rate, modulation type, spreading code, and secondary code. Several architectures are examined here aiming at a best model for multi-GNSS implementation, especially the underlying baseband and software realization platform. In this pursuit, it is found that the multi-core and multiple processor architectures are promising candidates. General purpose processors or digital signal processors demand excessive resources and power consumption. Alternative architectures are presented along with the general cost function, used to evaluate architecture efficiency. Taking into account (1) the superiority of a hardware time-interleaving technique, (2) RAM-based design versus register-based design, and (3) careful consideration of modern GNSS signal attributes, the proposed programmable custom pipeline correlator core provides flexibility and significantly reduces resources and power.
TL;DR: This manuscript describes an efficient, reliable pipeline for spike sorting on dense multi-electrode arrays (MEAs), where neural signals appear across many electrodes and spike sorting currently represents a major computational bottleneck.
Abstract: Spike sorting is a critical first step in extracting neural signals from large-scale electrophysiological data. This manuscript describes an efficient, reliable pipeline for spike sorting on dense multi-electrode arrays (MEAs), where neural signals appear across many electrodes and spike sorting currently represents a major computational bottleneck. We present several new techniques that make dense MEA spike sorting more robust and scalable. Our pipeline is based on an efficient multi-stage "triage-then-cluster-then-pursuit" approach that initially extracts only clean, high-quality waveforms from the electrophysiological time series by temporarily skipping noisy or "collided" events (representing two neurons firing synchronously). This is accomplished by developing a neural network detection method followed by efficient outlier triaging. The clean waveforms are then used to infer the set of neural spike waveform templates through nonparametric Bayesian clustering. Our clustering approach adapts a "coreset" approach for data reduction and uses efficient inference methods in a Dirichlet process mixture model framework to dramatically improve the scalability and reliability of the entire pipeline. The "triaged" waveforms are then finally recovered with matching-pursuit deconvolution techniques. The proposed methods improve on the state-of-the-art in terms of accuracy and stability on both real and biophysically-realistic simulated MEA data. Furthermore, the proposed pipeline is efficient, learning templates and clustering much faster than real-time for a 500-electrode dataset, using primarily a single CPU core.
TL;DR: In this article, a comprehensive pipeline flow model with gas composition tracking resulting from coupling of mass and chemical energy transport models has been developed to study the effect of the variation in gas composition on the operation strategy of the pipeline system.