HAL-Paris 13
Not a member yet
33893 research outputs found
Sort by
Diagnostic performance of 2023 endocarditis criteria in patients with and without cardiac implantable electronic devices
No full textInternational audienceBackground The European Society of Cardiology (ESC) diagnostic criteria for infective endocarditis (IE) were updated in 2023. Using a prospective multicentre cohort of patients treated for IE, we compared the performance of the 2023 ESC classification to that of 2015 ESC, 2019 European Heart Rhythm Association (EHRA) and 2023 Duke-International Society for Cardiovascular Infectious Diseases (ISCVID) classifications in patients with or without cardiac implantable electronic devices (CIEDs). Methods A total of 1180 patients treated for IE between January 2017 and October 2022 were categorised as having either definite or possible/rejected IE within each classification. The gold standard diagnosis of IE was independently adjudicated by experts as ‘certain’ or not. Definite and ‘certain’ cases of IE were cross-tabulated to calculate sensitivity, specificity and accuracy. Suspicion of lead IE was defined by specific anatomical or metabolic features of cardiac lead infection on imaging; or by pulmonary embolism without valvular vegetation; or by positive culture of a lead extracted during open-heart surgery. Results In the 269 patients with CIEDs, 208 (77.3%) were adjudicated as ‘certain IE’; 2023 Duke-ISCVID and 2023 ESC classifications achieved higher sensitivity (97.1% and 93.8%, respectively) but markedly lower specificity (31.1% and 41.0%, respectively) than 2015 ESC and 2019 EHRA classifications. In the 159 patients with CIEDs and lead IE (78.6% ‘certain IE’), while sensitivity remained high for all classifications, specificity dropped for 2023 Duke-ISCVID and 2023 ESC classifications (11.8% and 32.4%, respectively), but not for 2019 EHRA classification (67.6%). In the 911 patients without CIEDs (81.0% ‘certain IE’), sensitivity was higher than 96% for all classifications and specificity was 55.5%, 52.6% and 48.0% for 2015 ESC, 2023 Duke-ISCVID and 2023 ESC classifications, respectively. Conclusions The 2023 ESC and 2023 Duke-ISCVID classifications had a higher sensitivity but a lower specificity than the 2015 ESC classification in patients with CIEDs, especially among those with lead IE. In patients with CIEDs, the 2019 EHRA classification was the most accurate
Au delà de l’activité contrainte, un monde animé par les motivations des bénévoles
No full textInternational audienc
Automated calculation of healthcare quality and safety indicators for head and neck cancers: a multicentric study using electronic health records
No full textInternational audienc
Nefopam Misuse: A Cross‐Sectional Study in France
No full textInternational audienceABSTRACT Background Nefopam is a non‐opioid analgesic. Nefopam is increasingly prescribed in contraindicated populations, raising questions about the appropriate use of nefopam. This study aimed to evaluate nefopam use in France according to age, epilepsy status and socioeconomic status. Methods This retrospective observational study was conducted from January 1, 2016, to December 31, 2022. Data on nefopam use in France were retrieved from the national health insurance information system. Results Between 2016 and 2022, 51.8 million boxes of nefopam were dispensed to 6.5 million patients, with a mean age of 56 years and a male‐to‐female sex ratio of 0.6. Nefopam use increased by 121.7%, whereas codeine use and tramadol use decreased by 7% and 17% respectively, raising the market share of nefopam from 5.0% to 11.6%. Nefopam use increased 2.8‐fold in patients under 15 years and 1.7‐fold in patients over 65 years, after standardisation. Nefopam use increased 1.7‐fold in patients with epilepsy and 1.2‐fold in patients with a disadvantaged socioeconomic status. The cost of nefopam was 4.3 times higher than alternative drugs. During the period 2016–2022, 8.23% of the general population received 1 to 4 boxes of nefopam per year, 4.68% received 6 to 19 boxes per year, and 0.91% received ≥ 20 boxes per year. Conclusions This study highlights a widespread inappropriate use of nefopam, especially in contraindicated populations and vulnerable groups. The recent availability of the oral form in France may worsen these patterns. Close national and European monitoring is important to assess the extent of such practices and guide future regulation. Significance Statement This work is a snapshot of nefopam usage in France. It sheds light on contraindicated uses (in subjects under 15 years of age or with epilepsy), not recommended uses (chronic usage, or in the elderly) and reports the usage of this treatment in the most deprived patients. This article cautions against underestimating the dangers and pitfalls of this treatment, especially in light of mounting evidence of pharmacodependence
