Tool For Automatic Detection Research Articles

Misalignment Detection for Web-Scraped Corpora: A Supervised Regression Approach

To build state-of-the-art Neural Machine Translation (NMT) systems, high-quality parallel sentences are needed. Typically, large amounts of data are scraped from multilingual web sites and aligned into datasets for training. Many tools exist for automatic alignment of such datasets. However, the quality of the resulting aligned corpus can be disappointing. In this paper, we present a tool for automatic misalignment detection (MAD). We treated the task of determining whether a pair of aligned sentences constitutes a genuine translation as a supervised regression problem. We trained our algorithm on a manually labeled dataset in the FR–NL language pair. Our algorithm used shallow features and features obtained after an initial translation step. We showed that both the Levenshtein distance between the target and the translated source, as well as the cosine distance between sentence embeddings of the source and the target were the two most important features for the task of misalignment detection. Using gold standards for alignment, we demonstrated that our model can increase the quality of alignments in a corpus substantially, reaching a precision close to 100%. Finally, we used our tool to investigate the effect of misalignments on NMT performance.

Computer aided tool for automatic detection and delineation of nucleus from oral histopathology images for OSCC screening

Characterization of nucleus from oral tissue histology images is important for oral squamous cell carcinoma (OSCC) diagnosis. Therefore, the accurate delineation of nucleus is the foremost task to initiate the quantification of its various properties such as morphology, texture, intensity, etc. In this paper, we propose a two-stage computer-aided tool for automatic detection (the first stage) and delineation of the detected nucleus (second stage) from oral histological images to aid clinician in OSCC diagnosis. In the first stage, 81×81 patches were extracted from our image database followed by the wavelet-based downsampling process to generate 21×21 patches, which were used to train the deep convolution neural network (CNN) architecture. This approach was used to automatically detect nucleus from OSCC images and achieved 88.87% recall and 82.03% precision. After detection, 81×81 image patches were extracted from those detected nucleus regions and further processed through the second stage of proposed pipeline to segment the exact boundary of nucleus using active contour (Chan-Vese model) after applying edge enhancing nonsubsampled contourlet transform. The proposed segmentation methodology performed well with 94.22% Dice coefficient, 89.38% Jaccard index, 97.56% precision, and 91.58% recall. We view this work as the first attempt on oral tissue histology image computation for joint nucleus detection and segmentation to diagnose OSCC to the best of our knowledge.

PattRec: An easy-to-use CNV detection tool optimized for targeted NGS assays with diagnostic purposes

Genetic laboratories use custom-commercial targeted next-generation sequencing (tg-NGS) assays to identify disease-causing variants. Although the high coverage achieved with these tests allows for the detection of copy number variants (CNVs), which account for an important proportion of the genetic burden in human diseases, an easy-to-use tool for automatic CNV detection is still lacking. This article presents a new CNV detection tool optimized for tg-NGS data: PattRec. PattRec was evaluated using a wide range of data, and its performance compared with those of other CNV detection tools. The software includes features for selecting optimal controls, discarding polymorphic CNVs prior to analysis, and filtering out deletions based on SNV zygosity, and automatically creates an in-house CNV database. There is no need for high level bioinformatic expertise and users can choose color-coded xlsx output that helps to prioritize potentially pathogenic CNVs. PattRec is presented as a Java based GUI, freely available online: https://github.com/irotero/PattRec.

Semi-Supervised Target-Dependent Sentiment Classification for Micro-Blogs

The wealth of opinions available in the social media motivated researchers to develop automatic opinion detection tools. Many such tools are currently available online for opinion mining in short text, known as micro-blogs, but their efficacies are still limited. Current tools focus on detecting sentiment polarity expressed in a micro-blog regardless of the topic (target) discussed. Little improved approaches have been proposed to detect sentiment towards a specific target, referred to as target-dependent sentiment classification. Our literature review has shown that all these target-dependent approaches use supervised learning techniques. Such techniques need a huge amount of labeled data for increasing classification accuracy. However, preparing labeled data from social media needs a lot of efforts. In this work, we address this issue by employing semisupervised learning techniques that have not been used before with target-dependent sentiment classification. To the best of our knowledge, our work is the first research that employs semisupervised learning techniques in this direction. Semi-supervised learning techniques have been known in the literature to improve classification accuracy in comparison with supervised learning techniques; however, they use same number of labeled samples plus many unlabelled ones. In this work, we propose a new semi-supervised learning technique that uses less number of labeled microblogs than that used with supervised learning techniques. Experiment results have shown that the proposed technique provides competitive accuracy.

