A comparative analysis of Spanish Clinical encoder-based models on NER and classification tasks

OBJECTIVES

This comparative analysis aims to assess the efficacy of encoder Language Models for clinical tasks in the Spanish language. The primary goal is to identify the most effective resources within this context.

IMPORTANCE

This study highlights a critical gap in NLP resources for the Spanish language, particularly in the clinical sector. Given the vast number of Spanish speakers globally and the increasing reliance on electronic health records, developing effective Spanish language models is crucial for both clinical research and healthcare delivery. Our work underscores the urgent need for specialized encoder models in Spanish that can handle clinical data with high accuracy, thus paving the way for advancements in healthcare services and biomedical research for Spanish-speaking populations.

MATERIALS AND METHODS

We examined 17 distinct corpora with a focus on clinical tasks. Our evaluation centered on Spanish Language Models and Spanish Clinical Language models (both encoder-based). To ascertain performance, we meticulously benchmarked these models across a curated subset of the corpora. This extensive study involved fine-tuning over 3000 models.

RESULTS

Our analysis revealed that the best models are not clinical models, but general-purpose models. Also, the biggest models are not always the best ones. The best-performing model, RigoBERTa 2, obtained an average F1 score of 0.880 across all tasks.

DISCUSSION

Our study demonstrates the advantages of dedicated encoder-based Spanish Clinical Language models over generative models. However, the scarcity of diverse corpora, mostly focused on NER tasks, underscores the need for further research. The limited availability of high-performing models emphasizes the urgency for development in this area.

CONCLUSION

Through systematic evaluation, we identified the current landscape of encoder Language Models for clinical tasks in the Spanish language. While challenges remain, the availability of curated corpora and models offers a foundation for advancing Spanish Clinical Language models. Future efforts in refining these models are essential to elevate their effectiveness in clinical NLP.

The landslide's conceptualizing economic decline and its framing effect: Mandarin evidence

Conceptual metaphors are essential for explaining and understanding social concerns. Natural disaster metaphors are commonly employed to access the abstract and negative impacts of social issues. Five of the top 10 most prevalent natural disaster frames in the Center for Chinese Linguistics (CCL)-earthquake, flood, fire hazard, drought, typhoon, landslide, volcano, sandstorm, tsunami, and debris flow-share a common economic target domain and show economic recession. Additionally, corpus-based research has revealed that the landslide frame is the most salient in figuratively representing economic declines. An experimental study derived from the corpus analysis has found that the landslide-framed economic crises posed more severity to participants and exerted a notable influence on their opinions and judgments. Therefore, when effective communication of economic hazards is to be realized, metaphorical representation of economic crises demands great consideration.

Part 2: The corpus and the angle regions

BACKGROUND/AIMS

Although the mandible is the largest and strongest bone of the facial skeleton, it is frequently broken. The fracture location in the mandible depends on the biomechanical features, direction and angle of the trauma, and masticatory muscles. This study aimed to evaluate the stresses caused by trauma to the corpus and angle regions from different angles.

MATERIALS AND METHODS

After computer-based mandible models were created using finite element analysis, a force of 2000 Newton(N) was simulated with the mouth open or closed to the corpus and the angle. To the corpus: at 90° (Model 1) in the lateromedial direction, 45° (Model 2) in the lateromedial-inferosuperior direction, and 90° (Model 3) in the inferosuperior direction. To angle: 90° (Model 4) in the lateromedial direction and 45° (Model 5) in the lateromedial-inferosuperior direction. The resulting stress intensity was assessed using FEA.

RESULTS

Following the simulated forces, the maximum stress in the mandible occurred in the condylar region, except in Model 3 (Left(L)Corpus2[36 megapascals(MPa)]) in the mouth-closed condition. After traumas in Model 1 (open-mouth: LCondyle2[547 MPa]) and Model 4 (closed-mouth: LCondyle2[607 MPa]), higher stress values occurred in the condyle. In the mouth open-closed state, there was no significant stress change in the condyle region in Model 1 (open-mouth: LCondyle2[547 MPa], closed-mouth:LCondyle2[546 MPa]) or in Model 2 (open mouth: Right(R)Condyle2[431 MPa], closed-mouth:LCondyle2[439 MPa]). In Model 3, lower stress values occurred in the closed-mouth rather than the open-mouth (LCondyle1[167 MPa]) state. In Models 4 and 5, the stress values increased in the mouth-closed condition compared with the mouth-open condition.

CONCLUSIONS

Stress in the mandible is affected by the location of the trauma and the angle of incidence of the blow. In trauma to both the corpus and the angle, the most common area to be fractured is the condyle.

Introducing the keyconcept approach to the analysis of language: the case of regulation in COVID-19 diaries

Using the Mass Observation corpus of 12th of May Diaries, we investigate concepts that are characteristic of the first coronavirus lockdown in the UK. More specifically, we extract and analyse concepts which are distinctive of the discourses produced in May 2020 in relation to concepts used in the 10 previous years, 2010-2019. In the current paper we focus on the concept of regulation, which we identify through a novel approach to querying semantic content in large datasets. Typically, linguists look at keywords to understand differences between two datasets. We demonstrate that taking the perspective of a keyconcept rather than the keyword in linguistic analysis is a beneficial way of identifying trends in broader patterns of thoughts and behaviours which reflect lived-experiences that are particularly prominent of a given dataset, which, in this current paper, is the COVID-19 era dataset. In order to contextualise the keyconcept analysis, we investigate the discourses surrounding the concept of regulation. We find that diarists communicate collective experience of limited individual agency, surrounded by feelings of fear and gratitude. Diarists' reporting on events is often fragmented, focused on new information, and firmly placed in a temporal frame.

Learning conversational dependency: Children's response using un in Japanese

This study investigates how Japanese-speaking children learn interactional dependencies in conversations that determine the use of un, a token typically used as a positive response for yes-no questions, backchannel, and acknowledgement. We hypothesise that children learn to produce un appropriately by recognising different types of cues occurring in the immediately preceding turns. We built a set of generalised linear models on the longitudinal conversation data from seven children aged 1 to 5 years and their caregivers. Our models revealed that children not only increased their un production, but also learned to attend relevant cues in the preceding turns to understand when to respond by producing un. Children increasingly produced un when their interlocutors asked a yes-no question or signalled the continuation of their own speech. These results illustrate how children learn the probabilistic dependency between adjacent turns, and become able to participate in conversational interactions.

Distribution characteristics of mitochondria-rich segments in the epididymis

The epididymis is a highly specialized tissue that plays vital roles in sperm maturation and storage. The spatio-temporal repertoire of epididymal cells and their gene expression in the epididymis remain less characterized. With the help of single-cell RNA sequencing (scRNA-seq), Shi et al., reveal a spatio- and segment-specific distribution pattern of mitochondria that adds another layer of complexity to our understanding of the epididymis. They unexpectedly find a higher abundance of mitochondria and mitochondrial transcription in the corpus and cauda compared to the caput of epididymis, which are believed to be responsible for providing the energy necessary for sperm maturation and motility.

