Search

This work focuses on automatic grading and revision feedback for open-ended (OE) student responses relating to mathematics and science topics. The dissertation will be divided into three closely related chapters, where each build upon the work of the previous chapter(s). In the first chapter, we focus on improving the performance of an Automatic Short Answer Grading (ASAG) model, where performance is measured by how closely the machine ratings match those of a human rater. We experiment with incorporating domain-related text into the model with sources such as the item text, context, and rubric, and investigate how this might affect the generalizability of the model by testing the ASAG model on out-of-training-sample questions. Further, we examine if representing not only the text, but also the ordinal structure of a scoring rubric is useful by using a graph sampling method prior to incorporating the text into the ASAG model. In the second chapter, we address the lack of explainability in typical ASAG models. We train a deep reinforcement learning (RL) agent to revise a student’s response by inserting key phrases, or deleting portions of the response, to achieve a high grade from an ASAG model in the least number of revisions. The intent is that the agent’s revisions may provide some explanation as to what was missing or unnecessarily included in a student’s response. We also examine the intelligibility of a Neural Additive Model (NAM) using the RL agent’s key phrases as input features. NAMs provide the explainability of additive models with the ability to approximate high dimensional functions and inputs like a neural network. Further, we use the RL agent to expose shortcomings in the ASAG model by finding revisions that achieve a high grade from the automatic grader but may not be considered good responses by a human rater. In the third chapter, we focus on whether the RL agent’s revisions can provide feedback to encourage a student to revise their own short response. We collected data with a randomized controlled design from students interacting with an online tutoring system. Students who receive an intervention from the RL agent’s revisions will be compared with both students who receive an intervention from a popular open-source chatbot, ChatGPT, and students who received a generic intervention. We examine student perceptions of the feedback, how prior knowledge interacts with the interventions, and the accuracy of ChatGPT’s feedback.

Automatic speech recognition (ASR) systems have improved significantly in the last decade due to advances in deep learning algorithms and easier access to very large databases. ASR systems, however, face two major challenges. The first challenge is in low-resource situations, such as child speech, where the accuracy degrades significantly, and the second challenge concerns low inference efficiency due to the autoregressive mechanism and size of ASR models. In this dissertation, we address these challenges by introducing novel techniques to improve the accuracy and the inference efficiency for ASR tasks, especially child ASR. To address the accuracy challenge, we introduce novel self-supervised learning (SSL) methods using un-annotated adult speech data and explore how these methods can improve the downstream child ASR tasks. Specifically, a bidirectional autoregressive predictive coding (Bi-APC) method is proposed for non-causal models pretraining with the usage of adult speech data. The pretrained model is then finetuned on supervised child speech-text pairs. We also propose a novel framework, domain responsible adaptation and finetuning (DRAFT), to reduce the domain shifting in pretrained speech models. The DRAFT framework is effective for APC that uses a causal transformer as the backbone, and for Bi-APC, Wav2vec2.0 and HuBERT methods, which use a non-causal transformer as the backbone. To address the inference efficiency challenge, we introduce a novel Connectionist Temporal Classification (CTC) Alignment-based Single-Step Non-Autoregressive Transformer (CASS-NAT) for end-to-end ASR. A comprehensive evaluation of CASS-NAT is performed in this dissertation. In CASS-NAT, the word embeddings in an autoregressive transformer (AT) are substituted with token-level acoustic embeddings (TAE) that are extracted based on the encoder outputs and CTC alignments. TAE can be obtained simultaneously without recurrent operations, leading to a parallel generation of output sequences. In addition, an error-based alignment sampling method is proposed to reduce alignment mismatch between the training and inference. CASS-NAT achieves ~20x speed up during inference without significant performance degradation compared to AT. We also propose a CASS-NAT variant (UniEnc-CASSNAT) that consists of only an encoder module. Together with the proposed multi-pass CTC training and iterative decoding, UniEnc-CASSNAT can perform as well as CASS-NAT with fewer model parameters. Beyond these two challenges, and in order to facilitate the development of better child ASR, we build the first child ASR benchmark for the research community. The benchmark includes comparisons of widely-used techniques for ASR such as data augmentation, parameter efficient finetuning (PEFT), self-supervised (HuBERT and WavLM) and supervised models (such as Whisper). All codes developed in this dissertation will be available.

