1
|
Excoffier JB, Salaün-Penquer N, Ortala M, Raphaël-Rousseau M, Chouaid C, Jung C. Analyse des patients hospitalisés pour COVID-19 lors du premier confinement de 2020 à l'aide de méthodes d'explicabilité. Rev Epidemiol Sante Publique 2022. [PMCID: PMC9634428 DOI: 10.1016/j.respe.2022.09.007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022] Open
Abstract
Introduction La pandémie de COVID-19 a rapidement mis une forte pression sur les centres hospitaliers et en particulier sur les services de réanimation [1]. Il y eut lors du premier confinement un besoin urgent d'outils permettant d'identifier les patients hospitalisés les plus à risque de subir une aggravation de leur état, ainsi qu'une meilleure compréhension de la typologie des patients COVID-19. Méthodes Les données contiennent des informations sur des patients hospitalisés au Centre Hospitalier Intercommunal de Créteil à cause de la COVID-19 lors de la première vague de l'épidémie (printemps 2020). Les variables explicatives disponibles sur les patients étaient l'âge, le sexe, plusieurs comorbidités et les résultats des examens radiologiques et biologiques. Un modèle d'ensemble d'arbres stimulé (« Boosted Tree Ensemble » [2, 3]) a été appliqué pour détecter si l'état du patient allait s'aggraver pendant l'hospitalisation. L'analyse des effets de chaque variable explicative ainsi que des effets d'interaction entre deux variables ont été effectuées en utilisant des méthodes d'explicabilité, domaine aussi appelée intelligence artificielle explicable [4]. Une stratification de la typologie des patients [5] a été réalisée en utilisant techniques de regroupement (clustering) et de sélection d'instances. Résultats Il y avait 409 patients, dont 176 (43 %) avaient subi une aggravation pendant leur séjour hospitalier. La précision globale (« accuracy ») du modèle prédictif était de 75 % pour le modèle de risque tandis que le score ROC AUC était de 81 %. Les variables explicatives les plus importantes étaient l'âge, la gravité du scanner thoracique et les variables biologiques telles que la CRP, la saturation en oxygène et les éosinophiles. Plusieurs variables ont montré de forts effets non linéaires, en particulier pour la sévérité du scanner, comme indiqué dans la Figure 1. Des effets d'interaction ont également été détectés entre l'âge et le sexe ainsi qu'entre l'âge et les éosinophiles. Trois principaux sous-groupes de patients ont été identifiés. Le patient le plus représentatif de chaque groupe est indiqué dans la Figure 2. Le premier groupe présentait un risque très faible d'aggravation de l'état de santé (pas de facteur de risque), le deuxième groupe présentait un risque plus élevé d'aggravation, mais toujours inférieur à 50 % (leur seul facteur de risque était un âge avancé), tandis que le troisième groupe avait le pronostic le plus défavorable (plusieurs facteurs de risque comprenant un âge avancé, plusieurs comorbidités, une sévérité CT élevée et des valeurs biologiques anormales). Discussion Les méthodes d'explicabilité ainsi que les techniques de regroupement et de sélection d'instances ont permis de mieux comprendre les effets des variables explicatives. Cela a aussi permis de déterminer les principales typologies des patients hospitalisés, facilitant ainsi la définition et l'amélioration des protocoles médicaux pour fournir les soins les plus appropriés à chaque profil [6]. Mots clés COVID19; Intelligence artificielle; Explicabilité; Clustering Déclaration de liens d'intérêts Les auteurs n'ont pas précisé leurs éventuels liens d'intérêts.