Kindness in British communities: Discursive practices of promoting kindness during the Covid pandemic

This research adopts Critical Discourse Analysis as a perspective to explore how kindness was expressed and promoted in university communities and city communities from January to March in 2020 when the Covid pandemic broke out in the UK and provide a window on British culture in which kindness was expressed and promoted through discourse. It combines a qualitative method with a corpus-based quantitative method. It is found that kindness was meant for providing support and showing compassion and inclusion to community members and that strategies in lexis, syntax and metaphor can reproduce or resist the expression and promotion of kindness in communities. During the pandemic, the intentional kindness expressed by community authorities was respect of diversity rather than inclusion of different values or ethnicity and no substantial support was provided to vulnerable members even though authorities were trying to impress the public by claiming that they were making constant efforts to support the community. Case studies revealed that we should caution against the use of passivation and the pronouns like they.

First Steps Towards a Risk of Bias Corpus of Randomized Controlled Trials

Risk of bias (RoB) assessment of randomized clinical trials (RCTs) is vital to conducting systematic reviews. Manual RoB assessment for hundreds of RCTs is a cognitively demanding, lengthy process and is prone to subjective judgment. Supervised machine learning (ML) can help to accelerate this process but requires a hand-labelled corpus. There are currently no RoB annotation guidelines for randomized clinical trials or annotated corpora. In this pilot project, we test the practicality of directly using the revised Cochrane RoB 2.0 guidelines for developing an RoB annotated corpus using a novel multi-level annotation scheme. We report inter-annotator agreement among four annotators who used Cochrane RoB 2.0 guidelines. The agreement ranges between 0% for some bias classes and 76% for others. Finally, we discuss the shortcomings of this direct translation of annotation guidelines and scheme and suggest approaches to improve them to obtain an RoB annotated corpus suitable for ML.

Supervised Relation Extraction Between Suicide-Related Entities and Drugs: Development and Usability Study of an Annotated PubMed Corpus

BACKGROUND

Drug-induced suicide has been debated as a crucial issue in both clinical and public health research. Published research articles contain valuable data on the drugs associated with suicidal adverse events. An automated process that extracts such information and rapidly detects drugs related to suicide risk is essential but has not been well established. Moreover, few data sets are available for training and validating classification models on drug-induced suicide.

OBJECTIVE

This study aimed to build a corpus of drug-suicide relations containing annotated entities for drugs, suicidal adverse events, and their relations. To confirm the effectiveness of the drug-suicide relation corpus, we evaluated the performance of a relation classification model using the corpus in conjunction with various embeddings.

METHODS

We collected the abstracts and titles of research articles associated with drugs and suicide from PubMed and manually annotated them along with their relations at the sentence level (adverse drug events, treatment, suicide means, or miscellaneous). To reduce the manual annotation effort, we preliminarily selected sentences with a pretrained zero-shot classifier or sentences containing only drug and suicide keywords. We trained a relation classification model using various Bidirectional Encoder Representations from Transformer embeddings with the proposed corpus. We then compared the performances of the model with different Bidirectional Encoder Representations from Transformer-based embeddings and selected the most suitable embedding for our corpus.

RESULTS

Our corpus comprised 11,894 sentences extracted from the titles and abstracts of the PubMed research articles. Each sentence was annotated with drug and suicide entities and the relationship between these 2 entities (adverse drug events, treatment, means, and miscellaneous). All of the tested relation classification models that were fine-tuned on the corpus accurately detected sentences of suicidal adverse events regardless of their pretrained type and data set properties.

CONCLUSIONS

To our knowledge, this is the first and most extensive corpus of drug-suicide relations.

A corpus-assisted discourse study of Chinese university students' perceptions of sustainability

Education for sustainable development (ESD) in the higher education context plays a critical role in advancing the cause of sustainable development. However, previous research on university students' perceptions of sustainable development is limited. This study used a corpus-assisted eco-linguistic approach to investigate students' perceptions of sustainability issues and responsible actors to address these issues. This quantitative and qualitative study is based on a corpus of 501 collaborative essays on sustainability written by ~2,000 Chinese university students collected with their permission. The results show that the students had a comprehensive perception of the three dimensions of sustainable development. Environmental issues have received the most attention from students, followed by economic and social issues. With regard to perceived actors, students were inclined to view themselves as active participants in the cause of sustainable development, rather than as observers. They called for coordinated action of all relevant parties, such as the government, business sectors, institutions, and individuals. On the other hand, the author also noticed a tendency toward superficial green talk and anthropocentrism in students' discourse. This study aims to contribute to sustainability education by integrating findings into English as a foreign language (EFL) classes. Implications for sustainability education in the context of higher education are also discussed.

Applying data-driven learning in self-translation of academic discourse: A case study of a Chinese medical student

This article reports on an experiment on the use of data-driven learning (DDL) in the revision of self-translation by a Chinese medical student. The think-aloud method is employed to investigate the difficulties the student encountered in self-translation and the effectiveness of DDL in improving the quality of self-translation. Results show that difficulties in the self-translation of medical abstracts are mostly associated with markers of rhetorical moves, terminologies, and conventional academic expressions and that they can be effectively solved by such corpus consultation strategies as checking possible options in bilingual dictionaries, using the most certain keywords to find collocations, and using the most possible accompanying words to find contexts. A comparison of translations before and after the application of DDL reveals that it could help improve translation quality in lexical choices, syntactic structures, and discourse practice. An immediate interview shows that the participant holds a positive attitude toward DDL.

How interaction molds semantics: The mood functions of Chinese "sum-up" adverbs

"Hezhe" (), "ganqing" (), and "nao le bantian" () are common mood expressions in modern Chinese which have a common function of summarizing the information before (so called 'sum-up') as well as similar pragmatic functions. Study on these mood adverbs could reveal how the interactional mechanism molds the original semantic meanings of mood words and leads to new pragmatic functions. Five verbal corpora are applied to collect the real materials of usage containing the above three mood adverbs. Functional analysis and data statistics have been carried out to categorize the pragmatic functions of these words, calculate their distributions, and reconstruct their evolutional approach through an interactional perspective. We have found a core theoretical viewpoint that the similar functions of the three words emerged through the same pragmatic mechanism called "violation". These functions are: (1) "unexpectation" from the violation of psychological expectation; (2) "criticism" from the violation of universal principles; (3) "humor" from the violation of communicative principles. A statistic work of several corpora showed that these functions of the above three words appear broadly in verbal materials, with differences in their proportions according to the communication types and genres. In Chinese teaching to speakers of other languages, more attention should be paid to these words stemming from dialects, especially during intermediate and advanced levels.

Hierarchical Annotation for Building A Suite of Clinical Natural Language Processing Tasks: Progress Note Understanding

Applying methods in natural language processing on electronic health records (EHR) data is a growing field. Existing corpus and annotation focus on modeling textual features and relation prediction. However, there is a paucity of annotated corpus built to model clinical diagnostic thinking, a process involving text understanding, domain knowledge abstraction and reasoning. This work introduces a hierarchical annotation schema with three stages to address clinical text understanding, clinical reasoning, and summarization. We created an annotated corpus based on an extensive collection of publicly available daily progress notes, a type of EHR documentation that is collected in time series in a problem-oriented format. The conventional format for a progress note follows a Subjective, Objective, Assessment and Plan heading (SOAP). We also define a new suite of tasks, Progress Note Understanding, with three tasks utilizing the three annotation stages. The novel suite of tasks was designed to train and evaluate future NLP models for clinical text understanding, clinical knowledge representation, inference, and summarization.