Analysis and Detection of Health-Related Misinformation on Chinese Social Media

With the mobile Internet development, e-health has become increasingly connected with people's daily life. However, health information on Internet is severely corrupted by misinformation, especially for the aged. It is necessary to analyze the characteristics of health-related misinformation on Internet and to design automated detection tools. In this study, we focus on analyzing common characteristics of reliable and unreliable health-related information on Chinese online social media, and exploring possible detection method using machine learning algorithms. We first collect a dataset containing both reliable and unreliable health-related articles from multiple Chinese online social media sites, with 2,296 reliable and 2,085 unreliable included. Then we analyze their differences with respect to writing style, text topic and feature distribution by both intuitive and statistical analysis. We also manually select 104 linguistic and statistical features that are useful for machine learning classifiers. Lastly, we propose a Health-related Misinformation Detection framework (HMD) that includes a feature-based method and a text-based method for detecting unreliable health-related information. Experiments verifies the performance of our proposed HMD method.

Software Design Smell Detection: a systematic mapping study

Design Smells are indicators of situations that negatively affect software quality attributes such as understandability, testability, extensibility, reusability, and maintainability in general. Improving maintainability is one of the cornerstones of making software evolution easier. Hence, Design Smell Detection is important in helping developers when making decisions that can improve software evolution processes. After a long period of research, it is important to organize the knowledge produced so far and to identify current challenges and future trends. In this paper, we analyze 18 years of research into Design Smell Detection. There is a wide variety of terms that have been used in the literature to describe concepts which are similar to what we have defined as “Design Smells,” such as design defect, design flaw, anomaly, pitfall, antipattern, and disharmony. The aim of this paper is to analyze all these terms and include them in the study. We have used the standard systematic literature review method based on a comprehensive set of 395 articles published in different proceedings, journals, and book chapters. We present the results in different dimensions of Design Smell Detection, such as the type or scope of smell, detection approaches, tools, applied techniques, validation evidence, type of artifact in which the smell is detected, resources used in evaluation, supported languages, and relation between detected smells and software quality attributes according to a quality model. The main contributions of this paper are, on the one hand, the application of domain modeling techniques to obtain a conceptual model that allows the organization of the knowledge on Design Smell Detection and a collaborative web application built on that knowledge and, on the other, finding how tendencies have moved across different kinds of smell detection, as well as different approaches and techniques. Key findings for future trends include the fact that all automatic detection tools described in the literature identify Design Smells as a binary decision (having the smell or not), which is an opportunity to evolve to fuzzy and prioritized decisions. We also find that there is a lack of human experts and benchmark validation processes, as well as demonstrating that Design Smell Detection positively influences quality attributes.

EPINETLAB: A Software for Seizure-Onset Zone Identification From Intracranial EEG Signal in Epilepsy.

The pre-operative workup of patients with drug-resistant epilepsy requires in some candidates the identification from intracranial EEG (iEEG) of the seizure-onset zone (SOZ), defined as the area responsible of the generation of the seizure and therefore candidate for resection. High-frequency oscillations (HFOs) contained in the iEEG signal have been proposed as biomarker of the SOZ. Their visual identification is a very onerous process and an automated detection tool could be an extremely valuable aid for clinicians, reducing operator-dependent bias, and computational time. In this manuscript, we present the EPINETLAB software, developed as a collection of routines integrated in the EEGLAB framework that aim to provide clinicians with a structured analysis pipeline for HFOs detection and SOZ identification. The tool implements an analysis strategy developed by our group and underwent a preliminary clinical validation that identifies the HFOs area by extracting the statistical properties of HFOs signal and that provides useful information for a topographic characterization of the relationship between clinically defined SOZ and HFO area. Additional functionalities such as inspection of spectral properties of ictal iEEG data and import and analysis of source-space magnetoencephalographic (MEG) data were also included. EPINETLAB was developed with user-friendliness in mind to support clinicians in the identification and quantitative assessment of HFOs in iEEG and source space MEG data and aid the evaluation of the SOZ for pre-surgical assessment.

