In this paper, we present an overview of the MedVidQA 2022 shared task, collocated with the 21st BioNLP workshop at ACL 2022. The shared task addressed two of the challenges faced by medical video question answering: (I) a video classification task that explores new approaches to medical video understanding (labeling), and (ii) a visual answer localization task. Visual answer localization refers to the identification of the relevant temporal segments (start and end timestamps) in the video where the answer to the medical question is being shown or illustrated. A total of thirteen teams participated in the shared task challenges, with eleven system descriptions submitted to the workshop. The descriptions present monomodal and multi-modal approaches developed for medical video classification and visual answer localization. This paper describes the tasks, the datasets, evaluation metrics, and baseline systems for both tasks. Finally, the paper summarizes the techniques and results of the evaluation of the various approaches explored by the participating teams.

The task of verifying the truthfulness of claims in textual documents, or fact-checking, has received significant attention in recent years. Many existing evidence-based factchecking datasets contain synthetic claims and the models trained on these data might not be able to verify real-world claims. Particularly few studies addressed evidence-based fact-checking of health-related claims that require medical expertise or evidence from the scientific literature. In this paper, we introduce HEALTHVER, a new dataset for evidence-based fact-checking of health-related claims that allows to study the validity of real-world claims by evaluating their truthfulness against scientific articles. Using a three-step data creation method, we first retrieved real-world claims from snippets returned by a search engine for questions about COVID-19. Then we automatically retrieved and re-ranked relevant scientific papers using a T5 relevance-based model. Finally, the relations between each evidence statement and the associated claim were manually annotated as SUPPORT, REFUTE and NEUTRAL. To validate the created dataset of 14,330 evidence-claim pairs, we developed baseline models based on pretrained language models. Our experiments showed that training deep learning models on real-world medical claims greatly improves performance compared to models trained on synthetic and open-domain claims. Our results and manual analysis suggest that HEALTHVER provides a realistic and challenging dataset for future efforts on evidence-based fact-checking of health-related claims. The dataset, source code, and a leaderboard are available at https://github.com/sarrouti/healthver.

The MEDIQA 2021 shared tasks at the BioNLP 2021 workshop addressed three tasks on summarization for medical text: (i) a question summarization task aimed at exploring new approaches to understanding complex real-world consumer health queries, (ii) a multi-answer summarization task that targeted aggregation of multiple relevant answers to a biomedical question into one concise and relevant answer, and (iii) a radiology report summarization task addressing the development of clinically relevant impressions from radiology report findings. Thirty-five teams participated in these shared tasks with sixteen working notes submitted (fifteen accepted) describing a wide variety of models developed and tested on the shared and external datasets. In this paper, we describe the tasks, the datasets, the models and techniques developed by various teams, the results of the evaluation, and a study of correlations among various summarization evaluation measures. We hope that these shared tasks will bring new research and insights in biomedical text summarization and evaluation.

The growth of online consumer health questions has led to the necessity for reliable and accurate question answering systems. A recent study showed that manual summarization of consumer health questions brings significant improvement in retrieving relevant answers. However, the automatic summarization of long questions is a challenging task due to the lack of training data and the complexity of the related subtasks, such as the question focus and type recognition. In this paper, we introduce a reinforcement learning-based framework for abstractive question summarization. We propose two novel rewards obtained from the downstream tasks of (i) question-type identification and (ii) question-focus recognition to regularize the question generation model. These rewards ensure the generation of semantically valid questions and encourage the inclusion of key medical entities/foci in the question summary. We evaluated our proposed method on two benchmark datasets and achieved higher performance over state-of-the-art models. The manual evaluation of the summaries reveals that the generated questions are more diverse and have fewer factual inconsistencies than the baseline summaries. The source code is available here: https://github.com/shwetanlp/CHQ-Summ.

Deep neural networks have demonstrated high performance on many natural language processing (NLP) tasks that can be answered directly from text, and have struggled to solve NLP tasks requiring external (e.g., world) knowledge. In this paper, we present OSCR (Ontology-based Semantic Composition Regularization), a method for injecting task-agnostic knowledge from an Ontology or knowledge graph into a neural network during pre-training. We evaluated the performance of BERT pre-trained on Wikipedia with and without OSCR by measuring the performance when fine-tuning on two question answering tasks involving world knowledge and causal reasoning and one requiring domain (healthcare) knowledge and obtained 33.3%, 18.6%, and 4% improved accuracy compared to pre-training BERT without OSCR.