L1 acquisition of the tense-aspect markers -ess (past-perfective) and -ko iss (imperfective) in Korean

This study investigated whether Korean children follow the acquisition pattern predicted by the Aspect Hypothesis (Shirai & Andersen, 1995), and the relationship between caretakers' and children's speech. Accordingly, we analyzed a Korean corpus (Ryu-Corpus) on the CHILDES database (MacWhinney, 2000), which comprised longitudinal video-recorded interactions of three Korean children and their caregivers. Results indicate that the children used the past marker -ess principally with telic verbs, consistent with the Aspect Hypothesis. Each child's usage closely reflects the caretaker's frequency, yielding a high correlation (τb = 0.79). However, the acquisition of the imperfective marker -ko iss did not show a predicted association with activity verbs, contrary to the Aspect Hypothesis. Furthermore, caretakers' input did not correlate with the children's utterances of the imperfective marker (τb = 0.40). We argue that multiple factors such as input frequency, language-specific organization of aspectual semantics, and individual differences should be considered to explain tense-aspect acquisition.

A Corpus-Based Comparison of the Chief Executive Officer Statements in Annual Reports and Corporate Social Responsibility Reports

This study presents a corpus-based comparison of the Chief Executive Officer (CEO) statements between annual reports and corporate social responsibility (CSR) reports. Using a corpus of 209 CEO statements from annual reports and CSR reports of Chinese companies, this study employs the Discourse-Historical Approach of critical discourse analysis to investigate the nomination strategies and key topics in these two related reports. The results showed that corporate leaders tend to have different priorities in annual reports and CSR reports. In annual reports, corporate leaders highlight the economic and pragmatic concerns of stakeholders to create a professionally capable and objective corporate image. In CSR reports, corporate leaders highlight the ethical concerns of stakeholders to create a socially responsible corporate image and adopt a more engaging and affiliative voice through the use of first-person pronouns. This study has significance in understanding the differences in the related genres of annual reports and CSR reports for the stakeholders.

Organisation of the musculature of the rat stomach

The strengths, directions and coupling of the movements of the stomach depend on the organisation of its musculature. Although the rat has been used as a model species to study gastric function, there is no detailed, quantitative study of the arrangement of the gastric muscles in rat. Here we provide a descriptive and quantitative account, and compare it with human gastric anatomy. The rat stomach has three components of the muscularis externa, a longitudinal coat, a circular coat and an internal oblique (sling) muscle in the region of the gastro-oesophageal junction. These layers are similar to human. Unlike human, the rat stomach is also equipped with paired muscular oesophago-pyloric ligaments that lie external to the longitudinal muscle. There is a prominent muscularis mucosae throughout the stomach and strands of smooth muscle occur in the mucosa, between the glands of the corpus and antrum. The striated muscle of the oesophageal wall reaches to the stomach, unlike the human, in which the wall of the distal oesophagus is smooth muscle. Thus, the continuity of gastric and oesophageal smooth muscle bundles, that occurs in human, does not occur in rat. Circular muscle bundles extend around the circumference of the stomach, in the fundus forming a cap of parallel muscle bundles. This arrangement favours co-ordinated circumferential contractions. Small bands of muscle make connections between the circular muscle bundles. This is consistent with a slower conduction of excitation orthogonal to the circular muscle bundles, across the corpus towards the distal antrum. The oblique muscle merged and became continuous with the circular muscle close to the gastro-oesophageal junction at the base of the fundus, and in the corpus, lateral to the lesser curvature. Quantitation of muscle thickness revealed gradients of thickness of both the longitudinal and circular muscle. This anatomical study provides essential data for interpreting gastric movements.

Extraction of Temporal Structures for Clinical Events in Unlabeled Free-Text Electronic Health Records in Russian

The important information about a patient is often stored in a free-form text to describe the events in the patient's medical history. In this work, we propose and evaluate a hybrid approach based on rules and syntactical analysis to normalise temporal expressions and assess uncertainty depending on the remoteness of the event. A dataset of 500 sentences was manually labelled to measure the accuracy. On this dataset, the accuracy of extracting temporal expressions is 95,5%, and the accuracy of normalization is 94%. The event extraction accuracy is 74.80%. The essential advantage of this work is the implementation of the considered approach for the non-English language where NLP tools are limited.

Stomach serosal arteries distinguish gastric regions of the rat

Because the stomach in situ has few distinctive surface features and changes shape dramatically with food intake, we have used micro-CT imaging combined with two distinct contrast agents to (1) characterize the pattern of arteries, potential landmarks, on the stomach wall and (2) evaluate how meal-related shape changes affect the size of the different regions. Images generated with a contrast agent injected directly into the heart during perfusion enabled a thorough look at the organizational features of the stomach angioarchitecture. The stomach receives its blood supply primarily from two pairs of vessels, the gastric and gastroepiploic arteries. Each of the three regions of the stomach is delineated by a distinctive combination of arterial fields: the corpus, consistent with its dynamic secretory activity and extensive mucosa, is supplied by extensive arterial trees formed by the left and right gastric arteries, travelling, respectively, on the ventral and dorsal stomach surfaces. These major arteries course circularly from the lesser towards the greater curvature, distally along both left (or ventral) and right (or dorsal) walls of the corpus, and branch rostrally to supply the region. The muscular antrum is characterized by smaller arterial branches arising primarily from the right gastroepiploic artery that follows the distal greater curvature and secondarily from small, distally directed arteries supplied by the large vessels of the left and right gastric arteries. The forestomach, essentially devoid of mucosal tissue and separated from the corpus by the limiting ridge, is vascularized predominantly by a network of small arteries issued from the left gastroepiploic artery coursing around the proximal greater curvature, as well as from higher order and smaller branches issued by the gastric and celiac arteries. These distinctive arterial fields appear to distinguish the major gastric regions, irrespective of the degree of fill of the stomach. Volume assessments of stomach compartments were made from images of iodine-stained stomachs. By varying the delay time between eating and perfusion, we were able to probe the emptying behavior of the stomach and demonstrate that the regions of the stomach empty at different rates, thus changing the relative dimensions of the organ regions. Notably, and despite these shape changes, the gastric arteries appear to form a regular, particularly recognizable, and lateralized pattern corresponding to the corpus that should be of use in guiding surgical and experimental interventions.

LitCovid-AGAC: cellular and molecular level annotation data set based on COVID-19

Currently, coronavirus disease 2019 (COVID-19) literature has been increasing dramatically, and the increased text amount make it possible to perform large scale text mining and knowledge discovery. Therefore, curation of these texts becomes a crucial issue for Bio-medical Natural Language Processing (BioNLP) community, so as to retrieve the important information about the mechanism of COVID-19. PubAnnotation is an aligned annotation system which provides an efficient platform for biological curators to upload their annotations or merge other external annotations. Inspired by the integration among multiple useful COVID-19 annotations, we merged three annotations resources to LitCovid data set, and constructed a cross-annotated corpus, LitCovid-AGAC. This corpus consists of 12 labels including Mutation, Species, Gene, Disease from PubTator, GO, CHEBI from OGER, Var, MPA, CPA, NegReg, PosReg, Reg from AGAC, upon 50,018 COVID-19 abstracts in LitCovid. Contain sufficient abundant information being possible to unveil the hidden knowledge in the pathological mechanism of COVID-19.

