Clinical Trials Reports (CTRs) contain highly valuable health information from which Natural Language Inference (NLI) techniques determine if a given hypothesis can be inferred from a given premise. CTRs are abundant with domain terminology with particular terms that are difficult to understand without prior knowledge. Thus, we proposed to use domain ontologies as a source of external knowledge that could help with the inference process in theSemEval-2023 Task 7: Multi-evidence Natural Language Inference for Clinical Trial Data (NLI4CT). This document describes our participation in subtask 1: Textual Entailment, where Ontologies, NLP techniques, such as tokenization and named-entity recognition, and rule-based approaches are all combined in our approach. We were able to show that inputting annotations from domain ontologies improved the baseline systems.

The paper describes the participation of the Lasige-BioTM team at sub-tracks A and B of ProfNER, which was based on: i) a BiLSTM-CRF model that leverages contextual and classical word embeddings to recognize and classify the mentions, and ii) on a rule-based module to classify tweets. In the Evaluation phase, our model achieved a F1-score of 0.917 (0,031 more than the median) in sub-track A and a F1-score of 0.727 (0,034 less than the median) in sub-track B.

With the increasing number of publications about COVID-19, it is a challenge to extract personalized knowledge suitable for each researcher. This work aims to build a new semantic-based pipeline for recommending biomedical entities to scientific researchers. To this end, we developed a pipeline that creates an implicit feedback matrix based on Named Entity Recognition (NER) on a corpus of documents, using multidisciplinary ontologies for recognizing and linking the entities. Our hypothesis is that by using ontologies from different fields in the NER phase, we can improve the results for state-of-the-art collaborative-filtering recommender systems applied to the dataset created. The tests performed using the COVID-19 Open Research Dataset (CORD-19) dataset show that when using four ontologies, the results for precision@k, for example, reach the 80%, whereas when using only one ontology, the results for precision@k drops to 20%, for the same users. Furthermore, the use of multi-fields entities may help in the discovery of new items, even if the researchers do not have items from that field in their set of preferences.

Human phenotype-gene relations are fundamental to fully understand the origin of some phenotypic abnormalities and their associated diseases. Biomedical literature is the most comprehensive source of these relations, however, we need Relation Extraction tools to automatically recognize them. Most of these tools require an annotated corpus and to the best of our knowledge, there is no corpus available annotated with human phenotype-gene relations. This paper presents the Phenotype-Gene Relations (PGR) corpus, a silver standard corpus of human phenotype and gene annotations and their relations. The corpus consists of 1712 abstracts, 5676 human phenotype annotations, 13835 gene annotations, and 4283 relations. We generated this corpus using Named-Entity Recognition tools, whose results were partially evaluated by eight curators, obtaining a precision of 87.01%. By using the corpus we were able to obtain promising results with two state-of-the-art deep learning tools, namely 78.05% of precision. The PGR corpus was made publicly available to the research community.

Biomedical Question Answering (QA) aims at providing automated answers to user questions, regarding a variety of biomedical topics. For example, these questions may ask for related to diseases, drugs, symptoms, or medical procedures. Automated biomedical QA systems could improve the retrieval of information necessary to answer these questions. The MEDIQA challenge consisted of three tasks concerning various aspects of biomedical QA. This challenge aimed at advancing approaches to Natural Language Inference (NLI) and Recognizing Question Entailment (RQE), which would then result in enhanced approaches to biomedical QA. Our approach explored a common Transformer-based architecture that could be applied to each task. This approach shared the same pre-trained weights, but which were then fine-tuned for each task using the provided training data. Furthermore, we augmented the training data with external datasets and enriched the question and answer texts using MER, a named entity recognition tool. Our approach obtained high levels of accuracy, in particular on the NLI task, which classified pairs of text according to their relation. For the QA task, we obtained higher Spearman’s rank correlation values using the entities recognized by MER.

This paper describes our system, dubbed MoRS (Modular Ranking System), pronounced ‘Morse’, which participated in Task 3 of SemEval-2017. We used MoRS to perform the Community Question Answering Task 3, which consisted on reordering a set of comments according to their usefulness in answering the question in the thread. This was made for a large collection of questions created by a user community. As for this challenge we wanted to go back to simple, easy-to-use, and somewhat forgotten technologies that we think, in the hands of non-expert people, could be reused in their own data sets. Some of our techniques included the annotation of text, the retrieval of meta-data for each comment, POS tagging and Named Entity Recognition, among others. These gave place to syntactical analysis and semantic measurements. Finally we show and discuss our results and the context of our approach, which is part of a more comprehensive system in development, named MoQA.

This paper presents our approach to participate in the SemEval 2017 Task 12: Clinical TempEval challenge, specifically in the event and time expressions span and attribute identification subtasks (ES, EA, TS, TA). Our approach consisted in training Conditional Random Fields (CRF) classifiers using the provided annotations, and in creating manually curated rules to classify the attributes of each event and time expression. We used a set of common features for the event and time CRF classifiers, and a set of features specific to each type of entity, based on domain knowledge. Training only on the source domain data, our best F-scores were 0.683 and 0.485 for event and time span identification subtasks. When adding target domain annotations to the training data, the best F-scores obtained were 0.729 and 0.554, for the same subtasks. We obtained the second highest F-score of the challenge on the event polarity subtask (0.708). The source code of our system, Clinical Timeline Annotation (CiTA), is available at https://github.com/lasigeBioTM/CiTA.