Collapse
Affiliation(s)
| | | | | | - M. Raphaël-Rousseau
- Centre hospitalier intercommunal de Créteil (CHIC) - Site web, Créteil, France
| | - C. Chouaid
- Centre hospitalier intercommunal de Créteil (CHIC) - Site web, Créteil, France
| | - C. Jung
- Centre hospitalier intercommunal de Créteil (CHIC) - Site web, Créteil, France
| |
Collapse
|
2
|
Jung C, Excoffier JB, Raphaël-Rousseau M, Salaun-Penquer N, Ortala M, Chouaid C. Evolution du profil des patients hospitalisés au cours des trois premières vagues de COVID-19 par des techniques d'apprentissage automatique. Rev Epidemiol Sante Publique 2022. [PMCID: PMC9634431 DOI: 10.1016/j.respe.2022.09.006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022] Open
Abstract
Introduction La COVID-19 a rapidement évolué d'une épidémie locale à une pandémie mondiale, obligeant la plupart des pays à prendre de fortes mesures pour contenir la propagation au cours des différentes vagues et soulager la pression sur les centres hospitaliers, en particulier sur les unités de soins intensifs. Une abondante littérature a détaillé les caractéristiques des patients et les facteurs de protection et de risque lors du l'éclatement de la pandémie [1]. Malheureusement, très peu d'études ont ensuite été menées pour décrire l'évolution de ces caractéristiques au cours des vagues ultérieures [2,3]. De plus, comme les premiers facteurs de risque identifiés étaient pluriels (de l'âge aux comorbidités, multiples interactions) les méthodes classiques d'analyse ne suffisent pas à obtenir une compréhension précise de la population à risque de développer des formes sévères de COVID-19. Méthodes Les données ont été recueillies prospectivement au Centre hospitalier intercommunal de Créteil sur plus d'un an, correspondant aux trois premières vagues de COVID-19 en France. Les caractéristiques disponibles étaient l'âge, le sexe et de nombreuses comorbidités. La variable cible indiquant si le patient avait développé une forme sévère (ventilation mécanique, réanimation, décès) de COVID-19 pendant son l'hospitalisation. L'évolution des caractéristiques entre les cas non sévères et sévères au fil des vagues a été analysée en couplant un modèle d'apprentissage automatique [4] à une méthode d'explicabilité produisant des influences locales [5]. Ainsi, chaque patient se voit associé un niveau de risque (une probabilité d'être un cas sévère) et un score de contribution de chacune de ses variables explicatives, permettant de repérer les facteurs de protection et de risque. Résultats Il y avait 1076 patients sur les trois vagues: 429 pour la première vague, 214 pour la deuxième et 433 pour la troisième. Les formes sévères concernaient respectivement 29 %, 31 % et 18 % de chaque vague. Les facteurs de risque de la première vague comprenaient l'âge avancé (≥70 ans), être un homme et des comorbidités telles que le diabète et l'obésité, tandis que les problèmes cardiovasculaires apparaissaient comme un léger facteur de protection. Il y avait de plus des effets d'interaction entre l'âge et les autres variables importantes. La deuxième vague présentait moins de facteurs de risque, puisque seuls l'âge avancé (≥70 ans) et le fait d'être un homme étaient des informations importantes. Lors de la troisième vague, l'âge avancé (≥70 ans) a également été identifié comme un facteur de risque mais de manière plus hétérogène que pour les vagues précédentes. Être un homme et les comorbidités telles que l'obésité, la grossesse ainsi que les problèmes cardiovasculaires et pulmonaires sont également apparus comme des facteurs de risque mais il n'y avait pas d'interaction avec l'âge. Discussion La typologie des patients hospitalisés atteints de formes sévères de COVID-19 a rapidement évolué au fil des vagues. L'analyse a notamment mis en évidence que les facteurs de risque étaient beaucoup plus hétérogènes pour la troisième vague. Cette évolution peut être due aux changements des pratiques hospitalières à mesure que la maladie était mieux comprise ainsi qu'à la campagne de vaccination [6] ciblant en premier lieu les personnes comme à haut risque telles les personnes âgées ou présentant des comorbidités. Mots clés COVID; 19; Intelligence artificielle; Explicabilité Déclaration de liens d'intérêts Les auteurs n'ont pas précisé leurs éventuels liens d'intérêts.