ECO-CollecTF: A Corpus of Annotated Evidence-Based Assertions in Biomedical Manuscripts

Analysis of high-throughput experiments in the life sciences frequently relies upon standardized information about genes, gene products, and other biological entities. To provide this information, expert curators are increasingly relying on text mining tools to identify, extract and harmonize statements from biomedical journal articles that discuss findings of interest. For determining reliability of the statements, curators need the evidence used by the authors to support their assertions. It is important to annotate the evidence directly used by authors to qualify their findings rather than simply annotating mentions of experimental methods without the context of what findings they support. Text mining tools require tuning and adaptation to achieve accurate performance. Many annotated corpora exist to enable developing and tuning text mining tools; however, none currently provides annotations of evidence based on the extensive and widely used Evidence and Conclusion Ontology. We present the ECO-CollecTF corpus, a novel, freely available, biomedical corpus of 84 documents that captures high-quality, evidence-based statements annotated with the Evidence and Conclusion Ontology.

A Retrospective Experience of Helicobacter pylori Histology in a Large Sample of Subjects in Northern Italy

Updated data about the prevalence of Helicobacter pylori (H. pylori) and its correlation with histological results are scarce. The aim of our study was to provide current data on the impact of H. pylori in a third-level endoscopy service. We performed a large, retrospective study analyzing the results of all histological samples of gastroscopy from the year 2019. In total, 1512 subjects were included. The prevalence of H. pylori was 16.8%. A significant difference between the prevalence in subjects born in Italy and those from eastern Europe, south America, or Africa was found (p < 0.0001, p = 0.006, and p = 0.0006, respectively). An association was found between H. pylori and active superficial gastritis (p < 0.0001). Current H. pylori and/or a previous finding of H. pylori was related to antral atrophy (p < 0.0001). Fifteen patients had low-grade dysplasia. There were no statistically significant associations with current or past H. pylori infection. One patient presented gastric cardia adenocarcinoma with regular gastric mucosa. One patient, H. pylori positive, was diagnosed with gastric signet ring cell adenocarcinoma in a setting of diffuse atrophy, without metaplasia.. Our study provides updated, solid (biopsy diagnosis and large population) data on the prevalence of H. pylori infection in a representative region of southern Europe.

A novel corpus-based computing method for handling critical word-ranking issues: An example of COVID-19 research articles

A corpus is a massive body of structured textual data that are stored and operated electronically. It usually combines with statistics, machine learning algorithms, or artificial intelligence (AI) technologies to explore the semantic relationship between lexical units, and beneficial when applied to language learning, information processing, translation, and so forth. In the face of a novel disease, like, COVID-19, establishing medical-specific corpus will enhance frontline medical personnel's information acquisition efficiency, guiding them on the right approaches to respond to and prevent the novel disease. To effectively retrieve critical messages from the corpus, appropriately handling word-ranking issues is quite crucial. However, traditional frequency-based approaches may cause bias in handling word-ranking issues because they neither optimize the corpus nor integrally take words' frequency dispersion and concentration criteria into consideration. Thus, this paper develops a novel corpus-based approach that combines a corpus software and Hirsch index (H-index) algorithm to handle the aforementioned issues simultaneously, making word-ranking processes more accurate. This paper compiled 100 COVID-19-related research articles as an empirical example of the target corpus. To verify the proposed approach, this study compared the results of two traditional frequency-based approaches and the proposed approach. The results indicate that the proposed approach can refine corpus and simultaneously compute words' frequency dispersion and concentration criteria in handling word-ranking issues.

Modeling Input Factors in Second Language Acquisition of the English Article Construction

Based on the Competition Model, the current study investigated how cue availability and cue reliability as two important input factors influenced second language (L2) learners' cue learning of the English article construction. Written corpus data of university-level Chinese-L1 learners of English were sampled for a comparison of English majors and non-English majors who demonstrated two levels of L2 competence in English article usage. The path model analysis in structural equation modeling was utilized to investigate the relationship between the input factors and L2 usage (frequency and accuracy of article cue production). The findings contribute novel and scarce empirical evidence that confirms a central claim of the Competition Model, i.e., the changing importance of cue availability and cue reliability in the frequency and accuracy of production. Cue availability was found to determine L2 production frequency regardless of level of L2 competence. Cue reliability was the input factor that differentiated competence levels. When learners stayed at a relatively lower L2 proficiency, cue reliability played an important role in influencing L2 frequency of usage rather than accuracy of usage. When learners developed increased exposure to and stronger competence in the target language, cue reliability played a significant role in determining learners' success of cue learning. The study is methodologically innovative and expands the empirical applicability of the Competition Model to the domain of second language production and construction learning.

Helicobacter pylori and association between its positivity and anatomotopographic settlement in the stomach with the host age range

Helicobacter pylori (H. pylori) is a Gram-negative, helically shaped flagellated bacterium. Major diseases associated with H. pylori infection include peptic ulcer, gastric adenocarcinoma, and mucosa-associated lymphoid tissue lymphoma. The incidence of H. pylori in the anatomotopographic regions of the stomach, such as antrum, corpus, fundus, and incisura angularis, has been investigated. Do the rates of H. pylori in the settlements change over time according to the age ranges of the hosts? Does this change affect the diseases caused by or related to H. pylori? It is estimated that the outcomes, which have been obtained, may provide a new perspective in terms of understanding the etiopathogenesis of H. pylori-induced diseases. A comprehensive literature search of PubMed/MEDLINE databases had been conducted using a combination of terms, “Helicobacter pylori,” “Sydney System,” “stomach,” “pyloric antrum,” “gastric corpus,” “stomach cancer,” and “Helicobacter pylori and age.” There are very few articles examining the relationship between the topographic locations of H. pylori and host age range in the English language literature. Therefore, it is also purposed to emphasize the outcomes of our current research about the mentioned topic. In our opinion, similar studies should reveal the settlement and age range in the different geographic locations and societies as in our study. We believe that these findings will contribute to the efforts for understanding overtly of H. pylori-induced disease of the stomach.

Introducing the Ko Corpus of Korean Mother-Child Interaction

We describe a corpus of speech taking place between 30 Korean mother–child pairs, divided in three groups of Prelexical (M = 0;08), Early-Lexical (M = 1;02), and Advanced-Lexical (M = 2;03). In addition to the child-directed speech (CDS), this corpus includes two different formalities of adult-directed speech (ADS), i.e., family-directed ADS (ADS_Fam) and experimenter-directed ADS (ADS_Exp). Our analysis of the MLU in CDS, family-, and experimenter-directed ADS found significant differences between CDS and ADS_Fam, and between ADS_Fam and ADS_Exp, but not between CDS and ADS_Exp. Our finding suggests that researchers should pay more attention to controlling the level of formality in CDS and ADS when comparing the two registers for their speech characteristics. The corpus was transcribed in the CHAT format of the CHILDES system, so users can easily extract data related to verbal behavior in the mother–child interaction using the CLAN program of CHILDES.

