Semantic annotation of consumer health questions.

BACKGROUND: Consumers increasingly use online resources for their health information needs. While current search engines can address these needs to some extent, they generally do not take into account that most health information needs are complex and can only fully be expressed in natural language. Consumer health question answering (QA) systems aim to fill this gap. A major challenge in developing consumer health QA systems is extracting relevant semantic content from the natural language questions (question understanding). To develop effective question understanding tools, question corpora semantically annotated for relevant question elements are needed. In this paper, we present a two-part consumer health question corpus annotated with several semantic categories: named entities, question triggers/types, question frames, and question topic. The first part (CHQA-email) consists of relatively long email requests received by the U.S. National Library of Medicine (NLM) customer service, while the second part (CHQA-web) consists of shorter questions posed to MedlinePlus search engine as queries. Each question has been annotated by two annotators. The annotation methodology is largely the same between the two parts of the corpus; however, we also explain and justify the differences between them. Additionally, we provide information about corpus characteristics, inter-annotator agreement, and our attempts to measure annotation confidence in the absence of adjudication of annotations. RESULTS: The resulting corpus consists of 2614 questions (CHQA-email: 1740, CHQA-web: 874). Problems are the most frequent named entities, while treatment and general information questions are the most common question types. Inter-annotator agreement was generally modest: question types and topics yielded highest agreement, while the agreement for more complex frame annotations was lower. Agreement in CHQA-web was consistently higher than that in CHQA-email. Pairwise inter-annotator agreement proved most useful in estimating annotation confidence. CONCLUSIONS: To our knowledge, our corpus is the first focusing on annotation of uncurated consumer health questions. It is currently used to develop machine learning-based methods for question understanding. We make the corpus publicly available to stimulate further research on consumer health QA.

Text mining for pharmacovigilance: Using machine learning for drug name recognition and drug-drug interaction extraction and classification.

Pharmacovigilance (PV) is defined by the World Health Organization as the science and activities related to the detection, assessment, understanding and prevention of adverse effects or any other drug-related problem. An essential aspect in PV is to acquire knowledge about Drug-Drug Interactions (DDIs). The shared tasks on DDI-Extraction organized in 2011 and 2013 have pointed out the importance of this issue and provided benchmarks for: Drug Name Recognition, DDI extraction and DDI classification. In this paper, we present our text mining systems for these tasks and evaluate their results on the DDI-Extraction benchmarks. Our systems rely on machine learning techniques using both feature-based and kernel-based methods. The obtained results for drug name recognition are encouraging. For DDI-Extraction, our hybrid system combining a feature-based method and a kernel-based method was ranked second in the DDI-Extraction-2011 challenge, and our two-step system for DDI detection and classification was ranked first in the DDI-Extraction-2013 task at SemEval. We discuss our methods and results and give pointers to future work.

Consumer health information and question answering: helping consumers find answers to their health-related information needs.

OBJECTIVE: Consumers increasingly turn to the internet in search of health-related information; and they want their questions answered with short and precise passages, rather than needing to analyze lists of relevant documents returned by search engines and reading each document to find an answer. We aim to answer consumer health questions with information from reliable sources. MATERIALS AND METHODS: We combine knowledge-based, traditional machine and deep learning approaches to understand consumers' questions and select the best answers from consumer-oriented sources. We evaluate the end-to-end system and its components on simple questions generated in a pilot development of MedlinePlus Alexa skill, as well as the short and long real-life questions submitted to the National Library of Medicine by consumers. RESULTS: Our system achieves 78.7% mean average precision and 87.9% mean reciprocal rank on simple Alexa questions, and 44.5% mean average precision and 51.6% mean reciprocal rank on real-life questions submitted by National Library of Medicine consumers. DISCUSSION: The ensemble of deep learning, domain knowledge, and traditional approaches recognizes question type and focus well in the simple questions, but it leaves room for improvement on the real-life consumers' questions. Information retrieval approaches alone are sufficient for finding answers to simple Alexa questions. Answering real-life questions, however, benefits from a combination of information retrieval and inference approaches. CONCLUSION: A pilot practical implementation of research needed to help consumers find reliable answers to their health-related questions demonstrates that for most questions the reliable answers exist and can be found automatically with acceptable accuracy.

Bridging the Gap Between Consumers' Medication Questions and Trusted Answers.