Starting from JDK 9, the Java Platform Module System (JPMS) was introduced to facilitate resolving the monolithic ball-of-mud architecture of the original JDK by offering architectural modules in the Java language that provide various benefits for Java projects, especially in terms of protecting its internals from dangerous static or dynamic usage. However, a recent study has shown evidence that many existing projects bypass Java module boundaries to access internal APIs which, in turn, break the strong encapsulation (BSE) of Java modules. BSE leads to exceptions, errors, and other maintenance issues for Java systems. To address BSE, we propose a detection tool, BEAD, that leverages static analysis to identify BSE abuse instances within Java projects and evaluate BEAD on open-source Java projects. Our analysis revealed that JDK modules are designed with strong encapsulation principles, but BEAD detected numerous abuse instances, predominantly at compile-time, indicating prevalent unauthorized access to internal packages. Besides, our analysis found evidence of a correlation between the detected abuse instances and reported GitHub issues, validating the ecosystem impact of those detected abuse instances. Our findings demonstrate BEAD's effectiveness of detection capabilities in terms of BSE problem.

Some of the major results of future goals of an automatic program synthesis project are described in the two papers that comprise this document. The first paper gives a detailed algorithm for synthesizing a computer program from a trace of its behavior. Since the algorithm involves a search, the length of time required to do the synthesis of nontrivial programs can be quite large. Techniques are given for preprocessing the trace information to reduce enumeration, for pruning the search using a failure memory technique, and for utilizing multiple traces to the best advantage. The results of numerous tests are given to demonstrate the value of the techniques. The other paper gives a brief overview of the automatic programming system within which the above algorithm is being used. This system is still under development. (Author/CH)

This review aims to summarize and describe research on the topic of automatic group emotion recognition. In recent years, the topic of emotion analysis of groups or crowds has gained interest, with studies performing emotion detection in different contexts, using different datasets and modalities (such as images, video, audio, social media messages), and taking different approaches. Articles are included after an innovative search method, including Dense Query Extraction and automatic cross-referencing. Discussed are the types of groups and emotion models considered in automatic emotion recognition research, common datasets for all modalities, general approaches taken, and reported performances. These performances are discussed, followed by an analysis of the application possibilities of the discussed methods. To ensure clear, replicable, and comparable studies, we suggest research should test on multiple, common datasets and report on multiple metrics, when possible. Implementation details and code should be made available where possible. An area of interest for future work is to build systems with more real-world application possibilities, coping with changing group sizes, different emotional subgroups, and changing emotions over time, while having a higher robustness and working with datasets with reduced biases.

A systolic array architecture consists of a grid of simple processing elements (PE) connected through local interconnects. With a massive number of PEs and a local interconnection, such an architecture is capable of achieving high performance and energy efficiency. My dissertation focuses on extending the research of systolic array architecture in two fields: automatic systolic array synthesis and architecture optimization.The first part of the dissertation focuses on the automated systolic array synthesis. Designing high-performance systolic arrays requires an understanding of both the application characteristics and hardware architecture, requiring non-trivial efforts to reap its benefits. There exists a large body of past works on developing compilation frameworks for systolic arrays. However, these works failed to reach a balance between the generality, performance, and productivity, making them hard to use in practice. Our work advances this field by leveraging two compilation technologies, the polyhedral model and high-level synthesis (HLS). We propose a new compilation framework, AutoSA, which is built on the polyhedral framework and is capable of generating high-performance systolic arrays on FPGA in HLS languages. We show that AutoSA can handle applications with complex dependence structures and generate designs with performance comparable to or better than manual designs.AutoSA incorporates a broad set of hardware optimization techniques that open up a vast design space which is intractable to explore manually. To cope with this challenge, we propose an efficient auto-tuning framework, Odyssey, which finds optimal designs within seconds. With both AutoSA and Odyssey, we reduce the development cycles of systolic arrays from weeks to days, which significantly boosts the productivity compared to the prior works.In the second part of the dissertation, we present two application optimization studies that deploy systolic arrays for various applications and platforms.The first study investigates the architecture trade-offs when using systolic arrays for one important application, convolutional neural network (CNN). The results show that a single monolithic systolic array is insufficient to handle the divergent characteristics of different CNN layers. Therefore, we further explore the use of a multi-array architecture that implements several smaller systolic arrays with different configurations customized for each CNN layer. Multi-array systems help improve the throughput with a cost of longer latency. This work reveals the complexities and trade-offs when mapping a real-world application to systolic arrays.In addition to FPGA, systolic arrays can also be mapped to GPU as an overlay above the existing GPU architecture. The second work investigates the performance trade-offs when mapping systolic arrays to GPU. We achieve a performance speedup by leveraging the shuffle instructions on Nvidia GPUs to implement the inter-PE communication compared to baselines using the shared memory.Systolic array architecture plays an important role in the post-Moore's law era as one architecture candidate capable of delivering high performance and energy efficiency. The works presented in this dissertation provide comprehensive and efficient solutions to lowering the programming efforts and optimizing the performance of this architecture. We hope the promising results from these works will open the door to more deployment cases of systolic arrays in a broader range of applications and hardware platforms in the future.

