Methods to assess youth engagement in a text messaging supplement to an effective teen pregnancy program

Youth are prolific users of cell phone minutes and text messaging. Numerous programs using short message service text messaging (SMS) have been employed to help improve health behaviors and health outcomes. However, we lack information on whether and what type of interaction or engagement with SMS program content is required to realize any benefit.We explored youth engagement with an automated SMS program designed to supplement a 25-session youth development program with demonstrated efficacy for reductions in teen pregnancy. Using two years of program data, we report on youth participation in design of message content and response frequency to messages among youth enrolled in the intervention arm of a randomized controlled trial (RCT) as one indicator of engagement.There were 221 youth between the ages of 14–18 enrolled over two years in the intervention arm of the RCT. Just over half (51%) were female; 56% were Hispanic; and 27% African American. Youth were sent 40,006 messages of which 16,501 were considered bi-directional where youth were asked to text a response. Four-fifths (82%) responded at least once to a text. We found variations in response frequency by gender, age, and ethnicity. The most popular types of messages youth responded to include questions and quizzes. The first two months of the program in each year had the highest response frequency.An important next step is to assess whether higher response to SMS results in greater efficacy. This future work can facilitate greater attention to message design and content to ensure messages are engaging for the intended audience.     

Eye movements during processing of text requiring bridging inferences in adolescents with higher functioning autism spectrum disorders: A preliminary investigation

In this study, 18 children with higher functioning autism (HFASD) and 16 controls read short, two-sentence passages that necessitated a bridging inference for comprehension and responded to a general knowledge question that was either primed or unprimed by the inference. We measured mean reading times and correct responses to knowledge questions, as well as recorded the eye-movements of the participants while reading. Analysis of reaction times to general knowledge questions were faster for questions in relevant context as compared to those in irrelevant contexts, suggesting that children with HFASD were able to build the bridging inferences necessary for comprehension. Eye-movement data revealed that children with HFASD spent more time fixating on text, made more fixations overall, and made more regressions (i.e., moving backward within the text) while reading than did controls. Implications of these findings and suggestions for future research are offered.     

PKDE4J: Entity and relation extraction for public knowledge discovery

Due to an enormous number of scientific publications that cannot be handled manually, there is a rising interest in text-mining techniques for automated information extraction, especially in the biomedical field. Such techniques provide effective means of information search, knowledge discovery, and hypothesis generation. Most previous studies have primarily focused on the design and performance improvement of either named entity recognition or relation extraction. In this paper, we present PKDE4J, a comprehensive text-mining system that integrates dictionary-based entity extraction and rule-based relation extraction in a highly flexible and extensible framework. Starting with the Stanford CoreNLP, we developed the system to cope with multiple types of entities and relations. The system also has fairly good performance in terms of accuracy as well as the ability to configure text-processing components. We demonstrate its competitive performance by evaluating it on many corpora and found that it surpasses existing systems with average F-measures of 85% for entity extraction and 81% for relation extraction.     

Comparison of text-messaging to voice telephone interviews for active surveillance of adverse events following immunisation

ObjectivesIn 2013, the Follow-up and Active Surveillance of Trivalent Influenza Vaccine in Mums (FASTMum) program began using short message service (SMS) to collect adverse event information in pregnant women who recently received trivalent influenza vaccine (TIV). This study was designed to compare data collected via SMS and telephone for the purposes of monitoring vaccine safety.MethodsA number of 344 women who received TIV were randomly assigned to a telephone interview group. They were telephoned seven days post-vaccination and administered a standard survey soliciting any adverse events following immunisation (AEFI) they experienced. They were matched by brand of vaccine, age group, and residence to 344 women who were sent a SMS seven days post-vaccination. The SMS solicited similar information. AEFI reported by SMS and telephone interview were compared by calculating risk ratios.ResultsResponse rate was higher to SMS compared to telephone interview (90.1% vs. 63.9%). Women who were surveyed by SMS were significantly less likely to report an AEFI compared to women who were surveyed by telephone (RR: 0.41; 95% CI: 0.29–0.59). The greatest discrepancies between SMS and telephone interview were for self-reported injection site reactions (3.1% vs. 16.8%) and unsolicited (or “other”) events (11.4% vs. 4.1%). Data collected by SMS was significantly timelier.ConclusionsData collection by SMS results in significantly improved response rates and timeliness of vaccine safety data. Systems which incorporate SMS could be used to more rapidly detect safety signals and promote more rapid public health response to vaccine quality issues.     