Common Ground in Demonstrative Reference: The Case of Mano (Mande)

That demonstratives often have endophoric functions marking referents outside the physical space of interaction but accessible through cognition, especially memory, is well-known. These functions are often classified as independent from exophoric ones and are typically seen as secondary with respect to spatial deixis. However, data from multiple languages show that cognitive access to referents functions alongside of perceptual access, including vision. Cognitive access is enabled by prior interactions and prior familiarity with the referents. As a result of such interactions, the interlocutors share a great deal of knowledge about the referents, which facilitates reference to objects in the interactive field. The centrality of common ground in reference to an object at the interactive scene challenges the often assumed classification of demonstrative reference into exophoric and endophoric. I illustrate this idea throughout the paper by using first-hand data from Mano, a Mande language of Guinea. Adding another argument in favor of viewing demonstrative reference as a social, interactive process, the Mano data push the idea of salience of non-spatial parameters further and emphasizes the importance of short and long-term interactional history and cultural knowledge both for the choice of demonstratives in exophoric reference and for the structuring of the demonstrative paradigm.

A Novel GAN-Based Synthesis Method for In-Air Handwritten Words

In recent years, with the miniaturization and high energy efficiency of MEMS (micro-electro-mechanical systems), in-air handwriting technology based on inertial sensors has come to the fore. Most of the previous works have focused on character-level in-air handwriting recognition. In contrast, few works focus on word-level in-air handwriting tasks. In the field of word-level recognition, researchers have to face the problems of insufficient data and poor generalization performance of recognition methods. On one hand, the training of deep neural learning networks usually requires a particularly large dataset, but collecting data will take a lot of time and money. On the other hand, a deep recognition network trained on a small dataset can hardly recognize samples whose labels do not appear in the training set. To address these problems, we propose a two-stage synthesis method of in-air handwritten words. The proposed method includes a splicing module guided by an additional corpus and a generating module trained by adversarial learning. We carefully design the proposed network so that it can handle word sample inputs of arbitrary length and pay more attention to the details of the samples. We design multiple sets of experiments on a public dataset. The experimental results demonstrate the success of the proposed method. What is impressive is that with the help of the air-writing word synthesizer, the recognition model learns the context information (combination information of characters) of the word. In this way, it can recognize words that have never appeared in the training process. In this paper, the recognition model trained on synthetic data achieves a word-level recognition accuracy of 62.3% on the public dataset. Compared with the model trained using only the public dataset, the word-level accuracy is improved by 62%. Furthermore, the proposed method can synthesize realistic samples under the condition of limited of in-air handwritten character samples and word samples. It largely solves the problem of insufficient data. In the future, mathematically modeling the strokes between characters in words may help us find a better way to splice character samples. In addition, we will apply our method to various datasets and improve the splicing module and generating module for different tasks.

Contribution of the Semiological Approach to Deixis-Anaphora in Sign Language: The Key Role of Eye-Gaze

We address the issue of deixis–anaphora in sign language (SL) discourse, focusing on the role of eye-gaze. According to the Semiological Approach, SL structuring stems from a maximum exploitation of the visuo-gestural modality, which results in two modes of meaning production, depending on the signer’s semiotic intent. Involving both non-manual and manual parameters, the first mode, expressing the intent to say while showing, uses constructions based on structures, the termed “transfer structures.” The second one, expressing the intent to say without showing, involves lexical, pointing and fingerspelling units. In order to situate our descriptive concepts with respect to those used by SL linguists who, like us, adopt a cognitive–functionalist perspective, we expose a specific theoretical foundation of our approach, the “enunciation theories.” The concept of “enunciation” is decisive for understanding the role of eye-gaze, as being at the foundation of deixis and the key vector of referential creation and tracking in SL discourse. “Enunciation” entails the opposition between “Enunciation” and “Utterance” Domains. The first links, as co-enunciators, the signer/speaker and his/her addressee, establishing them by the very “act of enunciation” as 1st and 2nd person. The second is internal to the discourse produced. Grounding on corpora of narratives in several SLs (some with no historical link), we illustrate this crucial role of eye-gaze and the diversity of functions it fulfills. Our analyses, carried out in this perspective, attest to the multiple structural similarities between SLs, particularly with regard to transfer structures. This result strongly supports the typological hypothesis underlying our approach, namely, that these structures are common to all SLs. We thus show that an enunciative analysis, based on the key role of eye-gaze in these visual languages that are SLs, is able to give the simplest account of their own linguistic economy and, in particular, of deixis–anaphora in these languages.

Organizing an in-class hackathon to correct PDF-to-text conversion errors of Genomics & Informatics 1.0

This paper describes a community effort to improve earlier versions of the full-text corpus of Genomics & Informatics by semi-automatically detecting and correcting PDF-to-text conversion errors and optical character recognition errors during the first hackathon of Genomics & Informatics Annotation Hackathon (GIAH) event. Extracting text from multi-column biomedical documents such as Genomics & Informatics is known to be notoriously difficult. The hackathon was piloted as part of a coding competition of the ELTEC College of Engineering at Ewha Womans University in order to enable researchers and students to create or annotate their own versions of the Genomics & Informatics corpus, to gain and create knowledge about corpus linguistics, and simultaneously to acquire tangible and transferable skills. The proposed projects during the hackathon harness an internal database containing different versions of the corpus and annotations.

The acquisition and use of relative clauses in Turkish-learning children's conversational interactions: a cross-linguistic approach

Using a cross-linguistic approach, we investigated Turkish-speaking children's acquisition and use of relative clauses (RCs) by examining longitudinal child-caregiver interactions and cross-sectional peer conversations. Longitudinal data were collected from 8 children between the ages of 8 and 36 months. Peer conversational corpus came from 78 children aged between 43 and 64 months. Children produced RCs later than in English (Diessel, 2004) and Mandarin (Chen & Shirai, 2015), and demonstrated increasing semantic and structural complexity with age. Despite the morphosyntactic difficulty of object RCs, and prior experimental findings showing a subject RC advantage, preschool-aged children produced object RCs, which were highly frequent in child-directed speech, as frequently as subject RCs. Object RCs in spontaneous speech were semantically less demanding (with pronominal subjects and inanimate head nouns) than the stimuli used in prior experiments. Results suggest that multiple factors such as input frequency and morphosyntactic and semantic difficulty affect the acquisition patterns.

Region-specific microRNA signatures in the human epididymis

The epithelium of the human epididymis maintains an appropriate luminal environment for sperm maturation that is essential for male fertility. Regional expression of small noncoding RNAs such as microRNAs contributes to segment-specific gene expression and differentiated functions. MicroRNA profiles were reported in human epididymal tissues but not specifically in the epithelial cells derived from those regions. Here, we reveal miRNA signatures of primary cultures of caput, corpus, and cauda epididymis epithelial cells and of the tissues from which they were derived. We identify 324 epithelial cell-derived microRNAs and 259 tissue-derived microRNAs in the epididymis, some of which displayed regionalized expression patterns in cells and/or tissues. Caput cell-enriched miRNAs included miR-573 and miR-155. Cauda cell-enriched miRNAs included miR-1204 and miR-770. Next, we determined the gene ontology pathways associated with in silico predicted target genes of the differentially expressed miRNAs. The effect of androgen receptor stimulation on miRNA expression was also investigated. These data show novel epithelial cell-derived miRNAs that may regulate the expression of important gene networks that are responsible for the regionalized gene expression and function of the epididymis.