A dynamic machine tool circle test calibration method by R-test

With the continuous popularization of automation in the manufacturing industry and continuous improvement of production efficiency, new requirements have been put forward for the error detection of machine tool. As a new automated detection tool, R-test has been widely studied by scholars. This paper proposes a dynamic test method for R-test, and analyses the influence of three main error sources of servo error, pitch error and reading error. The corresponding elimination methods are also introduced. An experiment on a double turntable machine tool was conducted to verify the analysis. In addition, the selection of feed rate and initial position in circle test is also studied.

ADOPT: A tool for automatic detection of tectonic plates at the surface of convection models

AbstractMantle convection models with plate‐like behavior produce surface structures comparable to Earth's plate boundaries. However, analyzing those structures is a difficult task, since convection models produce, as on Earth, diffuse deformation and elusive plate boundaries. Therefore we present here and share a quantitative tool to identify plate boundaries and produce plate polygon layouts from results of numerical models of convection: Automatic Detection Of Plate Tectonics (ADOPT). This digital tool operates within the free open‐source visualization software Paraview. It is based on image segmentation techniques to detect objects. The fundamental algorithm used in ADOPT is the watershed transform. We transform the output of convection models into a topographic map, the crest lines being the regions of deformation (plate boundaries) and the catchment basins being the plate interiors. We propose two generic protocols (the field and the distance methods) that we test against an independent visual detection of plate polygons. We show that ADOPT is effective to identify the smaller plates and to close plate polygons in areas where boundaries are diffuse or elusive. ADOPT allows the export of plate polygons in the standard OGR‐GMT format for visualization, modification, and analysis under generic softwares like GMT or GPlates.

Validation of the CME Geomagnetic Forecast Alerts Under the COMESEP Alert System

Under the European Union 7th Framework Programme (EU FP7) project Coronal Mass Ejections and Solar Energetic Particles (COMESEP, http://comesep.aeronomy.be), an automated space weather alert system has been developed to forecast solar energetic particles (SEP) and coronal mass ejection (CME) risk levels at Earth. The COMESEP alert system uses the automated detection tool called Computer Aided CME Tracking (CACTus) to detect potentially threatening CMEs, a drag-based model (DBM) to predict their arrival, and a CME geoeffectiveness tool (CGFT) to predict their geomagnetic impact. Whenever CACTus detects a halo or partial halo CME and issues an alert, the DBM calculates its arrival time at Earth and the CGFT calculates its geomagnetic risk level. The geomagnetic risk level is calculated based on an estimation of the CME arrival probability and its likely geoeffectiveness, as well as an estimate of the geomagnetic storm duration. We present the evaluation of the CME risk level forecast with the COMESEP alert system based on a study of geoeffective CMEs observed during 2014. The validation of the forecast tool is made by comparing the forecasts with observations. In addition, we test the success rate of the automatic forecasts (without human intervention) against the forecasts with human intervention using advanced versions of the DBM and CGFT (independent tools available at the Hvar Observatory website, http://oh.geof.unizg.hr). The results indicate that the success rate of the forecast in its current form is unacceptably low for a realistic operation system. Human intervention improves the forecast, but the false-alarm rate remains unacceptably high. We discuss these results and their implications for possible improvement of the COMESEP alert system.

