Understanding Clinical Trial Reports: Extracting Medical Entities and Their Relations.

The best evidence concerning comparative treatment effectiveness comes from clinical trials, the results of which are reported in unstructured articles. Medical experts must manually extract information from articles to inform decision-making, which is time-consuming and expensive. Here we consider the end-to-end task of both (a) extracting treatments and outcomes from full-text articles describing clinical trials (entity identification) and, (b) inferring the reported results for the former with respect to the latter (relation extraction). We introduce new data for this task, and evaluate models that have recently achieved state-of-the-art results on similar tasks in Natural Language Processing. We then propose a new method motivated by how trial results are typically presented that outperforms these purely data-driven baselines. Finally, we run a fielded evaluation of the model with a non-profit seeking to identify existing drugs that might be re-purposed for cancer, showing the potential utility of end-to-end evidence extraction systems.

State of the evidence: a survey of global disparities in clinical trials.

INTRODUCTION: Ideally, health conditions causing the greatest global disease burden should attract increased research attention. We conducted a comprehensive global study investigating the number of randomised controlled trials (RCTs) published on different health conditions, and how this compares with the global disease burden that they impose. METHODS: We use machine learning to monitor PubMed daily, and find and analyse RCT reports. We assessed RCTs investigating the leading causes of morbidity and mortality from the Global Burden of Disease study. Using regression models, we compared numbers of actual RCTs in different health conditions to numbers predicted from their global disease burden (disability-adjusted life years (DALYs)). We investigated whether RCT numbers differed for conditions disproportionately affecting countries with lower socioeconomic development. RESULTS: We estimate 463 000 articles describing RCTs (95% prediction interval 439 000 to 485 000) were published from 1990 to July 2020. RCTs recruited a median of 72 participants (IQR 32-195). 82% of RCTs were conducted by researchers in the top fifth of countries by socio-economic development. As DALYs increased for a particular health condition by 10%, the number of RCTs in the same year increased by 5% (3.2%-6.9%), but the association was weak (adjusted R2=0.13). Conditions disproportionately affecting countries with lower socioeconomic development, including respiratory infections and tuberculosis (7000 RCTs below predicted) and enteric infections (9700 RCTs below predicted), appear relatively under-researched for their disease burden. Each 10% shift in DALYs towards countries with low and middle socioeconomic development was associated with a 4% reduction in RCTs (3.7%-4.9%). These disparities have not changed substantially over time. CONCLUSION: Research priorities are not well optimised to reduce the global burden of disease. Most RCTs are produced by highly developed countries, and the health needs of these countries have been, on average, favoured.

Trialstreamer: Mapping and Browsing Medical Evidence in Real-Time.

We introduce Trialstreamer, a living database of clinical trial reports. Here we mainly describe the evidence extraction component; this extracts from biomedical abstracts key pieces of information that clinicians need when appraising the literature, and also the relations between these. Specifically, the system extracts descriptions of trial participants, the treatments compared in each arm (the interventions), and which outcomes were measured. The system then attempts to infer which interventions were reported to work best by determining their relationship with identified trial outcome measures. In addition to summarizing individual trials, these extracted data elements allow automatic synthesis of results across many trials on the same topic. We apply the system at scale to all reports of randomized controlled trials indexed in MEDLINE, powering the automatic generation of evidence maps, which provide a global view of the efficacy of different interventions combining data from all relevant clinical trials on a topic. We make all code and models freely available alongside a demonstration of the web interface.

Trialstreamer: A living, automatically updated database of clinical trial reports.

OBJECTIVE: Randomized controlled trials (RCTs) are the gold standard method for evaluating whether a treatment works in health care but can be difficult to find and make use of. We describe the development and evaluation of a system to automatically find and categorize all new RCT reports. MATERIALS AND METHODS: Trialstreamer continuously monitors PubMed and the World Health Organization International Clinical Trials Registry Platform, looking for new RCTs in humans using a validated classifier. We combine machine learning and rule-based methods to extract information from the RCT abstracts, including free-text descriptions of trial PICO (populations, interventions/comparators, and outcomes) elements and map these snippets to normalized MeSH (Medical Subject Headings) vocabulary terms. We additionally identify sample sizes, predict the risk of bias, and extract text conveying key findings. We store all extracted data in a database, which we make freely available for download, and via a search portal, which allows users to enter structured clinical queries. Results are ranked automatically to prioritize larger and higher-quality studies. RESULTS: As of early June 2020, we have indexed 673 191 publications of RCTs, of which 22 363 were published in the first 5 months of 2020 (142 per day). We additionally include 304 111 trial registrations from the International Clinical Trials Registry Platform. The median trial sample size was 66. CONCLUSIONS: We present an automated system for finding and categorizing RCTs. This yields a novel resource: a database of structured information automatically extracted for all published RCTs in humans. We make daily updates of this database available on our website (https://trialstreamer.robotreviewer.net).