Collapse
Affiliation(s)
- C. Jung
- Centre de Recherche Clinique, CHI Créteil, France
| | | | | | | | | | - C. Chouaid
- Service de pneumologie, CHI Créteil, France,Inserm U955, UPEC, IMRB, Créteil, France
| |
Collapse
|
3
|
Tauzin M, Gouyon B, Hirt D, Carbajal R, Gouyon JB, Brunet AC, Ortala M, Goro S, Jung C, Durrmeyer X. Frequencies, Modalities, Doses and Duration of Computerized Prescriptions for Sedative, Analgesic, Anesthetic and Paralytic Drugs in Neonates Requiring Intensive Care: A Prospective Pharmacoepidemiologic Cohort Study in 30 French NICUs From 2014 to 2020. Front Pharmacol 2022; 13:939869. [PMID: 35924063 PMCID: PMC9341520 DOI: 10.3389/fphar.2022.939869] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/09/2022] [Accepted: 06/21/2022] [Indexed: 11/13/2022] Open
Abstract
Objectives: No consensus exists about the doses of analgesics, sedatives, anesthetics, and paralytics used in critically ill neonates. Large-scale, detailed pharmacoepidemiologic studies of prescription practices are a prerequisite to future research. This study aimed to describe the detailed prescriptions of these drug classes in neonates hospitalized in neonatal intensive care units (NICU) from computerized prescription records and to compare prescriptions by gestational age. Materials and Methods: We included all neonates requiring intensive care in 30 French level III units from 2014 through 2020 with a computerized prescription for an analgesic, sedative, anesthetic, or paralytic agent. We described frequencies of prescription, methods of administration, concomitant drug prescriptions, and dosing regimen, and compared them across gestational ages. Results: Among 65,555 neonates, 29,340 (44.8%) were prescribed at least one analgesic (acetaminophen in 37.2% and opioids in 17.8%), sedative (9.8%), anesthetic (8.5%), and/or paralytic agent (1%). Among preterm infants born before 28 weeks, 3,771/4,283 (88.0%) were prescribed at least one of these agents: 69.7% opioids, 41.2% sedatives, 32.5% anesthetics, and 5.8% paralytics. The most frequently prescribed agents were sufentanil (in 10.3% of neonates) and morphine (in 8.0% of neonates) for opioids, midazolam (9.3%) for sedatives, ketamine (5.7%) and propofol (3.3%) for anesthetics. In most neonates, opioids and sedatives were prescribed as continuous infusion, whereas anesthetics were prescribed as single doses. Opioids, sedatives and paralytics were mostly prescribed in association with another agent. Doses varied significantly by gestational age but within a limited range. Gestational age was inversely related to the frequency, cumulative dose and duration of prescriptions. For example, morphine prescriptions showed median (IQR) cumulative doses of 2601 (848–6750) vs. 934 (434–2679) µg/kg and median (IQR) durations of 7 (3–15) vs. 3 (2–5) days in infants born <28 vs. ≥ 37 weeks of gestation, respectively (p-value<0.001). Conclusion: The prescriptions of analgesic, sedative, anesthetic, or paralytic agent were frequent and often combined in the NICU. Lower gestational age was associated with higher frequencies, longer durations and higher cumulative doses of these prescriptions. Dose-finding studies to determine individualized dosing regimens and studies on long-term neurodevelopmental outcome according to received cumulative doses are required.
Collapse
Affiliation(s)
- Manon Tauzin
- Neonatal Intensive Care Unit, CHI Créteil, Créteil, France
- *Correspondence: Manon Tauzin,
| | - Béatrice Gouyon
- Centre d’Etudes Périnatales de L’Océan Indien (CEPOI, EA7388), Université de La Réunion, Saint Pierre, France
| | - Déborah Hirt
- Pharmacology Department, Hôpital Cochin APHP, Paris, France
| | - Ricardo Carbajal
- Pediatric Emergency Department, Assistance Publique-Hôpitaux de Paris, Hôpital Armand Trousseau- Sorbonne Université, Paris, France
- Institut National de La Santé et de La Recherche Médicale UMR1153, Paris, France
| | - Jean-Bernard Gouyon
- Centre d’Etudes Périnatales de L’Océan Indien (CEPOI, EA7388), Université de La Réunion, Saint Pierre, France
| | | | | | - Seydou Goro
- Clinical Research Center, CHI Créteil, Créteil, France
| | - Camille Jung
- Clinical Research Center, CHI Créteil, Créteil, France
| | - Xavier Durrmeyer
- Neonatal Intensive Care Unit, CHI Créteil, Créteil, France
- Faculté de Médecine de Créteil, IMRB, GRC CARMAS, Université Paris Est Créteil, Créteil, France
| |
Collapse
|
4
|
Excoffier JB, Salaün-Penquer N, Ortala M, Raphaël-Rousseau M, Chouaid C, Jung C. Analysis of COVID-19 inpatients in France during first lockdown of 2020 using explainability methods. Med Biol Eng Comput 2022; 60:1647-1658. [PMID: 35426076 PMCID: PMC9009979 DOI: 10.1007/s11517-022-02540-0] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/25/2021] [Accepted: 02/23/2022] [Indexed: 11/24/2022]