Chromatin organization at the nuclear periphery as revealed by image analysis of structured illumination microscopy data

The nuclear periphery (NP) plays a substantial role in chromatin organization. Heterochromatin at the NP is interspersed with active chromatin surrounding nuclear pore complexes (NPCs); however, details of the peripheral chromatin organization are missing. To discern the distribution of epigenetic marks at the NP of HeLa nuclei, we used structured illumination microscopy combined with a new MATLAB software tool for automatic NP and NPC detection, measurements of fluorescent intensity and statistical analysis of measured data. Our results show that marks for both active and non-active chromatin associate differentially with NPCs. The incidence of heterochromatin marks, such as H3K27me2 and H3K9me2, was significantly lower around NPCs. In contrast, the presence of marks of active chromatin such as H3K4me2 was only decreased very slightly around the NPCs or not at all (H3K9Ac). Interestingly, the histone demethylases LSD1 (also known as KDM1A) and KDM2A were enriched within the NPCs, suggesting that there was a chromatin-modifying mechanism at the NPCs. Inhibition of transcription resulted in a larger drop in the distribution of H1, H3K9me2 and H3K23me2, which implies that transcription has a role in the organization of heterochromatin at the NP.

A new two-dimensional matched filter based on the modified Chebyshev type I function for retinal vessels detection.

Retinal vessel segmentation takes a significant part in an automated diabetic retinopathy screening task. However, this can be a challenging job because of the low contrast retinal images and the presences of retinal pathologies. Hence, in this paper, we propose a novel matched filter based on the modified Chebyshev type I function for retinal blood vessels candidates detection. The proposed method is combined with the pre-processing and the post-processing phases to have a complete retinal vessel segmentation scheme. The retinal images from the DRIVE and STARE databases, which are equipped with the ground truths are used to evaluate our proposed method in the segmentation scheme. Using our method, the achieved average levels of sensitivity, specificity, and accuracy are 0.756, 0.973, and 0.954, for the DRIVE database, and 0.731, 0.972, and 0.953, for the STARE database, being better than other presented methods. The high results indicate that our method is reliable to be used in an automated detection tool for diabetic retinopathy.

ICARES: a real-time automated detection tool for clusters of infectious diseases in the Netherlands

BackgroundClusters of infectious diseases are frequently detected late. Real-time, detailed information about an evolving cluster and possible associated conditions is essential for local policy makers, travelers planning to visit the area, and the local population. This is currently illustrated in the Zika virus outbreak.MethodsIn the Netherlands, ICARES (Integrated Crisis Alert and Response System) has been developed and tested on three syndromes as an automated, real-time tool for early detection of clusters of infectious diseases. From local general practices, General Practice Out-of-Hours services and a hospital, the numbers of routinely used syndrome codes for three piloted tracts i.e., respiratory tract infection, hepatitis and encephalitis/meningitis, are sent on a daily basis to a central unit of infectious disease control. Historic data combined with information about patients’ syndromes, age cohort, gender and postal code area have been used to detect clusters of cases.ResultsDuring the first 2 years, two out of eight alerts appeared to be a real cluster. The first was part of the seasonal increase in Enterovirus encephalitis and the second was a remarkably long lasting influenza season with high peak incidence.ConclusionsThis tool is believed to be the first flexible automated, real-time cluster detection system for infectious diseases, based on physician information from both general practitioners and hospitals. ICARES is able to detect and follow small regional clusters in real time and can handle any diseases entity that is regularly registered by first line physicians. Its value will be improved when more health care institutions agree to link up with ICARES thus improving further the signal-to-noise ratio.

Development and Validation of Automated Detection Tools for Vocalizations of Rare and Endangered Anurans

AbstractFor many rare or endangered anurans, monitoring is achieved via auditory cues alone. Human-performed audio surveys are inherently biased, and may fail to detect animals when they are present. Automated audio recognition tools offer an alternative mode of observer-free monitoring. Few commercially available platforms for developing these tools exist, and little research has investigated whether these tools are effective at detecting rare vocalization events. We generated a recognizer for detecting the vocalization of the endangered Houston toad Anaxyrus houstonensis using SongScope© bioacoustics software. We developed our recognizer using a large sample of training data that included only the highest quality of recorded audio (i.e., low noise, no interfering vocalizations) divided into small, manageable batches. To track recognizer performance, we generated an independent set of test data through randomly sampling a large population of audio known to possess Houston toad vocalizations. We analyzed training data and test data recursively, using a criterion of zero tolerance for false-negative detections. For each step, we incorporated a new batch of training data into the recognizer. Once we included all training data, we manually verified recognizer performance against one full month (March 2014) of audio taken from a known breeding locality. The recognizer successfully identified 100% of all training data and 97.2% of all test data. However, there is a trade-off between reducing false-negative and increasing false-positive detections, which limited the usefulness of some features of SongScope. Methods of automated detection represent a means by which we may test the efficacy of the manual monitoring techniques currently in use. The ability to search any collection of audio recordings for Houston toad vocalizations has the potential to challenge the paradigms presently placed on monitoring for this species of conservation concern.

