Piecewise Affine Dynamical Models of Petri Nets – Application to Emergency Call Centers*

We study timed Petri nets, with preselection and priority routing. We represent the behavior of these systems by piecewise affine dynamical systems. We use tools from the theory of nonexpansive mappings to analyze these systems. We establish an equivalence theorem between priority-free fluid timed Petri nets and semi-Markov decision processes, from which we derive the convergence to a periodic regime and the polynomial-time computability of the throughput. More generally, we develop an approach inspired by tropical geometry, characterizing the congestion phases as the cells of a polyhedral complex. We illustrate these results by a current application to the performance evaluation of emergency call centers in the Paris area. We show that priorities can lead to a paradoxical behavior: in certain regimes, the throughput of the most prioritary task may not be an increasing function of the resources.

SARS CoV-2 Genomic Characteristics and Clinical Impact of SARS-CoV-2 Viral Diversity in Critically Ill COVID-19 Patients: A Prospective Multicenter Cohort Study

Background: SARS-CoV-2 variant of concern (VOC) α spread worldwide, including in France, at the beginning of 2021. This variant was suggested to be associated with a higher risk of mortality than other variants. Little information is available in the subset of patients with severe disease admitted in the intensive care unit (ICU). We aimed to characterize the genetic diversity of SARS-CoV-2 variants isolated from patients with severe COVID-19 in order to unravel the relationships between specific viral mutations/mutational patterns and clinical outcomes.Methods: Prospective multicentre observational cohort study. Patients aged ≥18 years admitted in 11 ICUs from Great Paris area hospitals between October 1, 2020, and May 30, 2021 (before the introduction of VOC δ (B.617.2) in France) for acute respiratory failure (SpO2≤90% and need for supplemental oxygen or ventilator support) were included. SARS-CoV-2 infection, determined by RT-PCR testing. The primary clinical endpoint was day-28 mortality. Full-length SARS-CoV-2 genomes were sequenced by means of next-generation sequencing (Illumina COVIDSeq).Results: 413 patients were included, 183 (44.3%) had been infected with pre-existing variants, 197 (47.7%) with variant α (B.1.1.7), and 33 (8.0%) with other variants. Patients infected with pre-existing variants were significantly older (64.9±11.9 vs 60.5±11.8 years; p=0.0005); they had significantly more frequent COPD (11.5% (n=21/183) vs 4.1% (n=8/197); p=0.009), and higher SOFA score (4 [3-8] vs 3 [2-4]; 0.0002). Day-28 mortality was not different between patients infected with pre-existing, α (B.1.1.7) or other variants (31.1% (n=57/183) vs 26.2% (n=51/197) vs 30.3% (n=10/33), respectively; p=0.550). There was no association between day-28 mortality with a specific variant or the presence of specific mutations in SARS CoV-2 genome, including 17 mutations selected in the spike protein and all 1017 non-synonymous mutations detected throughout the entire viral genome.Conclusions: At ICU admission, patients infected with pre-existing variants had a different clinical presentation from those infected with variant α (B.1.1.7) and other variants later in the course of the pandemic, but mortality did not differ between these groups. There was no association between a specific variant or SARS CoV-2 genome mutational pattern and day-28 mortality.

Mobile phones, non-human agents at the service of assisted reproduction: monitoring and gendered dual allegiance

For sub-Saharan women enrolled in a protocol for assisted reproductive technology (ART), the use of mobile phones entails dual allegiance: toward the services of reproductive medicine and toward their transnational family. Indispensable for medically monitoring women’s reproductive bodies, the mobile phone enters the process for producing female gametes and contributes to the gender asymmetry typical of biomedicalized procreation. It is also used to maintain contacts with transnational family members who, from a distance, obtrude in the woman’s reproductive life. The use of mobile phones extends biomedical power over the woman’s body into her everyday life and the normative power of her transnational family into reproduction. Paradoxically, the mobile telephone allows collateral relatives to support the woman seeking reproduction assistance while also “hypermedicalizing” the woman’s daily life. Also paradoxically, this everyday companion is conductive to individual autonomy while also being used for new forms of surveillance and control. The data come from fieldwork conducted in the greater Paris area between 2011 and 2013 within a network of ART professionals and their patients.

Stroke without Cerebral Vasculopathy in Sickle Cell Disease Children: Which Causes? Which Treatment?