Abstract
The COVID-19 pandemic rapidly puts a heavy pressure on hospital centers, especially on intensive care units. There was an urgent need for tools to understand typology of COVID-19 patients and identify those most at risk of aggravation during their hospital stay. Data included more than 400 patients hospitalized due to COVID-19 during the first wave in France (spring of 2020) with clinical and biological features. Machine learning and explainability methods were used to construct an aggravation risk score and analyzed feature effects. The model had a robust AUC ROC Score of 81%. Most important features were age, chest CT Severity and biological variables such as CRP, O2 Saturation and Eosinophils. Several features showed strong non-linear effects, especially for CT Severity. Interaction effects were also detected between age and gender as well as age and Eosinophils. Clustering techniques stratified inpatients in three main subgroups (low aggravation risk with no risk factor, medium risk due to their high age, and high risk mainly due to high CT Severity and abnormal biological values). This in-depth analysis determined significantly distinct typologies of inpatients, which facilitated definition of medical protocols to deliver the most appropriate cares for each profile. Graphical abstract represents main methods used and results found with a focus on feature impact on aggravation risk and identified groups of patients ![]()
Collapse
|
5
|
Excoffier JB, Escriva E, Aligon J, Ortala M. Local Explanation-Based Method for Healthcare Risk Stratification. Stud Health Technol Inform 2022; 294:555-556. [PMID: 35612141 DOI: 10.3233/shti220520] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 06/15/2023]
Abstract
Decision support tools in healthcare require a strong confidence in the developed Machine Learning (ML) models both in terms of performances and in their ability to provide users a deeper understanding of the underlying situation. This study presents a novel method to construct a risk stratification based on ML and local explanations. An open-source dataset was used to demonstrate the efficiency of this method that well identified the main subgroups of patients. Therefore, this method could help practitioners adjust and build protocols to improve care deliveries that would better reflect patient's risk level and profile.
Collapse
Affiliation(s)
| | - Elodie Escriva
- Kaduceo, Toulouse, France
- Université de Toulouse-Capitole, IRIT, (CNRS/UMR 5505), Toulouse, France
| | - Julien Aligon
- Université de Toulouse-Capitole, IRIT, (CNRS/UMR 5505), Toulouse, France
| | | |
Collapse
|
6
|
Lazzati A, Epaud S, Ortala M, Katsahian S, Lanoy E. Effect of bariatric surgery on cancer risk: results from an emulated target trial using population-based data. Br J Surg 2022; 109:433-438. [PMID: 35136932 DOI: 10.1093/bjs/znac003] [Citation(s) in RCA: 19] [Impact Index Per Article: 9.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/10/2021] [Revised: 12/08/2021] [Accepted: 01/04/2021] [Indexed: 12/31/2022]
Abstract
BACKGROUND The impact of weight loss induced by bariatric surgery on cancer occurrence is controversial. To study the causal effect of bariatric surgery on cancer risk from an observational database, a target-trial emulation technique was used to mimic an RCT. METHODS Data on patients admitted between 2010 and 2019 with a diagnosis of obesity were extracted from a national hospital discharge database. Criteria for inclusion included eligibility criteria for bariatric surgery and the absence of cancer in the 2 years following inclusion. The intervention arms were bariatric surgery versus no surgery. Outcomes were the occurrence of any cancer and obesity-related cancer; cancers not related to obesity were used as negative controls. RESULTS A total of 1 140 347 patients eligible for bariatric surgery were included in the study. Some 288 604 patients (25.3 per cent) underwent bariatric surgery. A total of 48 411 cancers were identified, including 4483 in surgical patients and 43 928 among patients who did not receive bariatric surgery. Bariatric surgery was associated with a decrease in the risk of obesity-related cancer (hazard ratio (HR) 0.89, 95 per cent c.i. 0.83 to 0.95), whereas no significant effect of surgery was identified with regard to cancers not related to obesity (HR 0.96, 0.91 to 1.01). CONCLUSION When emulating a target trial from observational data, a reduction of 11 per cent in obesity-related cancer was found after bariatric surgery.
