• Publicationsclose
• Research softwareclose
• 2019-2023close
• Other literature type close
• FR close
• Hal-Diderot close
• Mémoires en Sciences de l'Information et de la Communication close
• COVID-19 close
• Digital Humanities and Cultural Heritage close

International audience; During the early spring of 2020, as the COVID-19 (coronavirus disease 2019) pandemic spread across the globe like a viral tsunami, many countries urgently implemented unprecedented mitigation measures to slow it down, following the example of China, where the disease first emerged at the end of 2019. As a result, more than 3 billion unprepared people worldwide had to cope with living under stringent lockdown measures for weeks or months, depending on the country. Together with the deaths of hundreds of thousands of people from this disease, a frightening emotional shock has touched populations everywhere around the planet. Millions have lost their jobs, others have had their personal lives changed drastically; the elderly, for example, have sometimes been isolated from any visits. To our knowledge, no guidelines to prevent psychological trauma have been proposed, even for the most vulnerable individuals. Nonetheless, before this crisis began, several experts had warned about the potential impact of such measures on the populations’ mental health, based on previous experiences of quarantine, as summarised in a review published, as this pandemic reached Europe and before any lockdowns.1 This warning was confirmed a few days later by the publication of a nationwide study conducted in China reporting that the quarantine triggered a wide variety of psychological disorders.2 In France, one of the hardest hit countries in Europe, a nationwide lockdown started on 17 March and lasted till 11 May. Public health authorities made daily recommendations on how to prevent COVID-19 infection with almost no specific advice on psychological prevention.

International audience

La nécessité d’avoir des indicateurs de santé publique fiable est primordial pour avoir des réponses adaptées au niveau des moyens en particulier dans le contexte actuel de pandémie à COVID-19. En France, la régulation médicale permet d’obtenir une quantité de données importantes sur les appels passés. Nous avons obtenu grâce au « Natural langage processing » les données de 2010 à 2020 de la régulation au SMUR 33 concernant les symptômes grippaux de tous les patients soit plus de 2 millions textes libres, avec une catégorisation selon l’âge. Nous avons pris des patients de plus de 75 ans, mais également la catégorie des 65-75 ans. Les résultats montrent une augmentation globale du nombre d’appels, toutes causes confondues et pour syndrome grippal. Avec des catégories d’âge ayant des majorations différentes : les plus de 65 ans ayant le plus fort taux d’augmentation du nombre d’appel total et pour syndrome grippal. De plus, l’année 2020 avec l’émergence de la COVID 19 trouve une augmentation extrême des appels pour syndromes grippaux sur une courte période avec cependant une proportion moins importante des appels passés par les plus de 75 ans. L’utilisation de l’intelligence artificielle avec traitement naturel de langage des contenus des dossiers de régulation médicale permet une surveillance épidémiologique efficace et permet dans des situations sanitaires exceptionnelles d’avoir un outil supplémentaire efficient et accessible. The need for reliable public health indicators is crucial for appropriate resource responses, particularly in the current context of the COVID-19 pandemic. In France, medical regulation offers a significant amount of data on the calls made. Thanks to "Natural language processing", we have obtained data from 2010 to 2020 from SMUR 33 regulation concerning the flu symptoms of all patients, i.e. more than 2 million free texts, categorized according to age. We took patients over 75 years of age, but also the category 65-75 years of age. The results show an overall increase in the number of calls, all causes combined and for flu-like symptoms. With different age categories having different increases: the over 65s having the highest rate of increase in the total number of calls and for flu syndrome. Moreover, the year 2020 with the emergence of COVID 19 finds an extreme increase in the number of calls for flu syndromes over a short period of time with, however, a lower proportion of calls made by those over 75 years of age. The use of artificial intelligence with natural language processing of the contents of medical regulation files enables effective epidemiological surveillance and, in exceptional health situations, provides an additional efficient and accessible tool.

BackgroundA novel disease poses special challenges for informatics solutions. Biomedical informatics relies for the most part on structured data, which require a preexisting data or knowledge model; however, novel diseases do not have preexisting knowledge models. In an emergent epidemic, language processing can enable rapid conversion of unstructured text to a novel knowledge model. However, although this idea has often been suggested, no opportunity has arisen to actually test it in real time. The current coronavirus disease (COVID-19) pandemic presents such an opportunity. ObjectiveThe aim of this study was to evaluate the added value of information from clinical text in response to emergent diseases using natural language processing (NLP). MethodsWe explored the effects of long-term treatment by calcium channel blockers on the outcomes of COVID-19 infection in patients with high blood pressure during in-patient hospital stays using two sources of information: data available strictly from structured electronic health records (EHRs) and data available through structured EHRs and text mining. ResultsIn this multicenter study involving 39 hospitals, text mining increased the statistical power sufficiently to change a negative result for an adjusted hazard ratio to a positive one. Compared to the baseline structured data, the number of patients available for inclusion in the study increased by 2.95 times, the amount of available information on medications increased by 7.2 times, and the amount of additional phenotypic information increased by 11.9 times. ConclusionsIn our study, use of calcium channel blockers was associated with decreased in-hospital mortality in patients with COVID-19 infection. This finding was obtained by quickly adapting an NLP pipeline to the domain of the novel disease; the adapted pipeline still performed sufficiently to extract useful information. When that information was used to supplement existing structured data, the sample size could be increased sufficiently to see treatment effects that were not previously statistically detectable.

The high neutrophil to lymphocyte ratio observed in critically ill patients with COVID-19 is associated with excessive levels of reactive oxygen species (ROS), which promote a cascade of biological events that drive pathological host responses. ROS induce tissue damage, thrombosis and red blood cell dysfunction, which contribute to COVID-19 disease severity. We suggest that free radical scavengers could be beneficial for the most vulnerable patients. In this Comment article, Becker and colleagues consider how the excessive release of reactive oxygen species by neutrophils may perpetuate red blood cell dysfunction, thrombosis and tissue damage in severe cases of COVID-19.

Therapies - Sous presse. Epreuves corrigees par l'auteur. Disponible en ligne depuis le jeudi 10 juin 2021