Sickle cell disease (SCD) is a severe hemoglobinopathy due to abnormal hemoglobin S (HbS). One of the most serious complications of SCD is cerebral vasculopathy (CV) leading to ischemic stroke in 8% of homozygous SS children without prevention strategy. Typical CV in SCD is a stenosis of one or more intracranial or cervical arteries, including internal carotid arteries, anterior cerebral arteries and middle cerebral arteries. However, several studies have reported the occurrence of stroke in SCD children in the absence of typical cerebral vasculopathy. The aim of the study is to investigate the prevalence of stroke without CV in SCD children, and to compare them to "classical" stroke (i.e in a context of typical CV) in terms of suspected etiologies, triggering events, treatment and risk of recurrence. In a large cohort of about 1500 SCD children living in Paris area, France, we recorded all the strokes occurring between 2007 and 2020, excluding venous thrombosis and PRESS Syndrome. These children were followed in a university hospital, and all benefited from early screening for cerebral vasculopathy and an adapted stroke prevention program. We considered as a "typical" stroke any new ischemic lesion of the cerebral parenchyma associated with an acute neurological syndrome occurring in the territory of a pre-existing and/or non-regressive stenosis. 25 strokes occurred during the study period. 12 of them (48%) did not met the definition of a "typical" stroke related to sickle cell CV. The children with "atypical" stroke were older (9 years old +/- 4.6, vs 6.5 years old +/- 4.3 in the typical group, p = 0.0086) and less frequently of SS genotype (33% non-SS vs 8% non-SS in the typical group). They had lower leukocyte count (11.3 G/L +/- 4.6, vs 15.7 G/L +/- 2.3 in the typical group, p=0. 04) and higher hemoglobin level at the time of the stroke (11 g/dl +/- 3 vs 7.4 g/dl +/- 1.3 in the typical group, p= 0.027). 17% of atypical strokes had posterior ischemic lesions, 33% had anterior lesions and 17% had multiple systematized lesions, in counting junctional lesions. We also found 33% of ischemic lesions of the cerebellum. Considering a potential trigger of the stroke, 58% of atypical events were hospitalized in an anesthesia or intensive care unit at the time of the stroke or less than 7 days before, compared to 8% of children with a "classic" stroke (p= 0.011). The etiologies adopted by clinicians and radiologists for the atypical stroke were reversible cerebral vasoconstriction syndrome (RCVS) (Figure 1), cerebral fat embolism, hyperviscosity and vasculitis in systemic inflammatory context. The evolution in the atypical group was more favorable in terms of recurrence (0% within 2 years vs 42 % in the typical group, p= 0.045), although only 33% (3 children) of atypical strokes were still under exchange transfusion program 24 months after the stroke vs 92% in the group of typical stroke. Overall, in a cohort of SCD children with efficient stroke prevention program, atypical strokes account for nearly half of all acute ischemic neurological events, related to hyperviscosity, RVCS or inflammatory vasculitis. Physicians must be aware of the potential triggers and of the context in which such events occur. Nevertheless, stroke without CV may not require long-term transfusion program contrary to stroke with CV, given the very low risk of recurrence we highlighted. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Effect of 2020 containment strategies on trauma workflow in Ile-de-France region: another benefit of lockdown?

Background During the SARS-CoV-2 pandemic, the French Government imposed various containment strategies, such as severe lockdown (SL) or moderate lockdown (ML). The aim of this study was to evaluate the effect of both strategies on severe trauma admissions and ICU capacity in Ile-de-France region (Paris Area). Main text We conducted a multicenter cohort-based observational study from 1stJanuary 2017 to 31th December 2020, including all consecutive trauma patients admitted to the trauma centers of Ile-de-France region participating in the national registry (Traumabase®). Two periods were defined, the “non-pandemic period” (NPP) from 2017 to 2019, and the “pandemic period” (PP) concerning those admitted in 2020. The number of ICU beds released during 2020 pandemic period (overall period, SL and ML) was estimated by multiplying difference in trauma admissions by the median length of stay during the same week of pandemic period (ICU day-beds in 2020). A 15% yearly reduction of trauma patients was observed during the PP, associated with the release of 6422 ICU day-beds in 2020. During SL and ML, the observed decrease in trauma admission was respectively 49 and 39% compared with similar dates of the NPP. The number of beds released was 1531 days-beds in SL and 679 day-beds in ML. Those reductions respectively accounted for 4.5 and 6.0% of the overall ICU admission for COVID-19 in Ile-de-France. Conclusion The lockdown strategies during pandemic resulted in a reduction of severe trauma admissions. In addition to the social distancing effect, lockdown strategies freed up an important number of ICU beds in trauma centers, available for severe COVID-19 patients.

Evaluating the Performance of Machine Learning Approaches to Predict the Microbial Quality of Surface Waters and to Optimize the Sampling Effort

Exposure to contaminated water during aquatic recreational activities can lead to gastrointestinal diseases. In order to decrease the exposure risk, the fecal indicator bacteria Escherichia coli is routinely monitored, which is time-consuming, labor-intensive, and costly. To assist the stakeholders in the daily management of bathing sites, models have been developed to predict the microbiological quality. However, model performances are highly dependent on the quality of the input data which are usually scarce. In our study, we proposed a conceptual framework for optimizing the selection of the most adapted model, and to enrich the training dataset. This frameword was successfully applied to the prediction of Escherichia coli concentrations in the Marne River (Paris Area, France). We compared the performance of six machine learning (ML)-based models: K-nearest neighbors, Decision Tree, Support Vector Machines, Bagging, Random Forest, and Adaptive boosting. Based on several statistical metrics, the Random Forest model presented the best accuracy compared to the other models. However, 53.2 ± 3.5% of the predicted E. coli densities were inaccurately estimated according to the mean absolute percentage error (MAPE). Four parameters (temperature, conductivity, 24 h cumulative rainfall of the previous day the sampling, and the river flow) were identified as key variables to be monitored for optimization of the ML model. The set of values to be optimized will feed an alert system for monitoring the microbiological quality of the water through combined strategy of in situ manual sampling and the deployment of a network of sensors. Based on these results, we propose a guideline for ML model selection and sampling optimization.