Not so weak PICO: leveraging weak supervision for participants, interventions, and outcomes recognition for systematic review automation.

Objective: The aim of this study was to test the feasibility of PICO (participants, interventions, comparators, outcomes) entity extraction using weak supervision and natural language processing. Methodology: We re-purpose more than 127 medical and nonmedical ontologies and expert-generated rules to obtain multiple noisy labels for PICO entities in the evidence-based medicine (EBM)-PICO corpus. These noisy labels are aggregated using simple majority voting and generative modeling to get consensus labels. The resulting probabilistic labels are used as weak signals to train a weakly supervised (WS) discriminative model and observe performance changes. We explore mistakes in the EBM-PICO that could have led to inaccurate evaluation of previous automation methods. Results: In total, 4081 randomized clinical trials were weakly labeled to train the WS models and compared against full supervision. The models were separately trained for PICO entities and evaluated on the EBM-PICO test set. A WS approach combining ontologies and expert-generated rules outperformed full supervision for the participant entity by 1.71% macro-F1. Error analysis on the EBM-PICO subset revealed 18-23% erroneous token classifications. Discussion: Automatic PICO entity extraction accelerates the writing of clinical systematic reviews that commonly use PICO information to filter health evidence. However, PICO extends to more entities-PICOS (S-study type and design), PICOC (C-context), and PICOT (T-timeframe) for which labelled datasets are unavailable. In such cases, the ability to use weak supervision overcomes the expensive annotation bottleneck. Conclusions: We show the feasibility of WS PICO entity extraction using freely available ontologies and heuristics without manually annotated data. Weak supervision has encouraging performance compared to full supervision but requires careful design to outperform it.

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.

Machine Learning Assisted Citation Screening for Systematic Reviews.

Evidence-based practice is highly dependent upon up-to-date systematic reviews (SR) for decision making. However, conducting and updating systematic reviews, especially the citation screening for identification of relevant studies, requires much human work and is therefore expensive. Automating citation screening using machine learning (ML) based approaches can reduce cost and labor. Machine learning has been applied to automate citation screening but not for the SRs with very narrow research questions. This paper reports the results and observations for an ongoing research that aims to automate citation screening for SRs with narrow research questions using machine learning. The research also sheds light on the problem of class imbalance and class overlap on the performance of ML classifiers when applied to SRs with narrow research questions.

Automating Quality Control for Structured Standardized Radiology Reports Using Text Analysis.

Radiology reports describe the findings of a radiologist in an imaging examination, produced for another clinician in order to answer to a clinical indication. Sometimes, the report does not fully answer the question asked, despite guidelines for the radiologist. In this article, a system that controls the quality of reports automatically is described. It notably maps the free text onto MeSH terms and checks if the anatomy and disease terms match in the indication and conclusion of a report. The agreement between manual checks of experienced radiologists and the system is high with automatic checks requiring only a fraction of time. Being able to quality control all reports has the potential to improve report quality and thus limit misunderstandings, loosing time for requesting more information and possibly avoid medical mistakes.

NeuroTransDB: highly curated and structured transcriptomic metadata for neurodegenerative diseases.

Neurodegenerative diseases are chronic debilitating conditions, characterized by progressive loss of neurons that represent a significant health care burden as the global elderly population continues to grow. Over the past decade, high-throughput technologies such as the Affymetrix GeneChip microarrays have provided new perspectives into the pathomechanisms underlying neurodegeneration. Public transcriptomic data repositories, namely Gene Expression Omnibus and curated ArrayExpress, enable researchers to conduct integrative meta-analysis; increasing the power to detect differentially regulated genes in disease and explore patterns of gene dysregulation across biologically related studies. The reliability of retrospective, large-scale integrative analyses depends on an appropriate combination of related datasets, in turn requiring detailed meta-annotations capturing the experimental setup. In most cases, we observe huge variation in compliance to defined standards for submitted metadata in public databases. Much of the information to complete, or refine meta-annotations are distributed in the associated publications. For example, tissue preparation or comorbidity information is frequently described in an article's supplementary tables. Several value-added databases have employed additional manual efforts to overcome this limitation. However, none of these databases explicate annotations that distinguish human and animal models in neurodegeneration context. Therefore, adopting a more specific disease focus, in combination with dedicated disease ontologies, will better empower the selection of comparable studies with refined annotations to address the research question at hand. In this article, we describe the detailed development of NeuroTransDB, a manually curated database containing metadata annotations for neurodegenerative studies. The database contains more than 20 dimensions of metadata annotations within 31 mouse, 5 rat and 45 human studies, defined in collaboration with domain disease experts. We elucidate the step-by-step guidelines used to critically prioritize studies from public archives and their metadata curation and discuss the key challenges encountered. Curated metadata for Alzheimer's disease gene expression studies are available for download. Database URL: www.scai.fraunhofer.de/NeuroTransDB.html.