Recent work has shown that pre-trained Transformers obtain remarkable performance on many natural language processing tasks including automatic summarization. However, most work has focused on (relatively) data-rich single-document summarization settings. In this paper, we explore highly-abstractive multi-document summarization where the summary is explicitly conditioned on a user-given topic statement or question. We compare the summarization quality produced by three state-of-the-art transformer-based models: BART, T5, and PEGASUS. We report the performance on four challenging summarization datasets: three from the general domain and one from consumer health in both zero-shot and few-shot learning settings. While prior work has shown significant differences in performance for these models on standard summarization tasks, our results indicate that with as few as 10 labeled examples there is no statistically significant difference in summary quality, suggesting the need for more abstractive benchmark collections when determining state-of-the-art.

Efficient document summarization requires evaluation measures that can not only rank a set of systems based on an average score, but also highlight which individual summary is better than another. However, despite the very active research on summarization approaches, few works have proposed new evaluation measures in the recent years. The standard measures relied upon for the development of summarization systems are most often ROUGE and BLEU which, despite being efficient in overall system ranking, remain lexical in nature and have a limited potential when it comes to training neural networks. In this paper, we present a new hybrid evaluation measure for summarization, called HOLMS, that combines both language models pre-trained on large corpora and lexical similarity measures. Through several experiments, we show that HOLMS outperforms ROUGE and BLEU substantially in its correlation with human judgments on several extractive summarization datasets for both linguistic quality and pyramid scores.

Visual Question Generation (VQG), the task of generating a question based on image contents, is an increasingly important area that combines natural language processing and computer vision. Although there are some recent works that have attempted to generate questions from images in the open domain, the task of VQG in the medical domain has not been explored so far. In this paper, we introduce an approach to generation of visual questions about radiology images called VQGR, i.e. an algorithm that is able to ask a question when shown an image. VQGR first generates new training data from the existing examples, based on contextual word embeddings and image augmentation techniques. It then uses the variational auto-encoders model to encode images into a latent space and decode natural language questions. Experimental automatic evaluations performed on the VQA-RAD dataset of clinical visual questions show that VQGR achieves good performances compared with the baseline system. The source code is available at https://github.com/sarrouti/vqgr.

Automatic summarization research has traditionally focused on providing high quality general-purpose summaries of documents. However, there are many applications which require more specific summaries, such as supporting question answering or topic-based literature discovery. In this paper we study the problem of conditional summarization in which content selection and surface realization are explicitly conditioned on an ad-hoc natural language question or topic description. Because of the difficulty in obtaining sufficient reference summaries to support arbitrary conditional summarization, we explore the use of multi-task fine-tuning (MTFT) on twenty-one natural language tasks to enable zero-shot conditional summarization on five tasks. We present four new summarization datasets, two novel “online” or adaptive task-mixing strategies, and report zero-shot performance using T5 and BART, demonstrating that MTFT can improve zero-shot summarization quality.

Question understanding is one of the main challenges in question answering. In real world applications, users often submit natural language questions that are longer than needed and include peripheral information that increases the complexity of the question, leading to substantially more false positives in answer retrieval. In this paper, we study neural abstractive models for medical question summarization. We introduce the MeQSum corpus of 1,000 summarized consumer health questions. We explore data augmentation methods and evaluate state-of-the-art neural abstractive models on this new task. In particular, we show that semantic augmentation from question datasets improves the overall performance, and that pointer-generator networks outperform sequence-to-sequence attentional models on this task, with a ROUGE-1 score of 44.16%. We also present a detailed error analysis and discuss directions for improvement that are specific to question summarization.

This paper presents the MEDIQA 2019 shared task organized at the ACL-BioNLP workshop. The shared task is motivated by a need to develop relevant methods, techniques and gold standards for inference and entailment in the medical domain, and their application to improve domain specific information retrieval and question answering systems. MEDIQA 2019 includes three tasks: Natural Language Inference (NLI), Recognizing Question Entailment (RQE), and Question Answering (QA) in the medical domain. 72 teams participated in the challenge, achieving an accuracy of 98% in the NLI task, 74.9% in the RQE task, and 78.3% in the QA task. In this paper, we describe the tasks, the datasets, and the participants’ approaches and results. We hope that this shared task will attract further research efforts in textual inference, question entailment, and question answering in the medical domain.