Using automatic identification system (AIS) data, this article first has extended the definition of three widely used roadway congestion indices to maritime transportation systems (MTS), traffic speed index (TSI), traffic rate index (TRI), and dwell time index (DTI). Next, a methodology is developed to measure the indices based on AIS data, considering various factors, including path geometry, time of day, and the type and size of vessels, and finally the method has been applied to the AIS data of the Houston Ship Channel (HSC) to evaluate the applicability in real cases. The results show that although average TSI and TRI cannot represent waterway congestion, the real-time values (rather than the average) at the micro level can help finding location, time, and severity of traffic congestion. Besides, while TSI and TRI have shortcomings, both average and real-time dwell time index (DTI) can quantify traffic congestion and highlight severity in different waterway segments for different types of vessels. When congestion happens at some narrow waterways, vessels need to wait at sea buoy or docks, thus dwell time index (DTI) can quantify traffic congestion better than in-transit indices such as travel speed, TSI. According to HSC DTI, most tankers experience long waiting times at the sea buoy and Galveston Bay, while cargo vessels experience delays at Bayport and Barbour's Cut terminals. This paper helps the decision-makers quantify congestion in different sections of a waterway and provides measures to compare congestion for national competing projects at different waterways.

This second volume of a two-volume report on a fully automatic high quality translation (FAHQT) contains relevant papers contributed by specialists on the topic of machine translation. The papers presented here cover such topics as syntactical analysis in transformational grammar and in machine translation, lexical features in translation and paraphrasing, requirements for machine translation, current status of hardware and software as it affects FAHQT, bilingual computer dictionaries, and the shape of the dictionary for machine translation. Volume 1 (FL 002 936) includes papers as well as specific consideration of the FAHQT inquiry. (VM)

Low-power consumption and long-distance transmission are two problems that have to be solved by the application of broadband power line communication for the automatic meter reading system. To reduce the power consumption of the communication module, based on the analysis of the composition of the power consumption, some methods are proposed. From the communication chip level and the module circuit level, the design scheme of low-power consumption is given. To solve the problem of transmission distance, a frequency band of 2.44 MHz~5.6 MHz is used as the main working frequency band. The communication module supports multiple frequency bands. Using this feature, the optimal frequency band is adaptively selected for communication and automatic switching, which further improve the transmission distance. Field application shows that the above methods effectively decrease the power consumption of the communication module and extend the transmission distance.

Specifying utility functions is a key step towards applying the discrete choice framework for understanding the behaviour processes that govern user choices. However, identifying the utility function specifications that best model and explain the observed choices can be a very challenging and time-consuming task. This paper seeks to help modellers by leveraging the Bayesian framework and the concept of automatic relevance determination (ARD), in order to automatically determine an optimal utility function specification from an exponentially large set of possible specifications in a purely data-driven manner. Based on recent advances in approximate Bayesian inference, a doubly stochastic variational inference is developed, which allows the proposed DCM-ARD model to scale to very large and high-dimensional datasets. Using semi-artificial choice data, the proposed approach is shown to very accurately recover the true utility function specifications that govern the observed choices. Moreover, when applied to real choice data, DCM-ARD is shown to be able discover high quality specifications that can outperform previous ones from the literature according to multiple criteria, thereby demonstrating its practical applicability., 21 pages, 2 figures, 11 tables

This paper presents a simple control method for the fully automatic parking of an unmanned vehicle. This method is based on the switching control algorithm and backstepping theory. The fully automatic parking is hard to accomplish since it cannot converge to a specified norm, guarantee the convergence rate, and large uncertainties. Global exponential convergence to any prescribed norm bound (e-convergence) is guaranteed, and the convergence rate is explicitly given. For the design of steering laws, a method based on backstepping is proposed and analyzed in detail. The backstepping-based design of a multichain system is obtained. Moreover, an exponentially e-convergent control algorithm is adopted to guarantee both converge norm and convergence rate. Real road experiment results are presented and the results show the effectiveness of the control strategies.