This paper addresses the task of answering consumer health questions about medications. To better understand the challenge and needs in terms of methods and resources, we first introduce a gold standard corpus for Medication Question Answering created using real consumer questions. The gold standard (https://github.com/abachaa/Medication_QA_MedInfo2019) consists of six hundred and seventy-four question-answer pairs with annotations of the question focus and type and the answer source. We first present the manual annotation and answering process. In the second part of this paper, we test the performance of recurrent and convolutional neural networks in question type identification and focus recognition. Finally, we discuss the research insights from both the dataset creation process and our experiments. This study provides new resources and experiments on answering consumers' medication questions and discusses the limitations and directions for future research efforts.

Combining Open-domain and Biomedical Knowledge for Topic Recognition in Consumer Health Questions.

Determining the main topics in consumer health questions is a crucial step in their processing as it allows narrowing the search space to a specific semantic context. In this paper we propose a topic recognition approach based on biomedical and open-domain knowledge bases. In the first step of our method, we recognize named entities in consumer health questions using an unsupervised method that relies on a biomedical knowledge base, UMLS, and an open-domain knowledge base, DBpedia. In the next step, we cast topic recognition as a binary classification problem of deciding whether a named entity is the question topic or not. We evaluated our approach on a dataset from the National Library of Medicine (NLM), introduced in this paper, and another from the Genetic and Rare Disease Information Center (GARD). The combination of knowledge bases outperformed the results obtained by individual knowledge bases by up to 16.5% F1 and achieved state-of-the-art performance. Our results demonstrate that combining open-domain knowledge bases with biomedical knowledge bases can lead to a substantial improvement in understanding user-generated health content.

Towards natural language question generation for the validation of ontologies and mappings.

BACKGROUND: The increasing number of open-access ontologies and their key role in several applications such as decision-support systems highlight the importance of their validation. Human expertise is crucial for the validation of ontologies from a domain point-of-view. However, the growing number of ontologies and their fast evolution over time make manual validation challenging. METHODS: We propose a novel semi-automatic approach based on the generation of natural language (NL) questions to support the validation of ontologies and their evolution. The proposed approach includes the automatic generation, factorization and ordering of NL questions from medical ontologies. The final validation and correction is performed by submitting these questions to domain experts and automatically analyzing their feedback. We also propose a second approach for the validation of mappings impacted by ontology changes. The method exploits the context of the changes to propose correction alternatives presented as Multiple Choice Questions. RESULTS: This research provides a question optimization strategy to maximize the validation of ontology entities with a reduced number of questions. We evaluate our approach for the validation of three medical ontologies. We also evaluate the feasibility and efficiency of our mappings validation approach in the context of ontology evolution. These experiments are performed with different versions of SNOMED-CT and ICD9. CONCLUSIONS: The obtained experimental results suggest the feasibility and adequacy of our approach to support the validation of interconnected and evolving ontologies. Results also suggest that taking into account RDFS and OWL entailment helps reducing the number of questions and validation time. The application of our approach to validate mapping evolution also shows the difficulty of adapting mapping evolution over time and highlights the importance of semi-automatic validation.

Mathematical methods to analysis of topology, functional variability and evolution of metabolic systems based on different decomposition concepts.

Complexity of metabolic systems can be undertaken at different scales (metabolites, metabolic pathways, metabolic network map, biological population) and under different aspects (structural, functional, evolutive). To analyse such a complexity, metabolic systems need to be decomposed into different components according to different concepts. Four concepts are presented here consisting in considering metabolic systems as sets of metabolites, chemical reactions, metabolic pathways or successive processes. From a metabolomic dataset, such decompositions are performed using different mathematical methods including correlation, stiochiometric, ordination, classification, combinatorial and kinetic analyses. Correlation analysis detects and quantifies affinities/oppositions between metabolites. Stoichiometric analysis aims to identify the organisation of a metabolic network into different metabolic pathways on the hand, and to quantify/optimize the metabolic flux distribution through the different chemical reactions of the system. Ordination and classification analyses help to identify different metabolic trends and their associated metabolites in order to highlight chemical polymorphism representing different variability poles of the metabolic system. Then, metabolic processes/correlations responsible for such a polymorphism can be extracted in silico by combining metabolic profiles representative of different metabolic trends according to a weighting bootstrap approach. Finally evolution of metabolic processes in time can be analysed by different kinetic/dynamic modelling approaches.