9 results on '"D Cauchois"'
Search Results
2. Retour d'expérience sur le projet de chaînage entre les données hospitalières du Centre Oscar Lambret et les fichiers de décès de l'Insee
- Author
-
M. Barthoulot, F. Craynest, A. Hammoudi, E. Aymes, D. Cauchois, M. Cucchi, M. Le Deley, and E. Lartigau
- Subjects
Epidemiology ,Public Health, Environmental and Occupational Health - Published
- 2022
- Full Text
- View/download PDF
3. Plateforme de données de vie réelle ESME. Constitution d’une liste de sélection exhaustive multi-source
- Author
-
O. Payen, D Cauchois, A.V. Guizard, C. Courtinard, Gaëtane Simon, A Doly, M. Velten, D. Pérol, B Favier, T. Guesmia, M. Mons, I Piot, G Perrocheaux, M. Robain, D Berchery, and A Loeb
- Subjects
Epidemiology ,Public Health, Environmental and Occupational Health - Abstract
Introduction Les donnees de « vie reelle » sont des donnees cliniques et therapeutiques recueillies a grande echelle pour decrire la prise en charge reelle des patients et ses resultats. Ces donnees sont complementaires a celles des essais cliniques, chaque approche ayant son interet. L’exhaustivite de la population selectionnee est un point majeur pour la validite des donnees de vie reelle et repose sur une methodologie de selection parfois complexe. Methodes La selection des patients dans le premier projet ESME, patients avec une premiere prise en charge entre 2008 et 2013 pour un cancer du sein metastatique (CSM) dans les 20 sites des 18 centres de lutte contre le cancer (CLCC), a ete realisee a partir de differentes sources d’information structurees ou non structurees : programme medicalise des systemes d’information (PMSI), dossier pharmaceutique, bases locales specifiques, base des reunions de concertation pluridisciplinaire (RCP), moteurs de recherche internes. La qualification de chaque type de liste a ete prealablement realisee sur echantillons avant de decider l’utilisation de la source d’information dans la demarche. Les listes de preselection issues des differentes sources utilisees ont ete fusionnees afin de constituer une liste unique. La methodologie maximisait la sensibilite tout en essayant de conserver une bonne specificite. Le retour systematique aux Dossiers patients informatises (DPI) a permis de valider la selection de chaque patient. Des controles qualites des cas non selectionnes et selectionnes ainsi qu’un audit du processus complet de selection ont ete realises. Resultats Au total 34 484 patients ont ete preselectionnes. Ce travail de selection a necessite de s’appuyer sur les outils et les informations accessibles dans les centres. Apres retour a l’ensemble des dossiers des patients preselectionnes, 14 022 patients ont effectivement ete selectionnes. La contribution des sources de preselection a la selection finale est decrite dans la Tableau 1 . La proportion de patients effectivement selectionnes (40,7 %) differait d’un centre a l’autre (28 % a 58 %), cependant, le nombre final de patients effectivement selectionnes dans chaque centre correspondait a l’estimation initiale faite pour chaque centre a partir d’informations extrapolees du PMSI national. Conclusion L’utilisation de sources d’information differentes etait indispensable pour selectionner l’ensemble des patients de notre population, en particulier ceux suivis uniquement en consultation et accessibles a partir des moteurs de recherche uniquement. Le taux de conversion preselection/selection etait globalement inferieur a celui attendu. Nous considerons cependant avoir approche l’exhaustivite des patients a selectionner par l’approche multi-sources. Nous avons utilise les sources d’information communes a l’ensemble des centres et les sources d’information additionnelles propres aux centres afin de maximiser la sensibilite globale de la demarche. Cette methodologie de selection menee avec l’ensemble des sources d’information disponibles, variable d’un centre a l’autre, dont plusieurs sont standardisees dans tous les centres, possede une sensibilite tres elevee mais implique de poursuivre le travail pour ameliorer la specificite des patients preselectionnes et minimiser le travail de validation du dossier.
