Sequential Expectations: The Role of Prediction‐Based Learning in Language

Prediction‐based processes appear to play an important role in language. Few studies, however, have sought to test the relationship within individuals between prediction learning and natural language processing. This paper builds upon existing statistical learning work using a novel paradigm for studying the on‐line learning of predictive dependencies. Within this paradigm, a new “prediction task” is introduced that provides a sensitive index of individual differences for developing probabilistic sequential expectations. Across three interrelated experiments, the prediction task and results thereof are used to bridge knowledge of the empirical relation between statistical learning and language within the context of nonadjacency processing. We first chart the trajectory for learning nonadjacencies, documenting individual differences in prediction learning. Subsequent simple recurrent network simulations then closely capture human performance patterns in the new paradigm. Finally, individual differences in prediction performances are shown to strongly correlate with participants’ sentence processing of complex, long‐distance dependencies in natural language.     

The impact of iconic gestures on foreign language word learning and its neural substrate

Vocabulary acquisition represents a major challenge in foreign language learning. Research has demonstrated that gestures accompanying speech have an impact on memory for verbal information in the speakers' mother tongue and, as recently shown, also in foreign language learning. However, the neural basis of this effect remains unclear. In a within‐subjects design, we compared learning of novel words coupled with iconic and meaningless gestures. Iconic gestures helped learners to significantly better retain the verbal material over time. After the training, participants' brain activity was registered by means of fMRI while performing a word recognition task. Brain activations to words learned with iconic and with meaningless gestures were contrasted. We found activity in the premotor cortices for words encoded with iconic gestures. In contrast, words encoded with meaningless gestures elicited a network associated with cognitive control. These findings suggest that memory performance for newly learned words is not driven by the motor component as such, but by the motor image that matches an underlying representation of the word's semantics. Hum Brain Mapp, 2011. © 2010 Wiley‐Liss, Inc.     

Semi-supervised medical entity recognition: A study on Spanish and Swedish clinical corpora

ObjectiveThe goal of this study is to investigate entity recognition within Electronic Health Records (EHRs) focusing on Spanish and Swedish. Of particular importance is a robust representation of the entities. In our case, we utilized unsupervised methods to generate such representations.MethodsThe significance of this work stands on its experimental layout. The experiments were carried out under the same conditions for both languages. Several classification approaches were explored: maximum probability, CRF, Perceptron and SVM. The classifiers were enhanced by means of ensembles of semantic spaces and ensembles of Brown trees. In order to mitigate sparsity of data, without a significant increase in the dimension of the decision space, we propose the use of clustered approaches of the hierarchical Brown clustering represented by trees and vector quantization for each semantic space.ResultsThe results showed that the semi-supervised approaches significantly improved standard supervised techniques for both languages. Moreover, clustering the semantic spaces contributed to the quality of the entity recognition while keeping the dimension of the feature-space two orders of magnitude lower than when directly using the semantic spaces.ConclusionsThe contributions of this study are: (a) a set of thorough experiments that enable comparisons regarding the influence of different types of features on different classifiers, exploring two languages other than English; and (b) the use of ensembles of clusters of Brown trees and semantic spaces on EHRs to tackle the problem of scarcity of available annotated data.     

CHI: A contemporaneous health index for degenerative disease monitoring using longitudinal measurements

In this paper, we develop a novel formulation for contemporaneous patient risk monitoring by exploiting the emerging data-rich environment in many healthcare applications, where an abundance of longitudinal data that reflect the degeneration of the health condition can be continuously collected. Our objective, and the developed formulation, is fundamentally different from many existing risk score models for different healthcare applications, which mostly focus on predicting the likelihood of a certain outcome at a pre-specified time. Rather, our formulation translates multivariate longitudinal measurements into a contemporaneous health index (CHI) that captures patient condition changes over the course of progression. Another significant feature of our formulation is that, CHI can be estimated with or without label information, different from other risk score models strictly based on supervised learning. To develop this formulation, we focus on the degenerative disease conditions, for which we could utilize the monotonic progression characteristic (either towards disease or recovery) to learn CHI. Such a domain knowledge leads us to a novel learning formulation, and on top of that, we further generalize this formulation with a capacity to incorporate label information if available. We further develop algorithms to mitigate the challenges associated with the nonsmooth convex optimization problem by first identifying its dual reformulation as a constrained smooth optimization problem, and then, using the block coordinate descent algorithm to iteratively solve the optimization with a derived efficient projection at each iteration. Extensive numerical studies are performed on both synthetic datasets and real-world applications on Alzheimer’s disease and Surgical Site Infection, which demonstrate the utility and efficacy of the proposed method on degenerative conditions that include a wide range of applications.     