La Convention Citoyenne Étudiante de l'USPN : un dispositif pédagogique participatif pour repenser l'alimentation étudiante
No full textInternational audienceL’alimentation constitue un enjeu central des transitions écologique et sociale dans l’enseignement supérieur. À l’Université Sorbonne Paris Nord (USPN), deux enquêtes quantitatives ont permis de mieux comprendre les réalités alimentaires étudiantes, en documentant l’ampleur de l’insécurité alimentaire, les comportements nutritionnels et les facteurs associés. La première enquête, menée par l’Observatoire de la formation et de la vie étudiante (OFEVE) entre novembre 2022 et janvier 2023, a recueilli les réponses de 5 068 étudiant·es. La seconde, l’enquête CoCoNut (Comprendre les COmportements NUTritionnels des étudiant·es), s’est déroulée de septembre 2023 à juin 2024 et a rassemblé 474 participant·es. Ces deux enquêtes révèlent une insécurité alimentaire préoccupante : selon l’enquête OFEVE, 11 % des étudiant·es déclaraient ne pas avoir suffisamment à manger (insécurité alimentaire quantitative), tandis que 35 % disaient ne pas toujours pouvoir accéder aux aliments souhaités (insécurité alimentaire qualitative). Ces difficultés s’accompagnent de comportements nutritionnels souvent éloignés des recommandations. Nourrie de ces constats, une Convention Citoyenne Étudiante (CCE) a été organisée en novembre 2024, avec un objectif : impliquer directement les étudiant·es dans la co-construction de propositions visant à améliorer l’accès de la communauté étudiante à une alimentation satisfaisante, favorable à la santé et durable, ainsi qu’à promouvoir la pratique régulière d’activité physique . Inspirée et accompagnée par la CCE de l’UPEC, la CCE de l’USPN a été pensée comme un projet de recherche participative. Elle a mobilisé 27 étudiant·es sélectionné·es par appel à volontariat puis par tirage au sort, représentant la diversité des campus, disciplines et niveaux d’étude. Réuni·es sur trois journées, les participant·es ont été formé·es aux enjeux alimentaires et ont délibéré autour de cinq axes : 1. Réduction de la précarité alimentaire, 2. Amélioration de la qualité nutritionnelle des repas, 3. Promotion de l’alimentation durable, 4. Rapport au corps et comportements alimentaires et 5. Activité physique et mobilités actives.Les 31 propositions formulées touchent principalement aux déterminants environnementaux et politiques de l’alimentation. Elles incluent par exemple : la révision de l’offre alimentaire du CROUS, l’élargissement des horaires des restaurants universitaires et de la pause méridienne, l’affichage du Nutri-Score, ou encore l’organisation d’ateliers de cuisine. Un vote final, impliquant les étudiant·es et les partenaires institutionnels, a permis de prioriser les actions à mettre en œuvre.La CCE s’inscrit dans une démarche de transformation systémique des environnements alimentaires, en dépassant la seule responsabilisation individuelle. Elle combine pédagogie active, recherche appliquée et gouvernance participative. Ce format court, reproductible et adaptable, a suscité un fort engagement des participant·es et un intérêt marqué des parties prenantes universitaires.Nous proposons d’exposer cette expérience lors du colloque ETES 2025 afin d’en partager les enseignements. La Convention Citoyenne Étudiante de l’USPN constitue un outil pédagogique innovant pour enseigner les transitions écologiques et sociales dans le supérieur. En donnant une voix aux étudiant·es sur les enjeux alimentaires qui les concernent directement, elle favorise le développement de compétences citoyennes, l’ancrage local des politiques de transition, et la co-construction de solutions adaptées et durables
Translects: Postqueering Transgender in South African and Nigerian Autofictions: Postqueering Transgender in South African and Nigerian Auofictions
No full textInternational audienceTranslects—our coinage (Zabus & Das, 2020)— are transnational, transgender-inflected terms rooted in ancestral contexts. Hinging on ‘transing’ and ‘translating’, I examine the use of translects in ‘autofictions’—South African Zandile Ngozi Nkabinde’s Black Bull, Ancestors and Me (2008), which I contrast with South African Anastacia Thomson’s Always Anastacia (2015), and Nigerian-born, US-based, Igbo-Tamil writer, Akwaeke Emezi’s Freshwater (2018) and Dear Senthuran: A Black Spirit Memoir (2021)—to reflect on a ‘post-queer’ and post-secular turn in approaching transgender identities and personhoods, which translate into various shades of postcolonial naming practices in Sub-Saharan Africa
Latin-American cyborg methods: hacking digital methodologies in and for the scientific south