Collapse
Affiliation(s)
- Andrea Lazzati
- Department of General Surgery, Centre Hospitalier Intercommunal de Créteil, Créteil, France
- INSERM IMRB U955, Université Paris-Est Créteil, Créteil, France
| | | | | | - Sandrine Katsahian
- Assistance Publique Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Unité d'Épidémiologie et de Recherche Clinique, INSERM, Centre d'Investigation Clinique 1418, Module Épidémiologie Clinique, HEGP, Paris, France
- Université de Paris, Paris, France
- INSERM, UMRS 1138, Centre de Recherche des Cordeliers, Paris, France
| | | |
Collapse
|
7
|
Jung C, Excoffier JB, Raphaël-Rousseau M, Salaün-Penquer N, Ortala M, Chouaid C. Evolution of hospitalized patient characteristics through the first three COVID-19 waves in Paris area using machine learning analysis. PLoS One 2022; 17:e0263266. [PMID: 35192649 PMCID: PMC8863256 DOI: 10.1371/journal.pone.0263266] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/30/2021] [Accepted: 01/14/2022] [Indexed: 12/16/2022] Open
Abstract
Characteristics of patients at risk of developing severe forms of COVID-19 disease have been widely described, but very few studies describe their evolution through the following waves. Data was collected retrospectively from a prospectively maintained database from a University Hospital in Paris area, over a year corresponding to the first three waves of COVID-19 in France. Evolution of patient characteristics between non-severe and severe cases through the waves was analyzed with a classical multivariate logistic regression along with a complementary Machine-Learning-based analysis using explainability methods. On 1076 hospitalized patients, severe forms concerned 29% (123/429), 31% (66/214) and 18% (79/433) of each wave. Risk factors of the first wave included old age (≥ 70 years), male gender, diabetes and obesity while cardiovascular issues appeared to be a protective factor. Influence of age, gender and comorbidities on the occurrence of severe COVID-19 was less marked in the 3rd wave compared to the first 2, and the interactions between age and comorbidities less important. Typology of hospitalized patients with severe forms evolved rapidly through the waves. This evolution may be due to the changes of hospital practices and the early vaccination campaign targeting the people at high risk such as elderly and patients with comorbidities.
Collapse
Affiliation(s)
- Camille Jung
- Clinical Research Center, CHI Créteil, Créteil, France
| | | | | | | | | | - Christos Chouaid
- Department of pneumology, CHI Créteil, Créteil, France
- Inserm U955, UPEC, IMRB, Créteil, France
| |
Collapse
|
8
|
Rousseau M, Excoffier JB, Salaun-Penquer N, Ortala M, Chouaid C, Jung C. Evolution des facteurs de risque de forme grave de la COVID-19 à travers les trois vagues à partir des données du PMSI. Rev Epidemiol Sante Publique 2022. [PMCID: PMC8907805 DOI: 10.1016/j.respe.2022.01.088] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022] Open
Abstract
Introduction De nombreuses études ont établi une typologie des patients à risque de développer une forme grave de COVID-19. La grande majorité de ces études ne portent que sur la première vague de la COVID-19 alors que les profils à risque de développer des formes graves évoluent après les campagnes de vaccination. Cette étude a pour but d'analyser l’évolution des facteurs de risque cliniques de COVID-19 graves lors des différentes vagues afin d'ajuster l'offre de soins au contexte actuel. Méthodes Les données du PMSI du Centre hospitalier intercommunal de Créteil ont été analysées pendant les trois premières vagues de l'épidémie. Les variables cliniques disponibles étaient l’âge, le sexe et les comorbidités connues comme à risque de forme grave. Les formes graves étaient définies par le passage en réanimation, le recours à un support ventilatoire non invasif ou le décès du patient. L'évolution de la typologie des patients à risque a été analysée avec des méthodes classiques comme la régression multivariée ainsi qu'avec des techniques provenant de l'apprentissage automatique et de son sous-domaine qu'est l'explicabilité. Résultats Sur 1076 patients hospitalisés, les formes sévères concernaient 29 % (123/429) des patients de la vague 1, 31 % (66/214) de la vague 2 et 18 % (79/433) de la vague 3. Les facteurs de risque de la vague 1 étaient l'âge élevé ( ≥ 70 ans), le sexe masculin, le diabète et l'obésité, tandis que les problèmes cardiovasculaires apparaissaient comme des facteurs protecteurs. Les impacts de l'âge, du sexe étaient moins marqués pour la vague 3, de même que l'interaction entre l'âge et les comorbidités. Discussion/Conclusion Le profil des patients à risque a rapidement évolué au cours des vagues, la troisième vague ayant eu un écart bien moins net entre les formes sévères et non-sévères. Cette évolution peut provenir des changements des procédures médicales hospitalières ainsi que du début de la campagne de vaccination ciblant en premier lieu les personnes avec un risque élevé telles les personnes âgées ou présentant certaines comorbidités.