The scientific research process generally starts with the examination of the state of the art, which may involve a vast number of publications. Automatically summarizing scientific articles would help researchers in their investigation by speeding up the research process. The automatic summarization of scientific articles differs from the summarization of generic texts due to their specific structure and inclusion of citation sentences. Most of the valuable information in scientific articles is presented in tables, figures, and algorithm pseudocode. These elements, however, do not usually appear in a generic text. Therefore, several approaches that consider the particularity of a scientific article structure were proposed to enhance the quality of the generated summary, resulting in ad hoc automatic summarizers. This paper provides a comprehensive study of the state of the art in this field and discusses some future research directions. It particularly presents a review of approaches developed during the last decade, the corpora used, and their evaluation methods. It also discusses their limitations and points out some open problems. The conclusions of this study highlight the prevalence of extractive techniques for the automatic summarization of single monolingual articles using a combination of statistical, natural language processing, and machine learning techniques. The absence of benchmark corpora and gold standard summaries for scientific articles remains the main issue for this task.

The detection of objects on the seafloor is a complex task. The domain of the detection and classification of naval mines is additionally complicated by the high risk nature of the task. Autonomous underwater vehicles (AUVs) have been used in naval mine countermeasures (MCM) operations to search large areas using sensors such as sidescan or synthetic aperture sonars. These sensors generally have a high coverage rate, while sacrificing spatial resolution. Conversely, sensors with higher resolution but lower coverage (such as forward-looking sonars and electro-optical cameras) are employed for the later classification and identification stages of the MCM mission. However, to autonomously execute a target reacquisition mission, it is important to be able to collect and process data automatically and, in near real time, onboard an AUV. For this purpose, an automatic target recognition (ATR) system is required. This article proposes an ATR, which can be used onboard an autonomous vehicle, capable of detecting mine-like objects in forward-looking sonar data. The ATR combines a detector and a classifier, based on convolutional neural network models, with a probabilistic grid map that filters out false positives and combines reported detections at nearby locations. A strategy, combining a survey pattern with target-mapping maneuvers automatically activated by the ATR, has been designed to maximize the performance of this ATR. The whole system has been tested in simulation as well as using data from previous MCM exercises, the results of which are presented here This document is the results of the research project funded by the EU project Open Cooperation for European mAritime awareNess (OCEAN2020)

The day-to-day growth of online information necessitates intensive research in automatic text summarization (ATS). The ATS software produces summary text by extracting important information from the original text. With the help of summaries, users can easily read and understand the documents of interest. Most of the approaches for ATS used only local properties of text. Moreover, the numerous properties make the sentence selection difficult and complicated. So this article uses a graph based summarization to utilize structural and global properties of text. It introduces maximal clique based sentence selection (MCBSS) algorithm to select important and non-redundant sentences that cover all concepts of the input text for summary. The MCBSS algorithm finds novel information using maximal cliques (MCs). The experimental results of recall oriented understudy for gisting evaluation (ROUGE) on Timeline dataset show that the proposed work outperforms the existing graph algorithms Bushy Path (BP), Aggregate Similarity (AS), and TextRank (TR).

Automatic landing of aircraft is a challenging guidance and control task that requires multiple feedback loops and a precise sequence of actions. Increasing operational flexibility and reducing the...

Current pulses are convenient to be actively implemented by a Battery Management System (BMS). However, the Short-Term Features (STF) from currentpulses originate from various sensors with uneven qualities, which hinders one powerful and strong learner with STF for the battery SOH estimation. This paper thus proposes an optimized weak learner formulation procedure for Lithium-ion (Li-ion) battery SOH estimation, which further enables the automatic initialization and integration of the weak learners with STF into an efficient SOH estimation framework. A Pareto Front-based Selection Strategy (PFSS) is designed to select the representative solutions from the non-dominated solutions fed by a Knee point driven Evolutionary Algorithm (KnEA), which guarantees both the diversity and accuracy of the weak learners. Afterwards, the weak learners, whose coefficients are obtained by Selfadaptive Differential Evolution (SaDE), are integrated by a weight-based structure. The proposed method utilizes the weak learners with STF to boost the overall performance of SOH estimation. The validation of the proposed method is proved by LiFePO4/C batteries under accelerated cycling ageing test including one mission profile providing Primary Frequency Regulation (PFR) service to the grid and one constant current profile.

With the maturity of synthetic aperture radar (SAR) technology, the problem of imbalanced data has gradually emerged. This problem makes it difficult for the automatic target recognition (ATR) model to properly learn the classification boundaries of majority and minority category target samples. In this article, we propose an ATR model with new architecture, called the cost-sensitive awareness-based automatic target recognition (CA-ATR) model, which provides an effective way of solving the problem of imbalanced data. Aimed at the two issues caused by imbalanced data on ATR models, the proposed method solves the problems from both the data and algorithm levels. At the data level, CA-ATR avoids adverse correlations among the target samples through different oversampling methods. By making the ATR model cost-sensitive, the proposed method also avoids the empirical risk preference of the ATR model for majority category target samples at the algorithm-level. At the same time, CA-ATR can autonomously learn different cost-sensitive awareness from different imbalanced data sets. The awareness enables the ATR model to more accurately learn the classification boundaries between target samples that belong in different categories. Several experimental results show the superiority of the proposed approach based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. Compared with other imbalanced learning methods, the proposed method is able to solve different types of imbalanced data problems.

