Background: General-domain sentiment models have been found ineffective in distinguishing positivity and negativity in Intensive Care Unit (ICU) clinical notes and domain-specific sentiment models are recommended. Although there are multiple common approaches to sentiment analysis, there has been little work comparing and evaluating the effectiveness of specialized models, largely due to the difficulty of recruiting clinical annotators and of accessing clinical sentiment data. This study has three contributions: (1) MIMIC-III-Ext-Notes-Sentiment: the first public ICU-specific ground- truth dataset labeled by clinicians for investigating clinical sentiment polarity in ICU clinical notes. (2) SentimentICUModel: an effective model for classifying clinical sentiment in ICU narratives on the ground truth. (3) A guiding comparison of the effectiveness of a range of approaches to clinical sentiment classification on the dataset. Methods: We recruited five clinicians to annotate notes for the ground truth. Annotators indicated which pieces of note text influenced their labeling. Six clinical-specific models were compared on the ground truth. Results: The task of annotation was challenging due to clinicians’ workload and spanned 15 months. The ground truth data was formed based on inter-annotator agreement analysis. Clinicians’ extracts similarity aligned with their agreement level. Clinical language models provide comparable accuracy (up to 82%), with top score achieved by ClinicalT5 which is being released as SentimentICUModel. They outperform keyword-based lexicon (p $$ < 0.05$$ , [95% CI, −0.47, −0.28]). Conclusion: Clinical language models have demonstrated effectiveness in identifying clinical sentiment within clinical notes, enabling early detection of sudden changes and exploring different patterns in patients’ ICU stays.
Debates are an essential democratic institution in danger by the rise of Social Media. The advent of Fake News often referred to as the ‘crisis of trust’, has led to a substantial increase in debates that blend online and offline. It can be argued that blended approaches are not directly linked to increasing trustworthiness in the debate. To overcome this trust crisis and increase the reliability in debates, we introduce the HELIOSPHERE concept that seeks to use technological advances, such as Artificial Intelligence and Augmented Reality, to create a more fair, inclusive and transparent debate. The critical component for inclusiveness is Augmented Reality technology and 3D camera technology to hybridise the online and offline debating space and ensure that anyone who cannot be present can engage with the debate. For transparency and fairness, a key indicator of trust, an Artificial Intelligence dashboard is introduced to analyse and visualise speaking time, speaker gender, topic relatedness, bias detection sentiment in Real-Time. This work presents the overall theoretical concept focusing on academic and technical concepts to support reliable communication within debates.