Collapse
|
9
|
Abitbol E, Miere A, Excoffier JB, Mehanna CJ, Amoroso F, Kerr S, Ortala M, Souied EH. Deep learning-based classification of retinal vascular diseases using ultra-widefield colour fundus photographs. BMJ Open Ophthalmol 2022; 7:e000924. [PMID: 35141420 PMCID: PMC8819815 DOI: 10.1136/bmjophth-2021-000924] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/27/2021] [Accepted: 01/18/2022] [Indexed: 01/01/2023] Open
Abstract
Objective To assess the ability of a deep learning model to distinguish between diabetic retinopathy (DR), sickle cell retinopathy (SCR), retinal vein occlusions (RVOs) and healthy eyes using ultra-widefield colour fundus photography (UWF-CFP). Methods and Analysis In this retrospective study, UWF-CFP images of patients with retinal vascular disease (DR, RVO, and SCR) and healthy controls were included. The images were used to train a multilayer deep convolutional neural network to differentiate on UWF-CFP between different vascular diseases and healthy controls. A total of 224 UWF-CFP images were included, of which 169 images were of retinal vascular diseases and 55 were healthy controls. A cross-validation technique was used to ensure that every image from the dataset was tested once. Established augmentation techniques were applied to enhance performances, along with an Adam optimiser for training. The visualisation method was integrated gradient visualisation. Results The best performance of the model was obtained using 10 epochs, with an overall accuracy of 88.4%. For DR, the area under the receiver operating characteristics (ROC) curve (AUC) was 90.5% and the accuracy was 85.2%. For RVO, the AUC was 91.2% and the accuracy 88.4%. For SCR, the AUC was 96.7% and the accuracy 93.8%. For healthy controls, the ROC was 88.5% with an accuracy that reached 86.2%. Conclusion Deep learning algorithms can classify several retinal vascular diseases on UWF-CPF with good accuracy. This technology may be a useful tool for telemedicine and areas with a shortage of ophthalmic care.
Collapse
Affiliation(s)
- Elie Abitbol
- Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Creteil, France
| | - Alexandra Miere
- Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Creteil, France
| | | | - Carl-Joe Mehanna
- Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Creteil, France
| | - Francesca Amoroso
- Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Creteil, France
| | | | | | - Eric H Souied
- Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Creteil, France
| |
Collapse
|
10
|
Epaud S, Epaud R, Salaün-Penquer N, Belozertseva E, Remus N, Douvry B, Bequignon E, Coste A, Prulière-Escabasse V, Schlemmer F, Jung C, Ortala M, Maitre B, Delestrain C. Impact of a rare respiratory diseases reference centre set-up on primary ciliary dyskinesia care pathway. Eur Respir J 2021; 59:13993003.02413-2021. [PMID: 34711540 DOI: 10.1183/13993003.02413-2021] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/04/2021] [Accepted: 10/14/2021] [Indexed: 11/05/2022]
Affiliation(s)
- Salome Epaud
- Kaduceo SAS, Toulouse, France.,equal contributors
| | - Ralph Epaud
- Centre Hospitalier Intercommunal de Créteil, Service de Pédiatrie Générale, Créteil, , France .,University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,FHU SENEC, Créteil, France.,equal contributors
| | | | - Ekaterina Belozertseva
- Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,Clinical Research Centre, Intercommunal Hospital of Créteil, Créteil, France
| | - Natascha Remus
- Centre Hospitalier Intercommunal de Créteil, Service de Pédiatrie Générale, Créteil, , France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France
| | - Benoit Douvry
- Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service de Pneumologie, Créteil, , France
| | - Emilie Bequignon
- Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service d'ORL, Créteil, , France
| | - Andre Coste
- University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service d'ORL, Créteil, , France
| | - Virginie Prulière-Escabasse
- University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service de Pneumologie, Créteil, , France
| | - Frédéric Schlemmer
- University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,FHU SENEC, Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service de Pneumologie, Créteil, , France
