A growing body of work uses Natural Language Processing (NLP) methods to automatically generate medical notes from audio recordings of doctor-patient consultations. However, there are very few studies on how such systems could be used in clinical practice, how clinicians would adjust to using them, or how system design should be influenced by such considerations. In this paper, we present three rounds of user studies, carried out in the context of developing a medical note generation system. We present, analyse and discuss the participating clinicians’ impressions and views of how the system ought to be adapted to be of value to them. Next, we describe a three-week test run of the system in a live telehealth clinical practice. Major findings include (i) the emergence of five different note-taking behaviours; (ii) the importance of the system generating notes in real time during the consultation; and (iii) the identification of a number of clinical use cases that could prove challenging for automatic note generation systems.

Evaluating automatically generated text is generally hard due to the inherently subjective nature of many aspects of the output quality. This difficulty is compounded in automatic consultation note generation by differing opinions between medical experts both about which patient statements should be included in generated notes and about their respective importance in arriving at a diagnosis. Previous real-world evaluations of note-generation systems saw substantial disagreement between expert evaluators. In this paper we propose a protocol that aims to increase objectivity by grounding evaluations in Consultation Checklists, which are created in a preliminary step and then used as a common point of reference during quality assessment. We observed good levels of inter-annotator agreement in a first evaluation study using the protocol; further, using Consultation Checklists produced in the study as reference for automatic metrics such as ROUGE or BERTScore improves their correlation with human judgements compared to using the original human note.

In recent years, machine learning models have rapidly become better at generating clinical consultation notes; yet, there is little work on how to properly evaluate the generated consultation notes to understand the impact they may have on both the clinician using them and the patient’s clinical safety. To address this we present an extensive human evaluation study of consultation notes where 5 clinicians (i) listen to 57 mock consultations, (ii) write their own notes, (iii) post-edit a number of automatically generated notes, and (iv) extract all the errors, both quantitative and qualitative. We then carry out a correlation study with 18 automatic quality metrics and the human judgements. We find that a simple, character-based Levenshtein distance metric performs on par if not better than common model-based metrics like BertScore. All our findings and annotations are open-sourced.

Recent advances in Automatic Speech Recognition (ASR) have made it possible to reliably produce automatic transcripts of clinician-patient conversations. However, access to clinical datasets is heavily restricted due to patient privacy, thus slowing down normal research practices. We detail the development of a public access, high quality dataset comprising of 57 mocked primary care consultations, including audio recordings, their manual utterance-level transcriptions, and the associated consultation notes. Our work illustrates how the dataset can be used as a benchmark for conversational medical ASR as well as consultation note generation from transcripts.

We propose a method for evaluating the quality of generated text by asking evaluators to count facts, and computing precision, recall, f-score, and accuracy from the raw counts. We believe this approach leads to a more objective and easier to reproduce evaluation. We apply this to the task of medical report summarisation, where measuring objective quality and accuracy is of paramount importance.

Automatic summarisation has the potential to aid physicians in streamlining clerical tasks such as note taking. But it is notoriously difficult to evaluate these systems and demonstrate that they are safe to be used in a clinical setting. To circumvent this issue, we propose a semi-automatic approach whereby physicians post-edit generated notes before submitting them. We conduct a preliminary study on the time saving of automatically generated consultation notes with post-editing. Our evaluators are asked to listen to mock consultations and to post-edit three generated notes. We time this and find that it is faster than writing the note from scratch. We present insights and lessons learnt from this experiment.

Word embedding-based similarity measures are currently among the top-performing methods on unsupervised semantic textual similarity (STS) tasks. Recent work has increasingly adopted a statistical view on these embeddings, with some of the top approaches being essentially various correlations (which include the famous cosine similarity). Another excellent candidate for a similarity measure is mutual information (MI), which can capture arbitrary dependencies between the variables and has a simple and intuitive expression. Unfortunately, its use in the context of dense word embeddings has so far been avoided due to difficulties with estimating MI for continuous data. In this work we go through a vast literature on estimating MI in such cases and single out the most promising methods, yielding a simple and elegant similarity measure for word embeddings. We show that mutual information is a viable alternative to correlations, gives an excellent signal that correlates well with human judgements of similarity and rivals existing state-of-the-art unsupervised methods.

Similarity measures based purely on word embeddings are comfortably competing with much more sophisticated deep learning and expert-engineered systems on unsupervised semantic textual similarity (STS) tasks. In contrast to commonly used geometric approaches, we treat a single word embedding as e.g. 300 observations from a scalar random variable. Using this paradigm, we first illustrate that similarities derived from elementary pooling operations and classic correlation coefficients yield excellent results on standard STS benchmarks, outperforming many recently proposed methods while being much faster and trivial to implement. Next, we demonstrate how to avoid pooling operations altogether and compare sets of word embeddings directly via correlation operators between reproducing kernel Hilbert spaces. Just like cosine similarity is used to compare individual word vectors, we introduce a novel application of the centered kernel alignment (CKA) as a natural generalisation of squared cosine similarity for sets of word vectors. Likewise, CKA is very easy to implement and enjoys very strong empirical results.

A large body of research into semantic textual similarity has focused on constructing state-of-the-art embeddings using sophisticated modelling, careful choice of learning signals and many clever tricks. By contrast, little attention has been devoted to similarity measures between these embeddings, with cosine similarity being used unquestionably in the majority of cases. In this work, we illustrate that for all common word vectors, cosine similarity is essentially equivalent to the Pearson correlation coefficient, which provides some justification for its use. We thoroughly characterise cases where Pearson correlation (and thus cosine similarity) is unfit as similarity measure. Importantly, we show that Pearson correlation is appropriate for some word vectors but not others. When it is not appropriate, we illustrate how common non-parametric rank correlation coefficients can be used instead to significantly improve performance. We support our analysis with a series of evaluations on word-level and sentence-level semantic textual similarity benchmarks. On the latter, we show that even the simplest averaged word vectors compared by rank correlation easily rival the strongest deep representations compared by cosine similarity.

This paper presents a linguistic processing pipeline for Bulgarian including morphological analysis, lemmatization and syntactic analysis of Bulgarian texts. The morphological analysis is performed by three modules ― two statistical-based and one rule-based. The combination of these modules achieves the best result for morphological tagging of Bulgarian over a rich tagset (680 tags). The lemmatization is based on rules, generated from a large morphological lexicon of Bulgarian. The syntactic analysis is implemented via MaltParser. The two statistical morphological taggers and MaltParser are trained on datasets constructed within BulTreeBank project. The processing pipeline includes also a sentence splitter and a tokenizer. All tools in the pipeline are packed in modules that can also perform separately. The whole pipeline is designed to be able to serve as a back-end of a web service oriented interface, but it also supports the user tasks with a command-line interface. The processing pipeline is compatible with the Text Corpus Format, which allows it to delegate the management of the components to the WebLicht platform.

Our goal is to provide a web-based platform for the long-term preservation and distribution of a heterogeneous collection of linguistic resources. We discuss the corpus preprocessing and normalisation phase that results in sets of multi-rooted trees. At the same time we transform the original metadata records, just like the corpora annotated using different annotation approaches and exhibiting different levels of granularity, into the all-encompassing and highly flexible format eTEI for which we present editing and parsing tools. We also discuss the architecture of the sustainability platform. Its primary components are an XML database that contains corpus and metadata files and an SQL database that contains user accounts and access control lists. A staging area, whose structure, contents, and consistency can be checked using tools, is used to make sure that new resources about to be imported into the platform have the correct structure.