This paper describes our participation in SemEval-2017 Task 3 on Community Question Answering (cQA). The Question Similarity subtask (B) aims to rank a set of related questions retrieved by a search engine according to their similarity to the original question. We adapted our feature-based system for Recognizing Question Entailment (RQE) to the question similarity task. Tested on cQA-B-2016 test data, our RQE system outperformed the best system of the 2016 challenge in all measures with 77.47 MAP and 80.57 Accuracy. On cQA-B-2017 test data, performances of all systems dropped by around 30 points. Our primary system obtained 44.62 MAP, 67.27 Accuracy and 47.25 F1 score. The cQA-B-2017 best system achieved 47.22 MAP and 42.37 F1 score. Our system is ranked sixth in terms of MAP and third in terms of F1 out of 13 participating teams.

Text similarity measures are used in multiple tasks such as plagiarism detection, information ranking and recognition of paraphrases and textual entailment. While recent advances in deep learning highlighted the relevance of sequential models in natural language generation, existing similarity measures do not fully exploit the sequential nature of language. Examples of such similarity measures include n-grams and skip-grams overlap which rely on distinct slices of the input texts. In this paper we present a novel text similarity measure inspired from a common representation in DNA sequence alignment algorithms. The new measure, called TextFlow, represents input text pairs as continuous curves and uses both the actual position of the words and sequence matching to compute the similarity value. Our experiments on 8 different datasets show very encouraging results in paraphrase detection, textual entailment recognition and ranking relevance.

Readers usually rely on abstracts to identify relevant medical information from scientific articles. Abstracts are also essential to advanced information retrieval methods. More than 50 thousand scientific publications in PubMed lack author-generated abstracts, and the relevancy judgements for these papers have to be based on their titles alone. In this paper, we propose a hybrid summarization technique that aims to select the most pertinent sentences from articles to generate an extractive summary in lieu of a missing abstract. We combine i) health outcome detection, ii) keyphrase extraction, and iii) textual entailment recognition between sentences. We evaluate our hybrid approach and analyze the improvements of multi-factor summarization over techniques that rely on a single method, using a collection of 295 manually generated reference summaries. The obtained results show that the hybrid approach outperforms the baseline techniques with an improvement of 13% in recall and 4% in F1 score.

We describe a corpus of consumer health questions annotated with named entities. The corpus consists of 1548 de-identified questions about diseases and drugs, written in English. We defined 15 broad categories of biomedical named entities for annotation. A pilot annotation phase in which a small portion of the corpus was double-annotated by four annotators was followed by a main phase in which double annotation was carried out by six annotators, and a reconciliation phase in which all annotations were reconciled by an expert. We conducted the annotation in two modes, manual and assisted, to assess the effect of automatic pre-annotation and calculated inter-annotator agreement. We obtained moderate inter-annotator agreement; assisted annotation yielded slightly better agreement and fewer missed annotations than manual annotation. Due to complex nature of biomedical entities, we paid particular attention to nested entities for which we obtained slightly lower inter-annotator agreement, confirming that annotating nested entities is somewhat more challenging. To our knowledge, the corpus is the first of its kind for consumer health text and is publicly available.

This paper discusses the creation of a semantically annotated corpus of questions about patient data in electronic health records (EHRs). The goal is provide the training data necessary for semantic parsers to automatically convert EHR questions into a structured query. A layered annotation strategy is used which mirrors a typical natural language processing (NLP) pipeline. First, questions are syntactically analyzed to identify multi-part questions. Second, medical concepts are recognized and normalized to a clinical ontology. Finally, logical forms are created using a lambda calculus representation. We use a corpus of 446 questions asking for patient-specific information. From these, 468 specific questions are found containing 259 unique medical concepts and requiring 53 unique predicates to represent the logical forms. We further present detailed characteristics of the corpus, including inter-annotator agreement results, and describe the challenges automatic NLP systems will face on this task.

This paper presents a method for annotating question decomposition on complex medical questions. The annotations cover multiple syntactic ways that questions can be decomposed, including separating independent clauses as well as recognizing coordinations and exemplifications. We annotate a corpus of 1,467 multi-sentence consumer health questions about genetic and rare diseases. Furthermore, we label two additional medical-specific annotations: (1) background sentences are annotated with a number of medical categories such as symptoms, treatments, and family history, and (2) the central focus of the complex question (a disease) is marked. We present simple baseline results for automatic classification of these annotations, demonstrating the challenging but important nature of this task.