RIPPLELAB: A Comprehensive Application for the Detection, Analysis and Classification of High Frequency Oscillations in Electroencephalographic Signals.

High Frequency Oscillations (HFOs) in the brain have been associated with different physiological and pathological processes. In epilepsy, HFOs might reflect a mechanism of epileptic phenomena, serving as a biomarker of epileptogenesis and epileptogenicity. Despite the valuable information provided by HFOs, their correct identification is a challenging task. A comprehensive application, RIPPLELAB, was developed to facilitate the analysis of HFOs. RIPPLELAB provides a wide range of tools for HFOs manual and automatic detection and visual validation; all of them are accessible from an intuitive graphical user interface. Four methods for automated detection—as well as several options for visualization and validation of detected events—were implemented and integrated in the application. Analysis of multiple files and channels is possible, and new options can be added by users. All features and capabilities implemented in RIPPLELAB for automatic detection were tested through the analysis of simulated signals and intracranial EEG recordings from epileptic patients (n = 16; 3,471 analyzed hours). Visual validation was also tested, and detected events were classified into different categories. Unlike other available software packages for EEG analysis, RIPPLELAB uniquely provides the appropriate graphical and algorithmic environment for HFOs detection (visual and automatic) and validation, in such a way that the power of elaborated detection methods are available to a wide range of users (experts and non-experts) through the use of this application. We believe that this open-source tool will facilitate and promote the collaboration between clinical and research centers working on the HFOs field. The tool is available under public license and is accessible through a dedicated web site.

Instability in Distributed Mobility Management

Mobility support is critical to offering seamless data service to mobile devices in 3G/4G cellular networks. To accommodate policy requests by users and carriers, micro-mobility management scheme among cells (i.e., handoff) is designated to be configurable. Each cell and mobile device can configure or even customize its own handoff procedure. In this paper, we examine the handoff misconfiguration issues in 3G/4G networks. We show that they may incur handoff instability in the form of persistent loops, where the device oscillates between cells even without radio-link and location changes. Such instability is mainly triggered by uncoordinated parameter configurations and inconsistent decision logic in the hand- off procedure. It can degrade user data performance, incur excessive signaling overhead, and violate network's expected handoff goals. We derive the instability conditions, and validate them on two major US mobile carrier networks. We further design a soft- ware tool for automatic loop detection, and run it over operational networks. We discuss possible fixes to such uncoordinated configurations among devices and cells.

Automatic detection of carotid arteries in computed tomography angiography: a proof of concept protocol.

Atherosclerosis is one of the leading causes of mortality in the western world. Computed tomography angiography (CTA) is the conventional imaging method used for pre-surgery assessment of the blood flow within the carotid vessel. In this paper, we present a proof of concept of a novel, fast and operator independent protocol for the automatic detection (seeding) of the carotid arteries in CTA in the thorax and upper neck region. The dataset is composed of 14 patients' CTA images of the neck region. The performance of this method is compared with manual seeding by four trained operators. Inter-operator variation is also assessed based on the dataset. The minimum, average and maximum coefficient of variation among the operators was (0, 2, 5 %), respectively. The performance of our method is comparable with the state of the art alternative, presenting a detection rate of 75 and 71 % for the lowest and uppermost image levels, respectively. The mean processing time is 167s per patient versus 386s for manual seeding. There are no significant differences between the manual and automatic seed positions in the volumes (p = 0.29). A fast, operator independent protocol was developed for the automatic detection of carotid arteries in CTA. The results are encouraging and provide the basis for the creation of automatic detection and analysis tools for carotid arteries.

