A5034693908- Biomedical Text Mining and Ontologies
- Semantic Web and Ontologies
- Pharmacovigilance and Adverse Drug Reactions
- Electronic Health Records Systems
- Computational Drug Discovery Methods
- Natural Language Processing Techniques
- Data Quality and Management
- Topic Modeling
- Clinical practice guidelines implementation
- AI in cancer detection
- linguistics and terminology studies
- Biosimilars and Bioanalytical Methods
- Mobile Health and mHealth Applications
- Rough Sets and Fuzzy Logic
- Artificial Intelligence in Healthcare
- Academic integrity and plagiarism
- Cognitive Science and Mapping
- AI-based Problem Solving and Planning
- Renal and Vascular Pathologies
- Cognitive Computing and Networks
- Image Retrieval and Classification Techniques
- Scientific Computing and Data Management
- Service-Oriented Architecture and Web Services
- Genomics and Rare Diseases
- Health Systems, Economic Evaluations, Quality of Life
Sorbonne Université
2015-2025
Inserm
2015-2025
Laboratoire d'Informatique Médicale et d'Ingénierie des Connaissances en e-Santé
2016-2025
Université Sorbonne Paris Nord
2014-2024
Département d'Informatique
1997-2024
Université Paris Cité
2005-2020
Aristotle University of Thessaloniki
2020
Sorbonne Paris Cité
2014-2019
Institut National de Recherche en Santé Publique
2016
Université Sorbonne Nouvelle
2011-2015
Abstract Artificial intelligence (AI) is a powerful technology with the potential to disrupt cancer detection, diagnosis and treatment. However, development of new AI algorithms requires access large complex real-world datasets. Although such datasets are constantly being generated, them limited by data fragmentation across numerous repositories sites, heterogeneity, lack annotations, privacy issues. The European Cancer Imaging Initiative flagship Europe’s Beating Plan, aiming unlock power...
There are different approaches for repurposing clinical data collected in the Electronic Healthcare Record (EHR) use research. Semantic integration of "siloed" applications across domain boundaries is raison d'être standards-based profiles developed by Integrating Enterprise (IHE) initiative – an healthcare professionals and industry promoting coordinated established standards such as DICOM HL7 to address specific needs support optimal patient care. In particular, combination two IHE profile...
Artificial intelligence tools promise transformative impacts in drug development. Regulatory agencies face challenges integrating AI while ensuring reliability and safety clinical trial approvals, marketing authorizations, post-market surveillance. Incorporating these technologies into the existing regulatory framework agency practices poses notable challenges, particularly evaluating data models employed for purposes. Rapid adaptation of regulations internal processes is essential to keep...
Each year, the International Medical Informatics Association Yearbook recognizes significant scientific papers, labelled as "best papers", published previous year in subfields of biomedical informatics that correspond to different section topics journal. For each section, about fifteen pre-selected "candidate" best papers are externally peer-reviewed select actual papers. Although based on available literature, little is known pre-selection process.To move toward an explicit formalization...
Cardiovascular diseases are a major cause of death worldwide. Mobile health apps could help in preventing cardiovascular by improving modifiable risk factors such as eating habits, physical activity levels, and alcohol or tobacco consumption.The aim this study was to design mobile app, Prevent Connect, assess its quality for (1) assessing patient behavior 4 (unhealthy eating, sedentary lifestyle, alcohol, consumption) (2) suggesting personalized recommendations interventions risky...
Computational signal detection constitutes a key element of postmarketing drug monitoring and surveillance. Diverse data sources are considered within the 'search space' pharmacovigilance scientists, respective analysis methods employed, all with their qualities shortcomings, towards more timely accurate detection. Recent systematic comparative studies highlighted not only event-based data-source-based differential performance across but also complementarity. These findings reinforce...
Background:As more scientific work is published, it important to improve access the biomedical literature. Since 2000, when Medical Subject Headings (MeSH) Concepts were introduced, MeSH Thesaurus has been concept based. Nevertheless, information retrieval still performed at Descriptor or Supplementary Concept level.Objective:The study assesses benefit of using for indexing and retrieval.Methods:Three sets queries built thirty-two rare diseases twenty-two chronic diseases: (1) PubMed...
Pharmacovigilance signal report (PVSR) documents contain valuable condensed information published by drug monitoring organizations, typically in a free-text format. They provide initial insights into potential links between drugs and harmful effects. Still, their unstructured format prevents this from being integrated data-processing pipelines (e.g., to support either the investigation of safety signals or decision-making clinical context). OpenPVSignal is data model designed specifically...
L'utilisation des Systèmes d'Aide au Recrutement dans les Essais Cliniques (SAREC) appliqués aux données massives de santé pourrait faciliter l'inclusion patients essais cliniques. La revue exploratoire la littérature a porté sur l'intégration en vie courante SARECs dossiers informatisés (DPI) et ayant bénéficié d'une évaluation scientifique. MéTHODES: Dans cette revue, publications ont été extraites PubMed avec mots-clés « cliniques comme sujet », sélection » entrepôt ». Deux investigateurs...
For an intelligent entity to carry out tasks in the real world, perceived three-dimensional shapes are transformed into objects identified by their functional category which makes explicit roles or uses of actions. This study describes approach recognition functions that combines ideas about representations shapes, concepts, and object categories, with goal requirements A particular conceptual model compatibility between actions is introduced, outline solution given, experimental domain hand...
The aim of this study was to evaluate the relevance signals generated by a computerized drug-drug interaction detection system and design classification overridden alerts.Prospective over two months.Five hundred ten-bed university paediatric hospital.In Robert Debré Hospital physicians generate drug orders online using physician order entry that also detects interactions in real time. We analysed sample alerts physicians.We 613 alerts. defined three categories alerts: informational errors...
WHO-ART and MedDRA are medical terminologies used for the coding of adverse drug reactions in pharmacovigilance databases. proposes 13 Special Search Categories (SSC) grouping terms associated to specific conditions. For instance, SSC "Haemorrhage" includes 346 among which 55 also terms. itself does not provide such groupings. Our main contention is possibility classifying semantic categories by using knowledge extracted from SNOMED CT. A previous paper presents way term definitions have...
Objective: Driven by the need of pharmacovigilance centres and companies to routinely collect review all available data about adverse drug reactions (ADRs) events interest, we introduce validate a computational framework exploiting dominant as well emerging publicly sources for safety surveillance.Methods: Our approach relies on appropriate query formulation acquisition subsequent filtering, transformation joint visualization obtained data. We acquired from FDA Adverse Event Reporting System...