Tonality Tunes the Statistical Characteristics in Music: Computational Approaches on Statistical Learning

Statistical learning is a learning mechanism based on transition probability in sequences such as music and language. Recent computational and neurophysiological studies suggest that the statistical learning contributes to production, action, and musical creativity as well as prediction and perception. The present study investigated how statistical structure interacts with tonalities in music based on various-order statistical models. To verify this in all 24 major and minor keys, the transition probabilities of the sequences containing the highest pitches in Bach's Well-Tempered Clavier, which is a collection of two series (No. 1 and No. 2) of preludes and fugues in all of the 24 major and minor keys, were calculated based on nth-order Markov models. The transition probabilities of each sequence were compared among tonalities (major and minor), two series (No. 1 and No. 2), and music types (prelude and fugue). The differences in statistical characteristics between major and minor keys were detected in lower- but not higher-order models. The results also showed that statistical knowledge in music might be modulated by tonalities and composition periods. Furthermore, the principal component analysis detected the shared components of related keys, suggesting that the tonalities modulate statistical characteristics in music. The present study may suggest that there are at least two types of statistical knowledge in music that are interdependent on and independent of tonality, respectively.

Cellular and molecular characterization of a thick-walled variant reveal a pivotal role of shoot apical meristem in transverse development of bamboo culm

Little is known about the mechanisms underlying the development of bamboo culm. Using anatomical, mathematical modeling, and genomics methods, we investigated the role of shoot apical meristem (SAM) in the development of the transverse morphology of bamboo culm and explored the underlying cellular and molecular processes. We discovered that maintenance of SAM morphology that can produce circular culm and increase in SAM cell numbers, especially corpus cells, is the means by which bamboo makes a larger culm with a regular pith cavity and culm wall during development. A less cellular form of SAM with a lower proportion of corpus cells causes an abnormal higher ratio of wall component cells to pith cells, which breaks the balance of their interaction and triggers the random invasion of wall component cells into pith tissues during development, and finally results in the various thick culm walls of Phyllostachys nidularia f. farcta. The smaller SAM also results in a lower level of hormones such as cytokinin and auxin, and down-regulates hormone signaling and the downstream functional genes such as those related to metabolism, which finally results in a dwarf and smaller diameter culm with lower biomass. These results provide an important perspective on the culm development of bamboo, and support a plausible mechanism causing the size-reduced culm and various thick culm walls of P. nidularia f. farcta.

Sources of Microtemporal Clustering in Sociolinguistic Sequences

Persistence is the tendency of speakers to repeat the choice of sociolinguistic variant they have recently made in conversational speech. A longstanding debate is whether this tendency toward repetitiveness reflects the direct influence of one outcome on the next instance of the variable, which I call sequential dependence, or the shared influence of shifting contextual factors on proximal instances of the variable, which I call baseline deflection. I propose that these distinct types of clustering make different predictions for sequences of variable observations that are longer than the typical prime-target pairs of typical corpus persistence studies. In corpus ING data from conversational speech, I show that there are two effects to be accounted for: an effect of how many times the /ing/ variant occurs in the 2, 3, or 4-token sequence prior to the target (regardless of order), and an effect of whether the immediately prior (1-back) token was /ing/. I then build a series of simulations involving Bernoulli trials at sequences of different probabilities that incorporate either a sequential dependence mechanism, a baseline deflection mechanism, or both. I argue that the model incorporating both baseline deflection and sequential dependence is best able to produce simulated data that shares the relevant properties of the corpus data, which is an encouraging outcome because we have independent reasons to expect both baseline deflection and sequential dependence to exist. I conclude that this exploratory analysis of longer sociolinguistic sequences reflects a promising direction for future research on the mechanisms involved in the production of sociolinguistic variation.

Transcriptional networks in the human epididymis

BACKGROUND

The epithelial lining of the human epididymis is critical for sperm maturation. This process requires distinct specialized functions in the head, body, and tail of the duct. These region-specific properties are maintained by distinct gene expression profiles which are governed by transcription factor networks, non-coding RNAs, and other factors.

MATERIALS AND METHODS

We used genome-wide protocols including DNase-seq, RNA-seq and ChIP-seq to characterize open (active) chromatin, the transcriptome and occupancy of specific transcription factors (TFs) respectively, in caput, corpus, and cauda segments of adult human epididymis tissue and primary human epididymis epithelial (HEE) cell cultures derived from them. RNA-seq following TF depletion or activation, combined with gene ontology analysis also determined TF targets.

RESULTS

Among regional differentially expressed transcripts were epithelial-selective transcription factors (TFs), microRNAs, and antiviral response genes. Caput-enriched TFs included hepatocyte nuclear factor 1 (HNF1) and the androgen receptor (AR), both of which were also predicted to occupy cis-regulatory elements identified as open chromatin in HEE cells. HNF1 targets were identified genome-wide using ChIP-seq, in HEE cells. Next, siRNA-mediated depletion of HNF1 revealed a pivotal role for this TF in coordinating epithelial water and solute transport in caput epithelium. The importance of AR in HEE cells was shown by AR ChIP-seq, and by RNA-seq after synthetic androgen (R1881) treatment. AR has a distinct transcriptional program in the HEE cells and likely recruits different co-factors (RUNX1 and CEBPβ) in comparison to those used in prostate epithelium.

DISCUSSION AND CONCLUSION

Our data identify many transcription factors that regulate the development and differentiation of HEE cells. Moreover, a comparison between immature and adult HEE cells showed key TFs in the transition to fully differentiated function of this epithelium. These data may help identify new targets to treat male infertility and have the potential to open new avenues for male contraception.

Miliary Lung Metastases from Genital Large Cell Neuroendocrine Carcinomas

We herein report two cases of miliary lung metastases from genital carcinoma in uterine cervix and endometrium. Notably, these patients were unable to receive any anti-tumor chemotherapy due to rapid progression causing respiratory failure, and they ultimately died of disease progression within only a month after the first visit to our hospitals. A postmortem examination confirmed the diagnosis of genital large-cell neuroendocrine carcinoma (LCNEC). Chest physicians should be aware of genital LCNEC with a dismal prognostic entity as an important differential diagnosis of miliary lung metastases.

Cytotoxic Lesion of the Corpus Callosum Caused by Puumala Hantavirus Infection

We report the case of a 45-year-old male referred to our hospital with fever, asthenia, visual disturbances and increasing headaches. Diffusion-weighted imaging of the brain showed high signal intensity in the splenium of corpus callosum with low apparent diffusion coefficient values. Diagnosis of cytotoxic lesion of corpus callosum was made with Puumala Hantavirus infection serologically confirmed and should not be mistaken for ischemia. Patient was discharged 8 days after admission and imaging findings had resolved 3 weeks later.

Systematic review of hormonal and genetic factors involved in the nonsyndromic disorders of the lower jaw

Mandibular disorders are among the most common birth defects in humans, yet the etiological factors are largely unknown. Most of the neonates affected by mandibular abnormalities have a sequence of secondary anomalies, including airway obstruction and feeding problems, that reduce the quality of life. In the event of lacking corrective surgeries, patients with mandibular congenital disorders suffer from additional lifelong problems such as sleep apnea and temporomandibular disorders, among others. The goal of this systematic review is to gather evidence on hormonal and genetic factors that are involved in signaling pathways and interactions that are potentially associated with the nonsyndromic mandibular disorders. We found that members of FGF and BMP pathways, including FGF8/10, FGFR2/3, BMP2/4/7, BMPR1A, ACVR1, and ACVR2A/B, have a prominent number of gene-gene interactions among all identified genes in this review. Gene ontology of the 154 genes showed that the functional gene sets are involved in all aspects of cellular processes and organogenesis. Some of the genes identified by the genome-wide association studies of common mandibular disorders are involved in skeletal formation and growth retardation based on animal models, suggesting a potential direct role as genetic risk factors in the common complex jaw disorders. Developmental Dynamics 248:162-172, 2019. © 2018 Wiley Periodicals, Inc.