Automatic Feature Selection of Hardware Layout: A Step toward Robust Hardware Trojan Detection

Recently, the problem of hardware Trojan detection has gained a tangible significance in academia and industry. That problem, by its nature, is complex, time consuming and error prone due to design and fabrication outsourcing of hardware circuits to external untrusted foundries. Researchers have proposed different approaches, either destructive or non-destructive, to overcome that problem. The destructive approach depends on reverse engineering via decapsulation, delayering and layout identification. This paper presents a first trial of a new approach that can afford an automatic and robust solution for the step of layout identification. The proposed technique takes the underlying digital circuit as input, and automatically determines its basic features using Haar feature extractor. Based on that features, a decision tree is trained to act as a weak classifier, which is later boosted, by making use of AdaBoost learning algorithm, to produce a strong classifier in a chain of cascaded classifiers. Accordingly, a classification model is built up to provide an automatic hardware Trojan location and detection tool. To evaluate the proposed model, ISCAS89 benchmark dataset was used for training and testing. The hardware dataset has been altered deliberately to show different Trojan examples ---namely, Trojan insertion, Trojan deletion and Trojan parametric- inside hardware circuits. By investigating the underlying experimental results, the capabilities of the proposed model are evaluated, and the evaluation shows that the approach can detect different hardware Trojan types in different circuit layouts, with high accuracy rate. The proposed approach is not only automatic, but also robust and promising.

Monitoring Count Time Series inR: Aberration Detection in Public Health Surveillance

Public health surveillance aims at lessening disease burden, e.g., in case of infectious diseases by timely recognizing emerging outbreaks. Seen from a statistical perspective, this implies the use of appropriate methods for monitoring time series of aggregated case reports. This paper presents the tools for such automatic aberration detection offered by the R package surveillance. We introduce the functionality for the visualization, modelling and monitoring of surveillance time series. With respect to modelling we focus on univariate time series modelling based on generalized linear models (GLMs), multivariate GLMs, generalized additive models and generalized additive models for location, shape and scale. This ranges from illustrating implementational improvements and extensions of the well-known Farrington algorithm, e.g, by spline-modelling or by treating it in a Bayesian context. Furthermore, we look at categorical time series and address overdispersion using beta-binomial or Dirichlet-Multinomial modelling. With respect to monitoring we consider detectors based on either a Shewhart-like single timepoint comparison between the observed count and the predictive distribution or by likelihood-ratio based cumulative sum methods. Finally, we illustrate how surveillance can support aberration detection in practice by integrating it into the monitoring workflow of a public health institution. Altogether, the present article shows how well surveillance can support automatic aberration detection in a public health surveillance context.

Automatic detection of referral patients due to retinal pathologies through data mining.

With the increased prevalence of retinal pathologies, automating the detection of these pathologies is becoming more and more relevant. In the past few years, many algorithms have been developed for the automated detection of a specific pathology, typically diabetic retinopathy, using eye fundus photography. No matter how good these algorithms are, we believe many clinicians would not use automatic detection tools focusing on a single pathology and ignoring any other pathology present in the patient's retinas. To solve this issue, an algorithm for characterizing the appearance of abnormal retinas, as well as the appearance of the normal ones, is presented. This algorithm does not focus on individual images: it considers examination records consisting of multiple photographs of each retina, together with contextual information about the patient. Specifically, it relies on data mining in order to learn diagnosis rules from characterizations of fundus examination records. The main novelty is that the content of examination records (images and context) is characterized at multiple levels of spatial and lexical granularity: 1) spatial flexibility is ensured by an adaptive decomposition of composite retinal images into a cascade of regions, 2) lexical granularity is ensured by an adaptive decomposition of the feature space into a cascade of visual words. This multigranular representation allows for great flexibility in automatically characterizing normality and abnormality: it is possible to generate diagnosis rules whose precision and generalization ability can be traded off depending on data availability. A variation on usual data mining algorithms, originally designed to mine static data, is proposed so that contextual and visual data at adaptive granularity levels can be mined. This framework was evaluated in e-ophtha, a dataset of 25,702 examination records from the OPHDIAT screening network, as well as in the publicly-available Messidor dataset. It was successfully applied to the detection of patients that should be referred to an ophthalmologist and also to the specific detection of several pathologies.