Optimization of infobutton design and Implementation: A systematic review

ObjectiveInfobuttons are clinical decision tools embedded in the electronic health record that attempt to link clinical data with context sensitive knowledge resources. We systematically reviewed technical approaches that contribute to improved infobutton design, implementation and functionality.MethodsWe searched databases including MEDLINE, EMBASE, and the Cochrane Library database from inception to March 1, 2016 for studies describing the use of infobuttons. We selected full review comparative studies, usability studies, and qualitative studies examining infobutton design and implementation. We abstracted usability measures such as user satisfaction, impact, and efficiency, as well as prediction accuracy of infobutton content retrieval algorithms and infobutton adoption/interoperability.ResultsWe found 82 original research studies on infobuttons. Twelve studies met criteria for detailed abstraction. These studies investigated infobutton interoperability (1 study); tools to help tailor infobutton functionality (1 study); interventions to improve user experience (7 studies); and interventions to improve content retrieval by improving prediction of relevant knowledge resources and information needs (3 studies). In-depth interviews with implementers showed the Health Level Seven (HL7) Infobutton standard to be simple and easy to implement. A usability study demonstrated the feasibility of a tool to help medical librarians tailor infobutton functionality. User experience studies showed that access to resources with which users are familiar increased user satisfaction ratings; and that links to specific subsections of drug monographs increased information seeking efficiency. However, none of the user experience improvements led to increased usage uptake. Recommender systems based on machine learning algorithms outperformed hand-crafted rules in the prediction of relevant resources and clinicians’ information needs in a laboratory setting, but no studies were found using these techniques in clinical settings. Improved content indexing in one study led to improved content retrieval across three health care organizations.ConclusionBest practice technical approaches to ensure optimal infobutton functionality, design and implementation remain understudied. The HL7 Infobutton standard has supported wide adoption of infobutton functionality among clinical information systems and knowledge resources. Limited evidence supports infobutton enhancements such as links to specific subtopics, configuration of optimal resources for specific tasks and users, and improved indexing and content coverage. Further research is needed to investigate user experience improvements to increase infobutton use and effectiveness.     

Protection of Effective Component Group from Xiaoshuan Tongluo on Brain Injury after Chronic Hypoperfusion in Rats

Objective To investigate the protective effects of purified effective component group in extract from Xiaoshuan Tongluo(CGXT) formula on chronic brain ischemia in rats.Methods CGXT 75,150,and 300 mg/kg or vehicle were ig administered daily for four weeks to rats with bilateral common carotid arteries ligation(BCCAL) .From the day 24 to 28 after BCCAL,Morris water maze was performed to assess the learning and memory impairment of rats.Four weeks after BCCAL,brain gray and white matter damage were assessed.Results In Morris test,the mean escape latency of rats in the CGXT(150 and 300 mg/kg) groups was significantly shorter than that in the vehicle group.CGXT also attenuated the neuronal damage in hippocampus and cortex and reduced the pathological damage in the optic tract and corpus callosum.Conclusion CGXT could improve learning and memory impairment resulted from BCCAL in rats.These results provide the experimental basis for the clinical use of CGXT in stroke treatment and may help in investigation of multimodal therapy strategies in ischemic cerebrovascular diseases including stroke.     

Social exposure and emotion dysregulation: Main effects in relation to nonsuicidal self-injury

We examined the relation of interpersonal and media exposure to nonsuicidal self-injury (NSSI) among 340 university students in the southeastern United States (73.5% female, M age = 19.38 years, SD = 1.15). We also assessed interactions and main effects of each exposure and emotion dysregulation in relation to NSSI, testing the social learning hypothesis of NSSI. Most participants endorsed medium to high levels of exposure to NSSI via media sources. More than one-third of participants were somewhat or very familiar with someone who engaged in NSSI. Almost half reported occasional or frequent conversations about NSSI. Both exposure forms were significantly related to NSSI history. However, hurdle regression analyses revealed that interpersonal exposure and emotion dysregulation, but not media exposure, were significantly associated with NSSI history and frequency. We did not find evidence for an emotion dysregulation-by-interpersonal-exposure interaction. We discuss implications for theoretical models of NSSI, limitations, and future directions.     

Longitudinal analysis of discussion topics in an online breast cancer community using convolutional neural networks

Identifying topics of discussions in online health communities (OHC) is critical to various information extraction applications, but can be difficult because topics of OHC content are usually heterogeneous and domain-dependent. In this paper, we provide a multi-class schema, an annotated dataset, and supervised classifiers based on convolutional neural network (CNN) and other models for the task of classifying discussion topics. We apply the CNN classifier to the most popular breast cancer online community, and carry out cross-sectional and longitudinal analyses to show topic distributions and topic dynamics throughout members’ participation. Our experimental results suggest that CNN outperforms other classifiers in the task of topic classification and identify several patterns and trajectories. For example, although members discuss mainly disease-related topics, their interest may change through time and vary with their disease severities.     

Towards generalizable entity-centric clinical coreference resolution

ObjectiveThis work investigates the problem of clinical coreference resolution in a model that explicitly tracks entities, and aims to measure the performance of that model in both traditional in-domain train/test splits and cross-domain experiments that measure the generalizability of learned models.MethodsThe two methods we compare are a baseline mention-pair coreference system that operates over pairs of mentions with best-first conflict resolution and a mention-synchronous system that incrementally builds coreference chains. We develop new features that incorporate distributional semantics, discourse features, and entity attributes. We use two new coreference datasets with similar annotation guidelines – the THYME colon cancer dataset and the DeepPhe breast cancer dataset.ResultsThe mention-synchronous system performs similarly on in-domain data but performs much better on new data. Part of speech tag features prove superior in feature generalizability experiments over other word representations. Our methods show generalization improvement but there is still a performance gap when testing in new domains.DiscussionGeneralizability of clinical NLP systems is important and under-studied, so future work should attempt to perform cross-domain and cross-institution evaluations and explicitly develop features and training regimens that favor generalizability. A performance-optimized version of the mention-synchronous system will be included in the open source Apache cTAKES software.