Automatic target recognition has been one of the hottest research in synthetic aperture radar (SAR) data processing. Noticing that popular recognition methods cannot utilize multiple features of SAR complex data, a method fused scattering center feature and deep convolutional neural network (CNN) feature is proposed in this letter. This method contains three key parts, namely, scattering center extraction and reconstruction block, CNN feature extraction block, and final feature fusion and classification block. In this process, the scattering center feature and CNN feature are fused at the level of feature maps, which retain the space information of 2-D feature maps. What is more, the proposed half end-to-end strategy realizes the automatic update of weighting parameters in feature extraction network and subnetwork, which promotes a better recognition efficiency. Experimental results on measured SAR data show that the proposed method can achieve better accuracy than other single feature-based methods and feature fusion methods.

Communication plays a vital role in incorporating smartness into the interconnected power system. However, historical records prove that the data transfer has always been vulnerable to cyber-attacks. Unless these cyber-attacks are identified and cordoned off, they may lead to black-out and result in national security issues. This paper proposes an optimal two-stage Kalman filter (OTS-KF) for simultaneous state and cyber-attack estimation in automatic generation control (AGC) system. Biases/cyber-attacks are modeled as unknown inputs in the AGC dynamics. Five types of cyber-attacks, i.e., false data injection (FDI), data replay attack, denial of service (DoS), scaling, and ramp attacks, are injected into the measurements and estimated using OTS-KF. As the load variations of each area are seldom available, OTS-KF is reformulated to estimate the states and outliers along with the load variations of the system. The proposed technique is validated on the benchmark two-area, three-area, and five-area power system models. The simulation results under various test conditions demonstrate the efficacy of the proposed filter.

Academic studies on the Automatic Voltage Regulator (AVR) have focused on a linear mathematical model which lacks protective features. A more accurate model of the AVR system includes protective features as described in IEEE standards. The AVR models without protective features, namely limiters, result in less accurate control performance since the outputs of real controllers are always bounded. In the present study, the controller outputs are limited between -0.9pu and 1.0pu, and the upper bound of exciter output is limited by 3.1pu with the direction of IEEE standards. The effect of limiters on the controller performance is investigated. Two controllers with a novel sliding surface function are proposed based on a mathematical model of the AVR and its approximate reduced-order model. The proposed controllers having only two parameters are optimized with improved particle swarm optimization (PSO) algorithm. After optimization, the robustness of the controllers is compared with the results of various operating conditions identical to the previous studies. Although controller inputs and exciter outputs are limited, the maximum overshoot is measured as less than 0.1% at no-load conditions. Nominal time constants of the AVR constituents are perturbed from -50% to 50%, and ±10% load disturbance is applied to the output. The robustness of the controllers against parameter uncertainties is measured with an average overshoot at the output. Another superiority among the reported results in the literature is obtained from the proposed controllers, where the minimum average overshoot rates are obtained. In addition, when ±10% load is applied to the output of the AVR, the proposed controllers generate accurate control inputs to reject the load disturbance successfully. Furthermore, the proposed controllers keep the output within the ±5% band if there is a monotonic change at the output. All the results show that the proposed controllers with the improved PSO have drawn the best performance from the perspective of time-domain specifications in comparison with the recently reported controllers.

We propose a universal method for the evaluation of generalized standard materials that greatly simplifies the material law implementation process. By means of automatic differentiation and a numerical integration scheme, AutoMat reduces the implementation effort to two potential functions. By moving AutoMat to the GPU, we close the performance gap to conventional evaluation routines and demonstrate in detail that the expression level reverse mode of automatic differentiation as well as its extension to second order derivatives can be applied inside CUDA kernels. We underline the effectiveness and the applicability of AutoMat by integrating it into the FFT-based homogenization scheme of Moulinec and Suquet and discuss the benefits of using AutoMat with respect to runtime and solution accuracy for an elasto-viscoplastic example., 28 pages, 15 figures, 7 tables; new layout, more detailed proof of Theorem 1