- Published
- 2016
- Full Text
- View/download PDF
4. Organizing learning processes on risks by using the bow-tie representation
- Author
-
François-Régis Chevreau, D. Cauchois, Jean-Luc Wybo, Centre de recherche sur les Risques et les Crises (CRC), MINES ParisTech - École nationale supérieure des mines de Paris, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
- Subjects
Risk analysis ,Aramis method ,Safety Management ,Engineering ,Inservice Training ,Environmental Engineering ,Drug Industry ,Health, Toxicology and Mutagenesis ,0211 other engineering and technologies ,Guidelines as Topic ,02 engineering and technology ,Accident analysis ,Bow tie ,Risk Assessment ,Hazardous Substances ,Experience feedback process ,Order (exchange) ,0502 economics and business ,Accidents, Occupational ,Humans ,Environmental Chemistry ,Production (economics) ,European Union ,[SHS.GEST-RISQ]Humanities and Social Sciences/domain_shs.gest-risq ,Waste Management and Disposal ,Real-time operating system ,Simulation ,Bow-tie diagram ,050210 logistics & transportation ,021110 strategic, defence & security studies ,business.industry ,Decision Trees ,05 social sciences ,Representation (systemics) ,Organizational learning ,Environmental Exposure ,Problem-Based Learning ,Pollution ,Causality ,Risk analysis (engineering) ,Chemical Industry ,Models, Organizational ,Equipment Failure ,France ,business - Abstract
Outcome of the ARAMIS Project: Accidental Risk Assessment Methodology for Industries in the Framework of the SEVESO II Directive - New Stakes and Opportunities in the Control of Major Accident Hazards in Europe: Outputs from the ARAMIS Project; International audience; The Aramis method proposes a complete and efficient way to manage risk analysis by using the bow-tie representation. This paper shows how the bow-tie representation can also be appropriate for experience learning. It describes how a pharmaceutical production plant uses bow-ties for incident and accident analysis. Two levels of bow-ties are constructed: standard bow-ties concern generic risks of the plant whereas local bow-ties represent accident scenarios specific to each workplace. When incidents or accidents are analyzed, knowledge that is gained is added to existing local bow-ties. Regularly, local bow-ties that have been updated are compared to standard bow-ties in order to revise them. Knowledge on safety at the global and at local levels is hence as accurate as possible and memorized in a real time framework. As it relies on the communication between safety experts and local operators, this use of the bow-ties contributes therefore to organizational learning for safety.
- Published
- 2006
- Full Text
- View/download PDF
5. Choix d’un PACS : aspects pratiques de l’analyse des reponses a l’appel d’offre et choix d’un fournisseur
- Author
-
R. Buchaniec, Sophie Taïeb, S. Audebaud, P. Peugny, D. Cauchois, and L. Ceugnart
- Subjects
Radiological and Ultrasound Technology ,Radiology, Nuclear Medicine and imaging - Abstract
Objectifs Decrire la methodologie qui a permis de choisir le PACS du Centre Oscar Lambret (COL). Le groupe de travail comprenait : un radiologue, l’ingenieur biomedical, l’ingenieur informatique et un manipulateur radio futur coordonnateur de l’equipe d’administration du PACS. Materiels et methodes L’appel d’offre du PACS comportait 4 lots (cf. Appel d’offre d’un PACS: aspects pratiques de la redaction d’un cahier des charges fonctionnels-memes auteurs). Nous avons recu en septembre 2006 les reponses de 12 fournisseurs differents. Les criteres de choix retenus etaient: la completude de l’offre pour 8 points, la qualite du service apres vente pour 5 points, les couts (incluant la maintenance) pour 4 points et les possibilites d’evolution pour 3 points. Resultats A l’issue du premier tour d’analyse realise par le groupe de travail il restait 3 fournisseurs repondant aux criteres de choix retenus et valides positivement par les tests d’une vingtaine d’utilisateurs medecins non radiologues six mois auparavant. Ces 3 fournisseurs proposaient une solution immediatement operationnelle pour l’integration et la relecture des objets DICOM-RT part fondamentale de l’activite d’un centre de lutte contre le cancer. Une session de 2 jours pour chacun des 3 fournisseurs a ete faite comprenant: 3 demi-journees de test utilisateurs non radiologues (16) et radiologues (6) et une demi journee de debriefing du cahier des charges avec le groupe de travail. A l’issue de ce deuxieme tour il restait 2 fournisseurs: le groupe de travail a choisi le PACS qui satisfaisait 98 % des utilisateurs non radiologues (versus 80 % pour le second fournisseur). Conclusion Nous avons privilegie l’integration decentralisee des CD externes: integration plus souple et au fil de l’eau et le choix des utilisateurs non radiologues quant a l’ergonomie de la lecture des images.
- Published
- 2007
- Full Text
- View/download PDF
6. Matière organique dissoute dans l'estuaire maritime du St-Laurent — comparaison et choix des méthodes
- Author
-
D. Cauchois and M. Khalil
- Subjects
geography ,Oceanography ,geography.geographical_feature_category ,Dissolved organic carbon ,Environmental science ,Mineralogy ,Estuary - Abstract
To study quantitatively and qualitatively the dissolved organic matter in the St. Lawrence maritime estuary, a comparison between the different extracting methods has been undertaken. We propose two of them; one based on liquid–liquid extraction, the second by adsorption on a resin (Amberlite XAD-2). The methods used remove different spectrums of organic compounds from sea water. Hydrocarbons were found to constitute the major part of the extracts varying between 3 and 5 mg/liter. None of the methods used extracts more than 20% of the total organic matter present. The concentration of the organic chloro-compounds in the water column is in the order of 80 ppb.