Opinion: Strategy of Semi-Automatically Annotating Full Text Corpus of Genomics &amp; Informatics

There is a communal need for an annotated corpus consisting of the full texts of biomedical journal articles. In response to community needs, a prototype version of the full-text corpus of Genomics & Informatics, called GNI version 1.0, has recently been published, with 499 annotated full-text articles available as a corpus resource. However, GNI needs to be updated, as the texts were shallow-parsed and annotated with several existing parsers. I list issues associated with upgrading annotations and give an opinion on the methodology for developing the next version of the GNI corpus, based on a semi-automatic strategy for more linguistically rich corpus annotation.

Information Structure Preferences in Focus-Sensitive Ellipsis: How Defaults Persist

We compare the roles of overt accent and default focus marking in processing ellipsis structures headed by focus-sensitive coordinators (such as Danielle couldn't pass the quiz, let alone the final/Kayla). In a small auditory corpus study of radio transcripts, we establish that such structures overwhelmingly occur with contrastive pitch accents on the correlate and remnant ( the quiz and the final, or Danielle and Kayla), and that there is a strong bias to pair the remnant with the most local plausible correlate in production. In two auditory naturalness ratings experiments, we observe that marking a non-local correlate with contrastive pitch accent moderates, but does not fully overturn, the bias for local correlates in comprehension. We propose that the locality preference is due to a sentence-final default position for sentence accent, and that auditory processing is subject to "enduring focus," in which default positions for focus continue to influence the focus structure of the sentence even in the presence of overt accents. The importance of these results for models of auditory processing and of the processing of remnants in ellipsis structures is discussed.

Structure and Grammaticalization of Serial Verb Constructions in Sign Language of the Netherlands-A Corpus-Based Study

In serial verb constructions (SVCs), multiple independent lexical verbs are combined in a mono-clausal construction. SVCs express a range of grammatical meanings and are attested in numerous spoken languages all around the world. Yet, to date only few studies have investigated the existence and functions of SVCs in sign languages. For the most part, these studies—including a previous study on Sign Language of the Netherlands (NGT)—relied on elicited data. In this article, we offer a cross-modal typological contribution to the study of SVCs by investigating the phenomenon based on naturalistic corpus data from NGT. A search of the Corpus NGT yielded 41 mono-clausal utterances in which one of a closed set of verbs—namely go, give, take, and call—combines with another lexical verb. While the combinations we found are in important respects reminiscent of SVCs described for spoken languages, our data also confirm the previous finding that the fixed verb in the SVC serves to express agreement (by means of spatial modulation) when the other verb cannot do so. In addition, we identified some novel uses of the verbs go and give: (i) go functioning as a future tense marker and (ii) give functioning as a light verb. We will also discuss aspects of the grammaticalization of SVCs in NGT: from lexical verb to light verb to auxiliary, again offering some comparison to grammaticalization paths described for spoken languages.

TwiMed: Twitter and PubMed Comparable Corpus of Drugs, Diseases, Symptoms, and Their Relations

BACKGROUND

Work on pharmacovigilance systems using texts from PubMed and Twitter typically target at different elements and use different annotation guidelines resulting in a scenario where there is no comparable set of documents from both Twitter and PubMed annotated in the same manner.

OBJECTIVE

This study aimed to provide a comparable corpus of texts from PubMed and Twitter that can be used to study drug reports from these two sources of information, allowing researchers in the area of pharmacovigilance using natural language processing (NLP) to perform experiments to better understand the similarities and differences between drug reports in Twitter and PubMed.

METHODS

We produced a corpus comprising 1000 tweets and 1000 PubMed sentences selected using the same strategy and annotated at entity level by the same experts (pharmacists) using the same set of guidelines.

RESULTS

The resulting corpus, annotated by two pharmacists, comprises semantically correct annotations for a set of drugs, diseases, and symptoms. This corpus contains the annotations for 3144 entities, 2749 relations, and 5003 attributes.

CONCLUSIONS

We present a corpus that is unique in its characteristics as this is the first corpus for pharmacovigilance curated from Twitter messages and PubMed sentences using the same data selection and annotation strategies. We believe this corpus will be of particular interest for researchers willing to compare results from pharmacovigilance systems (eg, classifiers and named entity recognition systems) when using data from Twitter and from PubMed. We hope that given the comprehensive set of drug names and the annotated entities and relations, this corpus becomes a standard resource to compare results from different pharmacovigilance studies in the area of NLP.

A Panel of Serum Biomarkers (GastroPanel®) in Non-invasive Diagnosis of Atrophic Gastritis. Systematic Review and Meta-analysis

BACKGROUND/AIM

To meet the increasing demand of non-invasive tests for screening of gastric cancer (GC) risk, biomarker panel (GastroPanel®) (GP) was designed by Biohit Oyj as the first serological test for stomach health. The aim of the present study was to perform a systematic review and meta-analysis of all studies on GP in diagnosis of atrophic gastritis (AG).

MATERIALS AND METHODS

Studies were eligible, if i) GP was used to diagnose biopsy-confirmed AG of the corpus (AGC) and/or antrum (AGA) and ii) exact numbers were available to enable calculating sensitivity (SE) and specificity (SP). Comprehensive Meta-Analysis software was used with maximum likelihood meta-regression (R² analog). Effect size estimates (SE; SP, 95% confidence interval (CI)) were tested for homogeneity with Cochran's Q and I² statistics. Potential publication bias was estimated by funnel plot statistics.

RESULTS

Altogether, 27 studies were eligible comprising of 8,654 patients from different geographic regions. Significant heterogeneity between studies reporting AGC (n=27) or AGA (n=13) warranted random effects (RE) model for summary statistics. GP performs better in diagnosing AGC than AGA with 70.2% vs. 51.6% pooled SE and 93.9% vs. 84.1% pooled SP, respectively. Limited number of studies erodes the Q test's power to detect true heterogeneity in meta-analysis stratified by geographic study origin. Few hypothetical missing studies had only marginal effect on pooled estimates of SE and SP.

CONCLUSION

This first meta-analysis of GP literature corroborates the statement of international experts, advocating GP in diagnosis and screening of AG. Due to its high specificity for both AGA and AGC, GastroPanel® is truly a test for stomach health.

Generating a Corpus of Mobile Forensic Images for Masquerading user Experimentation

The Periodic Mobile Forensics (PMF) system investigates user behavior on mobile devices. It applies forensic techniques to an enterprise mobile infrastructure, utilizing an on-device agent named TractorBeam. The agent collects changed storage locations for later acquisition, reconstruction, and analysis. TractorBeam provides its data to an enterprise infrastructure that consists of a cloud-based queuing service, relational database, and analytical framework for running forensic processes. During a 3-month experiment with Purdue University, TractorBeam was utilized in a simulated operational setting across 34 users to evaluate techniques to identify masquerading users (i.e., users other than the intended device user). The research team surmises that all masqueraders are undesirable to an enterprise, even when a masquerader lacks malicious intent. The PMF system reconstructed 821 forensic images, extracted one million audit events, and accurately detected masqueraders. Evaluation revealed that developed methods reduced storage requirements 50-fold. This paper describes the PMF architecture, performance of TractorBeam throughout the protocol, and results of the masquerading user analysis.