No full textInternational audienceThis article aims to critically examine the multifaceted challenges faced by Latin American scientists when using computational methods for data production and analysis, amidst limited access to essential knowledge, know-how, tools, infrastructure, and resources. We use an experimental, collaborative, and multi-situated methodological strategy grounded in participatory observation, in-depth interviews, ethnography, and autoethnography, conducted across different scientific contexts and fields in Mexico, Guatemala, Ecuador, Argentina, and Chile. Based on what we call a “cyborg methods” approach, which aims to understand the use, development, and circulation of these tools as intertwined with our own activity as researchers and situated in our local context, we explore different strategies that Latin American scientists use to navigate the precariousness and complexity of conducting research in the South, with a particular emphasis on the innovative appropriation of digital tools within resource-constrained environments and its transformative potential in overcoming those limitations. Our findings are synthesized into five key areas: data scarcity, infrastructural constraints, lack of formal training, methodological innovations, and copyright infringement as a means to access essential tools and data
Evaluation of two-time zones online training to transform older people's care
No full textInternational audienc
Leveraging Guideline-Based Clinical Decision Support Systems with Large Language Models: A Case Study with Breast Cancer
No full textInternational audienceAbstract Background Multidisciplinary tumor boards (MTBs) have been established in most countries to allow experts collaboratively determine the best treatment decisions for cancer patients. However, MTBs often face challenges such as case overload, which can compromise MTB decision quality. Clinical decision support systems (CDSSs) have been introduced to assist clinicians in this process. Despite their potential, CDSSs are still underutilized in routine practice. The emergence of large language models (LLMs), such as ChatGPT, offers new opportunities to improve the efficiency and usability of traditional CDSSs. Objectives OncoDoc2 is a guideline-based CDSS developed using a documentary approach and applied to breast cancer management. This study aims to evaluate the potential of LLMs, used as question-answering (QA) systems, to improve the usability of OncoDoc2 across different prompt engineering techniques (PETs). Methods Data extracted from breast cancer patient summaries (BCPSs), together with questions formulated by OncoDoc2, were used to create prompts for various LLMs, and several PETs were designed and tested. Using a sample of 200 randomized BCPSs, LLMs and PETs were initially compared with regard to their responses to OncoDoc2 questions using classic metrics (accuracy, precision, recall, and F1 score). Best performing LLMs and PETs were further assessed by comparing the therapeutic recommendations generated by OncoDoc2, based on LLM inputs, to those provided by MTB clinicians using OncoDoc2. Finally, the best performing method was validated using a new sample of 30 randomized BCPSs. Results The combination of Mistral and OpenChat models under the enhanced Zero-Shot PET showed the best performance as a question-answering system. This approach gets a precision of 60.16%, a recall of 54.18%, an F1 score of 56.59%, and an accuracy of 75.57% on the validation set of 30 BCPSs. However, this approach yielded poor results as a CDSS, with only 16.67% of the recommendations generated by OncoDoc2 based on LLM inputs matching the gold standard. Conclusion All the criteria in the OncoDoc2 decision tree are crucial for capturing the uniqueness of each patient. Any deviation from a criterion alters the recommendations generated. Despite achieving a good accuracy rate of 75.57%, LLMs still face challenges in reliably understanding complex medical contexts and be effective as CDSSs
One Size Does Not Fit All: Improving Breast Cancer Patient Retrieval Using Variable-Tailored Similarity Measures
No full textInternational audienceClinical decision support systems (CDSSs) face challenges in managing complex patient cases that fall outside clinical practice guidelines. Using similarity-based approaches to retrieve comparable patients and apply the therapeutic care plans they received has the potential to support decision-making for complex cases. However, the choice of similarity metrics is difficult. We developed a hybrid similarity-based method (HM) which combines multiple similarity measures tailored to the specific characteristics of clinical variables aggregated with expert-defined weights. An interactive interface allowing clinicians to adjust weights in real time has been proposed. Assessed on breast cancer management, with expert-annotated datasets, HM achieved a recall of 0.6 for the top 5 most similar patients and retrieved 27 out of 45 clinically relevant cases. These results emphasize the importance of specific similarity measures in providing an effective, explainable, and customizable solution for patient retrieval of complex breast cancer cases