基于国际新闻的疾病趋势预测算法

文本分类作为自然语言处理领域的核心内容,已经成为文本处理的重点研究问题。该文主要针对社会上出现的大规模常见疾病进行预测。该文通过获取全球各大新闻媒体报道的新闻文本,分别统计新闻文本中出现次数排名前十的疾病,分析原始数据分布的特征。该文将基于 CNN 和 LSTM 网络的文本模型与基于 LSTM 网络的疾病趋势模型进行融合,综合分析文本中新闻文本的文本信息和疾病的时间序列,并使用了一种特殊的疾病选择策略。实验结果表明,该策略在 7 种不同的新闻数据集上获得了 70%以上的准确度。该文提出的融合策略和疾病选择策略对疾病的趋势预测具有一定的意义,有助于提高疾病趋势预测的准确性。     

新农合一卡通直补系统与HIS接口的设计与实现

针对SemRep不利于对所提取关系进行统计分析的问题,开发SemRep处理结果统计挖掘系统。介绍该系统各模块的设计情况,包括信息抽取、统计、查询、推导等。指出其功能能够达到最初设计目的。     

文本挖掘技术在药物研究中的应用

介绍生物医学文献中文本挖掘的过程,重点分析论述文本挖掘在药物研究中的应用,主要包括在药物命名识别、药物靶标发现、药物疗效评价和中医用药规律中的应用,最后指出文本挖掘在生物医学知识发现中需要解决的问题,并展望文本挖掘在药物研究领域的发展前景.     

Eventual situations for timeline extraction from clinical reports

Objective

To identify the temporal relations between clinical events and temporal expressions in clinical reports, as defined in the i2b2/VA 2012 challenge.

Design

To detect clinical events, we used rules and Conditional Random Fields. We built Random Forest models to identify event modality and polarity. To identify temporal expressions we built on the HeidelTime system. To detect temporal relations, we systematically studied their breakdown into distinct situations; we designed an oracle method to determine the most prominent situations and the most suitable associated classifiers, and combined their results.

Results

We achieved F-measures of 0.8307 for event identification, based on rules, and 0.8385 for temporal expression identification. In the temporal relation task, we identified nine main situations in three groups, experimentally confirming shared intuitions: within-sentence relations, section-related time, and across-sentence relations. Logistic regression and Naïve Bayes performed best on the first and third groups, and decision trees on the second. We reached a 0.6231 global F-measure, improving by 7.5 points our official submission.

Conclusions

Carefully hand-crafted rules obtained good results for the detection of events and temporal expressions, while a combination of classifiers improved temporal link prediction. The characterization of the oracle recall of situations allowed us to point at directions where further work would be most useful for temporal relation detection: within-sentence relations and linking History of Present Illness events to the admission date. We suggest that the systematic situation breakdown proposed in this paper could also help improve other systems addressing this task.     

黄精化合物对固齿的作用

以Swanson提出的开放式知识发现方法为基础,结合情报学的类比推理方法挖掘黄精化合物与补肾固齿关系的题录,找出黄精具有促进牙周组织细胞增殖或抑制阻止增殖的关联因子,明确黄精固齿的途径机理,完善其促牙周组织增殖、抗牙龈萎缩而固齿的假设模型,推导黄精多糖等促进牙周膜细胞的成骨作用。