A Corpus with Multi-Level Annotations of Patients, Interventions and Outcomes to Support Language Processing for Medical Literature.

We present a corpus of 5,000 richly annotated abstracts of medical articles describing clinical randomized controlled trials. Annotations include demarcations of text spans that describe the Patient population enrolled, the Interventions studied and to what they were Compared, and the Outcomes measured (the 'PICO' elements). These spans are further annotated at a more granular level, e.g., individual interventions within them are marked and mapped onto a structured medical vocabulary. We acquired annotations from a diverse set of workers with varying levels of expertise and cost. We describe our data collection process and the corpus itself in detail. We then outline a set of challenging NLP tasks that would aid searching of the medical literature and the practice of evidence-based medicine.

SKOPE-IT (Shareable Knowledge Objects as Portable Intelligent Tutors): overlaying natural language tutoring on an adaptive learning system for mathematics.

BACKGROUND: This study investigated learning outcomes and user perceptions from interactions with a hybrid intelligent tutoring system created by combining the AutoTutor conversational tutoring system with the Assessment and Learning in Knowledge Spaces (ALEKS) adaptive learning system for mathematics. This hybrid intelligent tutoring system (ITS) uses a service-oriented architecture to combine these two web-based systems. Self-explanation tutoring dialogs were used to talk students through step-by-step worked examples to algebra problems. These worked examples presented an isomorphic problem to the preceding algebra problem that the student could not solve in the adaptive learning system. RESULTS: Due to crossover issues between conditions, experimental versus control condition assignment did not show significant differences in learning gains. However, strong dose-dependent learning gains were observed that could not be otherwise explained by either initial mastery or time-on-task. User perceptions of the dialog-based tutoring were mixed, and survey results indicate that this may be due to the pacing of dialog-based tutoring using voice, students judging the agents based on their own performance (i.e., the quality of their answers to agent questions), and the students' expectations about mathematics pedagogy (i.e., expecting to solving problems rather than talking about concepts). Across all users, learning was most strongly influenced by time spent studying, which correlated with students' self-reported tendencies toward effort avoidance, effective study habits, and beliefs about their ability to improve in mathematics with effort. CONCLUSIONS: Integrating multiple adaptive tutoring systems with complementary strengths shows some potential to improve learning. However, managing learner expectations during transitions between systems remains an open research area. Finally, while personalized adaptation can improve learning efficiency, effort and time-on-task for learning remains a dominant factor that must be considered by interventions.

ElectronixTutor: an intelligent tutoring system with multiple learning resources for electronics.

BACKGROUND: The Office of Naval Research (ONR) organized a STEM Challenge initiative to explore how intelligent tutoring systems (ITSs) can be developed in a reasonable amount of time to help students learn STEM topics. This competitive initiative sponsored four teams that separately developed systems that covered topics in mathematics, electronics, and dynamical systems. After the teams shared their progress at the conclusion of an 18-month period, the ONR decided to fund a joint applied project in the Navy that integrated those systems on the subject matter of electronic circuits. The University of Memphis took the lead in integrating these systems in an intelligent tutoring system called ElectronixTutor. This article describes the architecture of ElectronixTutor, the learning resources that feed into it, and the empirical findings that support the effectiveness of its constituent ITS learning resources. RESULTS: A fully integrated ElectronixTutor was developed that included several intelligent learning resources (AutoTutor, Dragoon, LearnForm, ASSISTments, BEETLE-II) as well as texts and videos. The architecture includes a student model that has (a) a common set of knowledge components on electronic circuits to which individual learning resources contribute and (b) a record of student performance on the knowledge components as well as a set of cognitive and non-cognitive attributes. There is a recommender system that uses the student model to guide the student on a small set of sensible next steps in their training. The individual components of ElectronixTutor have shown learning gains in previous decades of research. CONCLUSIONS: The ElectronixTutor system successfully combines multiple empirically based components into one system to teach a STEM topic (electronics) to students. A prototype of this intelligent tutoring system has been developed and is currently being tested. ElectronixTutor is unique in its assembling a group of well-tested intelligent tutoring systems into a single integrated learning environment.