Automatic Fact Checking Using an Interpretable Bert-Based Architecture on COVID-19 Claims

We present a neural network architecture focused on verifying facts against evidence found in a knowledge base. The architecture can perform relevance evaluation and claim verification, parts of a well-known three-stage method of fact-checking. We fine-tuned BERT to codify claims and pieces of evidence separately. An attention layer between the claim and evidence representation computes alignment scores to identify relevant terms between both. Finally, a classification layer receives the vector representation of claims and evidence and performs the relevance and verification classification. Our model allows a more straightforward interpretation of the predictions than other state-of-the-art models. We use the scores computed within the attention layer to show which evidence spans are more relevant to classify a claim as supported or refuted. Our classification models achieve results compared to the state-of-the-art models in terms of classification of relevance evaluation and claim verification accuracy on the FEVER dataset.

Physiotherapy practices when treating patients with COVID-19 during a pandemic: A survey study.

Background Specific details pertaining to the clinical and other challenges faced by physiotherapists managing patients with COVID-19 during the pandemic are still largely unknown. Objectives To determine how physiotherapists clinically managed patients with COVID-19 in a hospital-based setting during the pandemic and to identify the personal and professional effects of working as a physiotherapist at this time. Methods Self-administered electronic cross-sectional survey. Participants included physiotherapists from around the world involved in the clinical management of patients with COVID-19. Results Of the 204 participants who returned the questionnaire, 39% worked as senior physiotherapists, 29% as consultant or specialist physiotherapists, 23% as general physiotherapists and 4% as graduate physiotherapists. Seventy-two percent of participants worked in the intensive care unit. The largest barrier to treating patients with COVID-19 was a lack of intensive care trained physiotherapists (70%). Eighty-three percent of participants reported performing activities outside of their typical work duties, including proning patients (55%), tutoring and advising other staff in the intensive care unit (55%) and adjusting or changing ventilator settings (52%). Almost all participants (90%) reported being aware of physiotherapy specific guidelines for treating patients with COVID-19, yet most participants performed techniques that were not recommended. Conclusions The experience of the pandemic highlighted the need for specialist training and availability of experienced cardiorespiratory physiotherapists to manage patients with COVID-19, specifically in intensive care. Furthermore, clear guidelines on the management of patients with COVID-19 should be established to ensure optimal management of patients and ensure the safety of physiotherapy staff.