| | - Camille Jung
- Clinical Research Centre, Intercommunal Hospital of Créteil, Créteil, France
| | | | - Bernard Maitre
- University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,FHU SENEC, Créteil, France.,Centre Hospitalier Intercommunal de Créteil, Service de Pneumologie, Créteil, , France.,equal contributors
| | - Céline Delestrain
- Centre Hospitalier Intercommunal de Créteil, Service de Pédiatrie Générale, Créteil, , France .,University Paris Est Créteil, INSERM, IMRB, Créteil, France.,Centre des Maladies Respiratoires Rares (RESPIRARE®), Créteil, France.,FHU SENEC, Créteil, France
| |
Collapse
|
11
|
Lazzati A, Salaün-Penquer N, Ortala M, Vignot M, De Filippo G, Jung C. Trends in metabolic bariatric surgery in adolescents in France: a nationwide analysis on an 11- year period. Surg Obes Relat Dis 2021. [DOI: 10.1016/j.soard.2021.05.027
expr 953237874 + 872256771] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/16/2023]
|
12
|
Lazzati A, Raphael Rousseau M, Bartier S, Dabi Y, Challine A, Haddad B, Herta N, Souied E, Ortala M, Epaud S, Masson M, Salaün-Penquer N, Coste A, Jung C. Impact of COVID-19 on surgical emergencies: nationwide analysis. BJS Open 2021; 5:6280342. [PMID: 34021327 PMCID: PMC8140197 DOI: 10.1093/bjsopen/zrab039] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/04/2021] [Accepted: 03/16/2021] [Indexed: 12/21/2022] Open
Abstract
Background The COVID-19 pandemic has had a major impact on healthcare in many countries. This study assessed the effect of a nationwide lockdown in France on admissions for acute surgical conditions and the subsequent impact on postoperative mortality. Methods This was an observational analytical study, evaluating data from a national discharge database that collected all discharge reports from any hospital in France. All adult patients admitted through the emergency department and requiring a surgical treatment between 17 March and 11 May 2020, and the equivalent period in 2019 were included. The primary outcome was the change in number of hospital admissions for acute surgical conditions. Mortality was assessed in the matched population, and stratified by region. Results During the lockdown period, 57 589 consecutive patients were admitted for acute surgical conditions, representing a decrease of 20.9 per cent compared with the 2019 cohort. Significant differences between regions were observed: the decrease was 15.6, 17.2, and 26.8 per cent for low-, intermediate- and high-prevalence regions respectively. The mortality rate was 1.92 per cent during the lockdown period and 1.81 per cent in 2019. In high-prevalence zones, mortality was significantly increased (odds ratio 1.22, 95 per cent c.i. 1.06 to 1.40). Conclusion A marked decrease in hospital admissions for surgical emergencies was observed during the lockdown period, with increased mortality in regions with a higher prevalence of COVID-19 infection. Health authorities should use these findings to preserve quality of care and deliver appropriate messages to the population.
Collapse
Affiliation(s)
- A Lazzati
- Department of General and Digestive Surgery, Intercommunal Hospital of Créteil, Créteil, France.,INSERM U955, IMRB, Créteil, France
| | - M Raphael Rousseau
- Department of Medical Informatics, Intercommunal Hospital of Créteil, Créteil, France
| | - S Bartier
- INSERM U955, IMRB, Créteil, France.,University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Oto-rhino-laryngology Head and Neck Surgery, Intercommunal Hospital of Créteil, Créteil, France.,Department of Oto-rhino-laryngology Head and Neck Surgery, Paris Public Hospitals, Henri Mondor Hospital, France.,CNRS, ERL 7240, Créteil, France
| | - Y Dabi
- University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Obstetrics and Gynaecology, Intercommunal Hospital of Créteil, Créteil, France
| | - A Challine
- Department of Digestive, Hepatobiliary and Pancreatic Surgery, AP-HP, Université de Paris, Cochin Hospital, France
| | - B Haddad
- University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Obstetrics and Gynaecology, Intercommunal Hospital of Créteil, Créteil, France
| | - N Herta
- University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Ophthalmology, Intercommunal Hospital of Créteil, Créteil, France
| | - E Souied