Prevalence of H. pylori Infection and Atrophic Gastritis in a Population-based Screening with Serum Biomarker Panel (GastroPanel®) in St. Petersburg

BACKGROUND/AIM

Russian Federation is among the high-incidence countries for gastric cancer (GC), with the incidence being projected to continue increasing. Using a non-invasive blood test with four stomach-specific biomarkers (pepsinogen-I (PG-I) and -II (PG-II), amidated gastrin-17 (G-17) and Helicobacter pylori (HP) IgG antibodies) in a hospital-based screening setting, we aimed to determine the prevalence of GC risk conditions: HP-infection and atrophic gastritis (AG).

PATIENTS AND METHODS

A population-derived cohort of 918 asymptomatic subjects (646 women and 272 men) with a mean age of 51.8 years (range=26-83) was examined with the GastroPanel® (GP) test. GP results were verified by gastroscopy and biopsies (the Updated Sydney System (USS) classification for all test-positive AG cases and for random 5% test-negatives (n=263) to correct for the verification bias.

RESULTS

Of the 918 subjects, only 199 (21.7%) tested completely normal, while 76.7% (704/918) had HP-infection. Altogether, in 99 subjects (10.8%), GP suggested AG: atrophic gastritis in the antrum (AGA) (n=21), atrophic gastritis in the corpus (AGC) (n=69) or atrophic pangastritis (AGpan) (n=9). The overall concordance between GP and USS classification was 82.5% (217/263) with weighted kappa intra-class correlation coefficient (ICC)=0.875 (95% confidence interval (CI)=0.840-0.901). The sensitivity/specificity balance in receiver operating characteristic (ROC) analysis for PG-I as a marker of moderate/severe AGC (AGC2+) had area under the curve (AUC)=0.895 (95%CI=0.837-0.953). Using the AGC2+ end-point, verification bias-corrected specificity of PGI reached 96.4% (95%CI=94.7-97.9) and that of PGI/PGII ratio 94.6% (95%CI=92.6-96.3), with inevitable erosion in sensitivities.

CONCLUSION

While capable of detecting the subjects at risk for GC (HP and/or AG), GP should be the cost-effective means to break the current ominous trend in GC incidence in Russian Federation.

SuperCAT: The (New and Improved) Corpus Analysis Toolkit

This paper reports SuperCAT, a corpus analysis toolkit. It is a radical extension of SubCAT, the Sublanguage Corpus Analysis Toolkit, from sublanguage analysis to corpus analysis in general. The idea behind SuperCAT is that representative corpora have no tendency towards closure-that is, they tend towards infinity. In contrast, non-representative corpora have a tendency towards closure-roughly, finiteness. SuperCAT focuses on general techniques for the quantitative description of the characteristics of any corpus (or other language sample), particularly concerning the characteristics of lexical distributions. Additionally, SuperCAT features a complete re-engineering of the previous SubCAT architecture.

Expression profiles of human epididymis epithelial cells reveal the functional diversity of caput, corpus and cauda regions

STUDY HYPOTHESIS

Region-specific transcriptional profiling of tissues and cultured epithelial cells from the human epididymis will predict functional specialization along the duct.

STUDY FINDING

We identified the molecular signature driving functions of the caput, corpus and cauda epithelium, and determined how these differ to establish the regional differentiation of the organ.

WHAT IS KNOWN ALREADY

The epithelium lining the human male genital ducts has a critical role in fertility. In particular, it controls the luminal environment in the epididymis, which is required for normal sperm maturation and reproductive competence. Studies in many animal species have largely informed our understanding of the molecular basis of epididymis function. However, there are substantial differences between species.

STUDY DESIGN, SAMPLES/MATERIALS, METHODS

Using RNA sequencing on biological replicates, we described gene expression profiles for tissue from each region of the epididymis and cultured epithelial cells derived from these regions. Bioinformatic tools were then utilized to identify differentially expressed genes (DEGs) between tissues and cells from the caput, corpus and cauda.

MAIN RESULTS AND THE ROLE OF CHANCE

The data showed that the caput is functionally divergent from the corpus and cauda, which have very similar transcriptomes. Interrogation of DEGs using gene ontology process enrichment analyses showed that processes of ion transport, response to hormone stimulus and urogenital tract development are more evident in the caput, while defense response processes are more important in the corpus/cauda. Consistent with these regional differences in epididymis function, we observed differential expression of transcription factors in the caput and corpus/cauda.

LIMITATIONS, REASONS FOR CAUTION

Cultured caput, corpus and cauda cells may not faithfully represent the same cells in the intact organ, due to loss of hormonal signals from the testis and communication from other cell types.

WIDER IMPLICATIONS OF THE FINDINGS

Our data provide a molecular characterization that will facilitate advances in understanding human epididymis epithelium biology in health and disease. They may also reveal the mechanisms coordinating epididymis luminal environment and sperm maturation.

LARGE SCALE DATA

Data deposited at http://www.ncbi.nlm.nih.gov/geo/GSE72986.

STUDY FUNDING AND COMPETING INTERESTS

This work was supported by the National Institutes of Health: R01HD068901 (PI: A.H.). The authors declare no conflict of interest.

Vagal Intramuscular Arrays: The Specialized Mechanoreceptor Arbors That Innervate the Smooth Muscle Layers of the Stomach Examined in the Rat

The fundamental roles that the stomach plays in ingestion and digestion notwithstanding, little morphological information is available on vagal intramuscular arrays (IMAs), the afferents that innervate gastric smooth muscle. To characterize IMAs better, rats were given injections of dextran biotin in the nodose ganglia, and, after tracer transport, stomach whole mounts were collected. Specimens were processed for avidin-biotin permanent labeling, and subsets of the whole mounts were immunohistochemically processed for c-Kit or stained with cuprolinic blue. IMAs (n = 184) were digitized for morphometry and mapping. Throughout the gastric muscle wall, IMAs possessed common phenotypic features. Each IMA was generated by a parent neurite arborizing extensively, forming an array of multiple (mean = 212) branches averaging 193 µm in length. These branches paralleled, and coursed in apposition with, bundles of muscle fibers and interstitial cells of Cajal. Individual arrays averaged 4.3 mm in length and innervated volumes of muscle sheet, presumptive receptive fields, averaging 0.1 mm(3) . Evaluated by region and by muscle sheet, IMAs displayed architectural adaptations to the different loci. A subset (32%) of circular muscle IMAs issued specialized polymorphic collaterals to myenteric ganglia, and a subset (41%) of antral longitudinal muscle IMAs formed specialized net endings associated with the serosal boundary. IMAs were concentrated in regional patterns that correlated with the unique biomechanical adaptations of the stomach, specifically proximal stomach reservoir functions and antral emptying operations. Overall, the structural adaptations and distributions of the IMAs were consonant with the hypothesized stretch receptor roles of the afferents.