Automatic modulation classification (AMC) is a process of automatic detection of modulation format imposed on the received signal with no prior information (carrier, signal power, phase offset) of the signal, also known as blind classification. In this paper, we proposed a new AMC algorithm, by combining the synergy of the meta-heuristic technique with Gabor feature extraction mainly used in texture analysis. Gabor filters are used to extract the features that are further optimized using the cuckoo search algorithm to increase the efficiency of the classification procedure. The classification approach is applied on digitally modulated signals having phase-shift keying, frequency-shift keying, and quadrature amplitude modulation schemes of order 2-64 over the nonfading channel (AWGN) and fading channel (Rayleigh). Simulations and performance comparison with the existing literature validate that the proposed solution has better classification accuracy with lower sample size and lower signal-to-noise ratio.

Text summarization automatically produces a summary containing important sentences and includes all relevant important information from the original document. One of the main approaches, when viewed from the summary results, are extractive and abstractive. An extractive summary is heading towards maturity and now research has shifted towards abstractive summation and real-time summarization. Although there have been so many achievements in the acquisition of datasets, methods, and techniques published, there are not many papers that can provide a broad picture of the current state of research in this field. This paper provides a broad and systematic review of research in the field of text summarization published from 2008 to 2019. There are 85 journal and conference publications which are the results of the extraction of selected studies for identification and analysis to describe research topics/trends, datasets, preprocessing, features, techniques, methods, evaluations, and problems in this field of research. The results of the analysis provide an in-depth explanation of the topics/trends that are the focus of their research in the field of text summarization; provide references to public datasets, preprocessing and features that have been used; describes the techniques and methods that are often used by researchers as a comparison and means for developing methods. At the end of this paper, several recommendations for opportunities and challenges related to text summarization research are mentioned.

Automatic exposure control (AEC) is used to optimize the X-ray tube output during computed tomography (CT) scans. However, calculation of the tube current by AEC can be affected when a patient is not aligned with the rotational center of the X-ray tube. An automatic couch height-positioning compensation mechanism provides a corrective function when the patient is off-center. In this study, we aimed to (a) evaluate the performance characteristics of the positioning compensation mechanism and (b) confirm whether our proposed compensation method can be properly applied to a noise-based AEC system even if the CT device is not equipped with a positioning compensation mechanism. An elliptical phantom was scanned at various table heights on systems without/with the positioning compensation mechanism. Expressions describing the offset from the gantry’s isocenter and adjusted standard deviation settings were derived and used in our proposed compensation method. A phantom was scanned at various table heights with our proposed compensation method, and volume CT dose index (CTDIvol) and image noise levels were obtained. An anthropomorphic chest phantom was also scanned using the proposed compensation method to verify its accuracy. When the positioning compensation mechanism was used, it yielded a constant CTDIvol and image noise levels at various table heights tested. A comparison between our proposed method and the positioning compensation mechanism for both the elliptical and chest phantoms yielded similar CTDIvol. Therefore, both automatic and manual positioning compensation methods are useful for avoiding AEC miscalculations in off-centered patient positioning cases.

Deep learning has attracted intensive attention in synthetic aperture radar (SAR) automatic target recognition (ATR). Usually, a considerable number of labeled samples are necessary to learn a deep model for obtaining good generalization capability. However, the process of sample labeling is time-consuming and costly. This letter proposes an active self-paced deep learning (ASPDL) approach to SAR ATR. In a nutshell, we first introduce the Bayesian inference into the process of deep model parameter optimization, aiming at learning a robust classification model in the case of a limited number of labeled samples. Next, a cost-effective sample selection strategy is presented to iteratively and actively select the informative samples from a pool of unlabeled samples for labeling. Concretely, high-confidence samples are actively selected through self-paced learning (SPL) way and automatically pseudo-labeled with the current classification model, whereas low-confidence samples are chosen through an active learning strategy and manually labeled. Finally, we update the parameters of the model by minimizing a dual-loss function using a new training set that is constructed by incorporating new labeled samples with original ones. Experiments on the moving and stationary target acquisition and recognition (MSTAR) benchmark data demonstrate that the proposed method can achieve better classification accuracy with relatively few labeled samples compared with some state-of-the-art methods.

Synthetic aperture radar automatic target recognition (SAR ATR) is a key technique of remote-sensing image recognition, which has many potential applications in the fields of military surveillance, national defense, civil application, and so on. With the development of science and technology, deep convolutional neural network (DCNN) has been widely applied for SAR ATR. However, it is difficult to use deep learning to train models with limited ray SAR images. To resolve this problem, we proposed an effectively lightweight attention mechanism CNN (AM-CNN) model for SAR ATR. Extensive experimental results on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set illustrate that the AM-CNN model can achieve a superior recognition performance, and the average recognition accuracy can reach 99.35% on the classification of 10 class targets. Compared with the traditional CNN and the state-of-the-art method, our model is significantly superior to improve performance and efficiency.