- University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Ophthalmology, Intercommunal Hospital of Créteil, Créteil, France
| | | | - S Epaud
- Kaduceo SAS, Toulouse, France
| | | | | | - A Coste
- INSERM U955, IMRB, Créteil, France.,University Paris-Est Creteil, School of Medicine, Créteil, France.,Department of Oto-rhino-laryngology Head and Neck Surgery, Intercommunal Hospital of Créteil, Créteil, France.,Department of Oto-rhino-laryngology Head and Neck Surgery, Paris Public Hospitals, Henri Mondor Hospital, France.,CNRS, ERL 7240, Créteil, France
| | - C Jung
- Clinical Research Centre, Intercommunal Hospital of Créteil, Créteil, France
| |
Collapse
|
13
|
Lazzati A, Salaün-Penquer N, Ortala M, Vignot M, De Filippo G, Jung C. Trends in metabolic bariatric surgery in adolescents in France: a nationwide analysis on an 11- year period. Surg Obes Relat Dis 2021; 17:1566-1575. [PMID: 34144914 DOI: 10.1016/j.soard.2021.05.027] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/31/2020] [Revised: 04/29/2021] [Accepted: 05/14/2021] [Indexed: 12/24/2022]
Abstract
BACKGROUND As the prevalence of obesity in adolescents has reached an alarming level of 16%, the rate of metabolic bariatric surgery (MBS) in this population is also rising in several countries. OBJECTIVES This study aimed to compare the trends in types of MBS, short-term safety, and revisional rates, in younger adolescents aged < 18 years, compared with older adolescents (aged 18-19 yr) and adults aged >20 years. SETTING Clinical research center, general hospital in France. METHODS Using a national administrative database (Programme de Médicalisation des Systèmes d'Information [PMSI]), data regarding all patients undergoing MBS between 2008 and 2018 in France were examined. Demographic parameters, body mass index (BMI), co-morbidities, types of surgery, early complications, and long-term revisional rates were analyzed, comparing younger adolescents (<18 yr), older adolescents (18-19 yr), and adults (≥20 yr). RESULTS The number of bariatric procedures in adolescents initially increased from 59 in 2008 to 135 in 2014, and then progressively declined to 56 procedures in 2018. Adjustable gastric banding (AGB) decreased from 83.1% (n = 49) of procedures to 32.1% (n = 18) of procedures during the study period, while sleeve gastrectomy (SG) increased from 6.8% (n = 4) to 46.4% (n = 26). In the early postoperative period, younger adolescents undergoing MBS experienced fewer episodes of reoperation (1.0% versus 1.3% in older adolescents and 2.6% in adults, P < .001) and intensive care unit (ICU) stays (.2% versus .2% in older adolescents and .6% in adults, P < .001), and no deaths were observed in younger adolescents (.02% in older adolescents and .1% in adults, P = .18). At 10 years, the AGB removal rate was lower in younger adolescents (24.8%) compared with that in older adolescents (29.6%) and adults (50.3%, P < .001). Similarly, rates of revisional surgery after SG were different in the 3 groups: 2.9%, 4.6% and 12.2% in younger adolescents, older adolescents, and adults, respectively. CONCLUSION Despite significantly lower early complication rates and long-term revisional rates in young adolescents (<18 yr), we observed a progressive decrease in the utilization of MBS in this population in France, compared with adults (≥20 yr) and older adolescents (18-19 yr).
Collapse
Affiliation(s)
- Andrea Lazzati
- Department of Digestive Surgery, Intercommunal Hospital of Créteil, Créteil, France.
| | | | | | - Marina Vignot
- Clinical Research Center, Intercommunal Hospital of Créteil, Créteil, France
| | - Gianpaolo De Filippo
- Assistance Publique-Hôpitaux de Paris, Hôpital Robert-Debré, Service d'Endocrinologie et Diabétologie Pédiatrique, Paris, France; French Clinical Research Group in Adolescent Medicine and Health, Paris, France
| | - Camille Jung
- Clinical Research Center, Intercommunal Hospital of Créteil, Créteil, France; Department of Pediatrics, Intercommunal Hospital of Créteil, Créteil, France
| |
Collapse
|
14
|
Paolino L, Pravettoni R, Epaud S, Ortala M, Lazzati A. Comparison of Surgical Activity and Scientific Publications in Bariatric Surgery: an Epidemiological and Bibliometric Analysis. Obes Surg 2020; 30:3822-3830. [DOI: 10.1007/s11695-020-04703-0] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/14/2022]
|