- Published
- 1974
- Full Text
- View/download PDF
7. [What are the prognostic factors in operable breast cancer without histologic axillary lymph node invasiveness. Results of an uni- and multifactorial analysis]
- Author
-
B, Chevallier, F, Heintzman, V, Mosseri, Y, Graic, P, Bastit, J P, Dauce, P, Maisse, D, Cauchois, J P, Cabanne, and B, Asselain
- Subjects
Adult ,Aged, 80 and over ,Actuarial Analysis ,Risk Factors ,Humans ,Breast Neoplasms ,Female ,Adenocarcinoma ,Middle Aged ,Neoplasm Recurrence, Local ,Prognosis ,Aged ,Retrospective Studies - Abstract
Between October 1977 and December 1983, 379 consecutive patients have been treated for unilateral, non metastatic breast cancer, either with conservative (n = 205) or radical surgery (n = 174), with axillary dissection in all the cases. None of them had histologically proved lymph node involvement. Adjuvant radiotherapy was given in 268 cases. Estrogen receptor (ER) and progesterone receptor (PR) levels were measured on each tumor. Levels greater than 5 fmoles/mg cytosolic protein were considered as positive for both ER and PR. At 5 years, overall survival (OS) and disease-free survival (DFS) are respectively 88% and 79%. Unifactorial analysis using KAPLAN and MEIER estimates and Logrank test revealed that OS was significantly related to age, tumor size, histopathological grading, ER and PR. DFS was significantly related to the same factors. Menopausal status, number of intra mammary tumor foci, previous familial history of breast cancer were not significant. Multifactorial analysis revealed that DFS was significantly related to age (bad prognosis [bp]: less than or equal to 37 years old), tumor size, histopathological grading (bp: SBR = 3) and that OS was significantly related to tumor size and PR (bp: PR less than or equal to 5 fmoles/mg protein). A prognostic score was obtained which sampled our patients into 3 significantly different (P less than 0.0001) groups with high, intermediate and low risk of relapse. These results suggest that tumor size, histopathological grading and PR have their own prognostic weight in histologically node negative breast cancer. Grouping these factors together allows to define a high risk relapse group that could benefit from adjuvant treatment.
- Published
- 1989
8. Organizing learning processes on risks by using the bow-tie representation.
- Author
-
Chevreau FR, Wybo JL, and Cauchois D
- Subjects
- Causality, Chemical Industry organization & administration, Drug Industry, Environmental Exposure prevention & control, Equipment Failure, European Union, France, Guidelines as Topic, Hazardous Substances, Humans, Models, Organizational, Accidents, Occupational prevention & control, Chemical Industry standards, Decision Trees, Inservice Training, Problem-Based Learning, Risk Assessment methods, Safety Management methods
- Abstract
The Aramis method proposes a complete and efficient way to manage risk analysis by using the bow-tie representation. This paper shows how the bow-tie representation can also be appropriate for experience learning. It describes how a pharmaceutical production plant uses bow-ties for incident and accident analysis. Two levels of bow-ties are constructed: standard bow-ties concern generic risks of the plant whereas local bow-ties represent accident scenarios specific to each workplace. When incidents or accidents are analyzed, knowledge that is gained is added to existing local bow-ties. Regularly, local bow-ties that have been updated are compared to standard bow-ties in order to revise them. Knowledge on safety at the global and at local levels is hence as accurate as possible and memorized in a real time framework. As it relies on the communication between safety experts and local operators, this use of the bow-ties contributes therefore to organizational learning for safety.
- Published
- 2006
- Full Text
- View/download PDF
9. [What are the prognostic factors in operable breast cancer without histologic axillary lymph node invasiveness. Results of an uni- and multifactorial analysis].
- Author
-
Chevallier B, Heintzman F, Mosseri V, Graic Y, Bastit P, Dauce JP, Maisse P, Cauchois D, Cabanne JP, and Asselain B
- Subjects
- Actuarial Analysis, Adenocarcinoma pathology, Adult, Aged, Aged, 80 and over, Breast Neoplasms pathology, Female, Humans, Middle Aged, Neoplasm Recurrence, Local, Prognosis, Retrospective Studies, Risk Factors, Adenocarcinoma mortality, Breast Neoplasms mortality
- Abstract
Between October 1977 and December 1983, 379 consecutive patients have been treated for unilateral, non metastatic breast cancer, either with conservative (n = 205) or radical surgery (n = 174), with axillary dissection in all the cases. None of them had histologically proved lymph node involvement. Adjuvant radiotherapy was given in 268 cases. Estrogen receptor (ER) and progesterone receptor (PR) levels were measured on each tumor. Levels greater than 5 fmoles/mg cytosolic protein were considered as positive for both ER and PR. At 5 years, overall survival (OS) and disease-free survival (DFS) are respectively 88% and 79%. Unifactorial analysis using KAPLAN and MEIER estimates and Logrank test revealed that OS was significantly related to age, tumor size, histopathological grading, ER and PR. DFS was significantly related to the same factors. Menopausal status, number of intra mammary tumor foci, previous familial history of breast cancer were not significant. Multifactorial analysis revealed that DFS was significantly related to age (bad prognosis [bp]: less than or equal to 37 years old), tumor size, histopathological grading (bp: SBR = 3) and that OS was significantly related to tumor size and PR (bp: PR less than or equal to 5 fmoles/mg protein). A prognostic score was obtained which sampled our patients into 3 significantly different (P less than 0.0001) groups with high, intermediate and low risk of relapse. These results suggest that tumor size, histopathological grading and PR have their own prognostic weight in histologically node negative breast cancer. Grouping these factors together allows to define a high risk relapse group that could benefit from adjuvant treatment.
- Published
- 1989
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.