This paper proposes a semi-automatic Fontan surgery planning method for designing and manufacturing hemodynamically optimized patient-specific grafts. Fontan surgery is a palliative procedure for patients with a single ventricle heart defect by creating a new path using a vascular graft for the deoxygenated blood to be directed to the lungs, bypassing the heart. However, designing patient-specific grafts with optimized hemodynamic performance is a complex task due to the variety of patient-specific anatomies, confined surgical planning space, and the requirement of simultaneously considering multiple design criteria for vascular graft optimization. To address these challenges, we used parameterized Fontan pathways to explore patient-specific vascular graft design spaces and search for optimal solutions by formulating a nonlinear constrained optimization problem, which minimizes indexed power loss (iPL) of the Fontan model by constraining hepatic flow distribution (HFD), percentage of abnormal wall shear stress (%WSS) and geometric interference between Fontan pathways and the heart models (InDep) within clinically acceptable thresholds. Gaussian process regression was employed to build surrogate models of the hemodynamic parameters as well as InDep and [Formula: see text] (conduit model smoothness indicator) for optimization by pattern search. We tested the proposed method on two patient-specific models (n=2). The results showed the automatically optimized (AutoOpt) Fontan models hemodynamically outperformed or at least are comparable to manually optimized Fontan models with significantly reduced surgical planning time (15 hours versus over 2 weeks). We also demonstrated feasibility of manufacturing the AutoOpt Fontan conduits by using electrospun nanofibers.

With the advancement of communication technologies and the development of the smart grid, today's physical power systems present an ever-growing dependency on cyber resources. It increases cyber vulnerabilities, causing safety and system stability concerns. In this article, a data-driven resilient automatic generation control (AGC) scheme is proposed under a false data injection attack (FDIA). The key idea is to identify the relationship between AGC signals and system operational conditions. This is achieved by the proposed regression-based FDIA signal predictions, including sequence-to-point prediction and the long short-term memory network-based prediction. It allows us to reconstruct the AGC control signals without being affected by FDIAs and to attenuate attacks in the closed control loop, thus alleviating the impact of FDIA on system performance. Numerical results carried out on the benchmark systems validate the effectiveness of the proposed method.

A phoneme is the smallest perceptually distinct sound unit that can be distinguished among words in a particular language. Every language has its own set of phonemes, and all possible words can be considered as ordered sequences of phonemes.The total number of phonemes contained in a language is always very few in comparison to the size of the vocabulary supported by the language. These facts have made phoneme recognition an attractive proposition in the entire journey of the Automatic Speech Processing (ASP) till date. As a result, the classification and recognition of phonemes are considered as the primary tasks of automatic speech recognition (ASR) systems irrespective of application domain. The dynamic nature of phonemes and several sources of their variability create lots of barriers in accurate identification of phonemes from an acoustic signal. The contribution of Machine Learning (ML) based techniques in overcoming these obstructions in automatic phoneme recognition (APR) is remarkable. Nowadays with lot of data availability, ML based ASR is preferred because of its simplicity over acoustic-phonetic based methods. The ML based techniques do not follow the conventional method based on identification of acoustic properties. Rather, ML techniques build their own trained model (algorithm) using readily available data. They do so by finding out the hidden patterns in speech signals, and acquire predictive intelligence through learning. Therefore, ML techniques can be said to provide a more generalized model for phoneme classification. In this paper, we present a comprehensive survey of ML tools to build phoneme recognizers. We also highlight some applications of speech (especially phoneme) recognition which illustrate the current scope as well as future prospects of APR.

In order to address the problems of time-consuming, low accuracy and poor convergence effect of traditional image automatic annotation algorithm, an Automatic annotation algorithm of medical radiological images based on convolutional neural network (CNN) is proposed. First of all, the image gradient information model was constructed, the edge contour feature of medical radiation image was initialized, the automatic segmentation model of medical radiation image was established by block template matching method, and the automatic segmentation processing of medical radiation image was completed. Secondly, by fusing the contour and gray information of image segmentation, the multi-resolution feature is extracted by using the three-dimensional distributed pixel sequence of image. The fusion feature decomposition of the image was obtained based on CNN, and the automatic annotation of medical radiation image was completed. The results show that the image segmentation effect of the proposed algorithm is good, the number of feature points is accurate, and the accuracy of multi-resolution feature extraction is as high as 98.7%. The convergence of image annotation is good, short time-consumption, and the F1 measurement value of the algorithm is high, and the overall performance is good.

