Negation detection in Dutch clinical texts: an evaluation of rule-based and machine learning methods.

When developing models for clinical information retrieval and decision support systems, the discrete outcomes required for training are often missing. These labels need to be extracted from free text in electronic health records. For this extraction process one of the most important contextual properties in clinical text is negation, which indicates the absence of findings. We aimed to improve large scale extraction of labels by comparing three methods for negation detection in Dutch clinical notes. We used the Erasmus Medical Center Dutch Clinical Corpus to compare a rule-based method based on ContextD, a biLSTM model using MedCAT and (finetuned) RoBERTa-based models. We found that both the biLSTM and RoBERTa models consistently outperform the rule-based model in terms of F1 score, precision and recall. In addition, we systematically categorized the classification errors for each model, which can be used to further improve model performance in particular applications. Combining the three models naively was not beneficial in terms of performance. We conclude that the biLSTM and RoBERTa-based models in particular are highly accurate accurate in detecting clinical negations, but that ultimately all three approaches can be viable depending on the use case at hand.

Transforming epilepsy research: A systematic review on natural language processing applications.

Despite improved ancillary investigations in epilepsy care, patients' narratives remain indispensable for diagnosing and treatment monitoring. This wealth of information is typically stored in electronic health records and accumulated in medical journals in an unstructured manner, thereby restricting complete utilization in clinical decision-making. To this end, clinical researchers increasing apply natural language processing (NLP)-a branch of artificial intelligence-as it removes ambiguity, derives context, and imbues standardized meaning from free-narrative clinical texts. This systematic review presents an overview of the current NLP applications in epilepsy and discusses the opportunities and drawbacks of NLP alongside its future implications. We searched the PubMed and Embase databases with a "natural language processing" and "epilepsy" query (March 4, 2022) and included original research articles describing the application of NLP techniques for textual analysis in epilepsy. Twenty-six studies were included. Fifty-eight percent of these studies used NLP to classify clinical records into predefined categories, improving patient identification and treatment decisions. Other applications of NLP had structured clinical information retrieval from electronic health records, scientific papers, and online posts of patients. Challenges and opportunities of NLP applications for enhancing epilepsy care and research are discussed. The field could further benefit from NLP by replicating successes in other health care domains, such as NLP-aided quality evaluation for clinical decision-making, outcome prediction, and clinical record summarization.

Spatiotemporal variations of public opinion on social distancing in the Netherlands: Comparison of Twitter and longitudinal survey data.

Background: Social distancing has been implemented by many countries to curb the COVID-19 pandemic. Understanding public support for this policy calls for effective and efficient methods of monitoring public opinion on social distancing. Twitter analysis has been suggested as a cheaper and faster-responding alternative to traditional survey methods. The current empirical evidence is mixed in terms of the correspondence between the two methods. Objective: We aim to compare the two methods in the context of monitoring the Dutch public's opinion on social distancing. For this comparison, we quantified the temporal and spatial variations in public opinion and their sensitivities to critical events using data from both Dutch Twitter users and respondents from a longitudinal survey. Methods: A longitudinal survey on a representative Dutch sample (n = 1,200) was conducted between July and November 2020 to measure opinions on social distancing weekly. From the same period, near 100,000 Dutch tweets were categorized as supporting or rejecting social distancing based on a model trained with annotated data. Average stances for the 12 Dutch provinces and over the 20 weeks were computed from the two data sources and were compared through visualizations and statistical analyses. Results: Both data sources suggested strong support for social distancing, but public opinion was much more varied among tweets than survey responses. Both data sources showed an increase in public support for social distancing over time, and a strong temporal correspondence between them was found for most of the provinces. In addition, the survey but not Twitter data revealed structured differences among the 12 provinces, while the two data sources did not correspond much spatially. Finally, stances estimated from tweets were more sensitive to critical events happened during the study period. Conclusions: Our findings indicate consistencies between Twitter data analysis and survey methods in describing the overall stance on social distancing and temporal trends. The lack of spatial correspondence may imply limitations in the data collections and calls for surveys with larger regional samples. For public health management, Twitter analysis can be used to complement survey methods, especially for capturing public's reactivities to critical events amid the current pandemic.