The diseases and injuries of the knee joint are the most common orthopedic disorders. Personalized knee models can be helpful in the process of early intervention and lasting treatment techniques development. Fully automatic reconstruction of knee joint anatomical structures from medical images (CT, MRI, ultrasound) remains a challenge. For this reason, most of state-of-the-art knee joint models contain simplifications such as representation of muscles and ligaments as line segments connecting two points which replace attachment areas. The paper presents algorithms for automatic detection of such points on knee CT images. This paper presents three approaches to automatic detection of ligaments and tendons attachment sites on the patients CT images: qualitative anatomical descriptions, analysis of bones curvature, and quantitative anatomical descriptions. Combinations of these approaches result in new automatic detection algorithms. Each algorithm exploits anatomical peculiarities of each attachment site, e.g., bone curvature and number of other attachments in a neighborhood of the site. The experimental dataset consisted of 26 anonymized CT sequences containing right and left knee joints in different resolutions. The proposed algorithms take into account bone surface curvatures and spatial differences in locations of medial and lateral parts of both knees. The algorithms for detection of quadriceps femoris, popliteus, biceps femoris tendons, and lateral collateral and medial collateral ligaments attachment sites are provided, as well as examples of their application. Two algorithms are validated by comparison with known statistics of ligaments lengths and also using ground truth annotations for anatomical landmarks approved by clinical experts. The algorithms simplify generation of patient-specific knee joint models demanded in personalized biomechanical models. The algorithms in the current implementation have two important limitations. First, the correctness of the produced results depends on the bones segmentation quality. Second, the presented algorithms detect a point of the attachment site, which is not necessarily its center. Therefore, manual correction of the attachment site location may be required for attachments with relatively large area.

BackgroundStandardized coding of plays an important role in radiology reports’ secondary use such as data analytics, data-driven decision support, and personalized medicine. RadLex, a standard radiological lexicon, can reduce subjective variability and improve clarity in radiology reports. RadLex coding of radiology reports is widely used in many countries, but translation and localization of RadLex in China are far from being established. Although automatic RadLex coding is a common way for non-standard radiology reports, the high-accuracy cross-language RadLex coding is hardly achieved due to the limitation of up-to-date auto-translation and text similarity algorithms and still requires further research.MethodsWe present an effective approach that combines a hybrid translation and a Multilayer Perceptron weighting text similarity ensemble algorithm for automatic RadLex coding of Chinese structured radiology reports. Firstly, a hybrid way to integrate Google neural machine translation and dictionary translation helps to optimize the translation of Chinese radiology phrases to English. The dictionary is made up of 21,863 Chinese–English radiological term pairs extracted from several free medical dictionaries. Secondly, four typical text similarity algorithms are introduced, which are Levenshtein distance, Jaccard similarity coefficient, Word2vec Continuous bag-of-words model, and WordNet Wup similarity algorithms. Lastly, the Multilayer Perceptron model has been used to synthesize the contextual, lexical, character and syntactical information of four text similarity algorithms to promote precision, in which four similarity scores of two terms are taken as input and the output presents whether the two terms are synonyms.ResultsThe results show the effectiveness of the approach with an F1-score of 90.15%, a precision of 91.78% and a recall of 88.59%. The hybrid translation algorithm has no negative effect on the final coding, F1-score has increased by 21.44% and 8.12% compared with the GNMT algorithm and dictionary translation. Compared with the single similarity, the result of the MLP weighting similarity algorithm is satisfactory that has a 4.48% increase compared with the best single similarity algorithm, WordNet Wup.ConclusionsThe paper proposed an innovative automatic cross-language RadLex coding approach to solve the standardization of Chinese structured radiology reports, that can be taken as a reference to automatic cross-language coding.

The paper discusses a new approach to improving the accuracy of reproducing the task signal by a precision circular transfer table based on the implementation in the form of a mechatronic module. This design eliminates the disadvantages of the traditional design, such as backlash, failure when fixing the face-plate. The mechatronic module is a gearless connection between the face-plate and the electric motor. The new design required a new approach to the operation of the automatic tracking and stage positioning control system. In the process of stage positioning, thermal deformations occur, and therefore, a feature is the inclusion of an additional automatic control system for the cooling of an executive synchronous electric motor with permanent magnets in the rotor. The paper deals with the issue of the interrelationship between the positioning and tracking systems with the thermal deformation compensation system. Simulation of the operation of two systems has been carried out.