Bioprojet (France)

International audience

This chapter provides a Management Model for Artifitial Intelligence (MMAI) that, through its application and analysis, enables corporate decision-makers, academics, consultants, and others interested in the topic to seek the best alternatives for AI management, helping them streamline their organizations and achieve the highest standards of profitability, competitiveness, and sustainability. The MMAI is composed of nine key factors: 1) Clear objectives, 2) Ethics and regulation, 3) Data governance, 4) Transparency and explainability, 5) Performance and scalability, 6) Continuous training, 7) Risk management, 8) Integration and compatibility, and 9) Stakeholder engagement. Each factor, in turn, contains a series of variables that explain it. In total, there are twenty-seven (27) variables explained in detail in the theoretical framework of this chapter, which measure four levels ranging from 0 to 100%.

Refined probabilistic models have been proposed for steel resistance, dead loads, live loads, snow loads, and wind loads, which can be used in reliability analyses by the FORM and SORM methods. The theoretical distributions were selected using the Anderson–Darling goodness-of-fit criterion. It was found that steel resistance is most accurately represented by a lognormal distribution. The analysis was carried out on the basis of extensive experimental data on the yield strength of steel grade S235. It is shown that the use of the Generalized Extreme Value (GEV) distribution for snow and wind loads provides the best agreement with observed annual maxima obtained from 50-year meteorological records. The results improve the accuracy of probabilistic design calculations and can be applied in the calibration of code-specified (baseline) load and resistance values.

The purpose of the work is to study the problem of developing an information model of bridge structures in the design process in the field of road construction. The use of information modeling technology in the field of road management allows for continuous monitoring of a construction facility at all stages of its life cycle. For the successful implementation of the information model design, it is necessary to organize all the processes of interaction between the design participants. It isrequired to create a specialized knowledge library in the form of a catalog with standard solutions that can be used in future projects, with the possibility of changing component parameters. The article discusses the technology of creating an information model of a bridge structure using the Renga software package. The scenario of the organization of design based on information modeling technology is shown. In this paper, the dependencies of the Renga software package tools with the elements of the bridge structure were selected. A 3D model of the bridge is presented with information on each element, which makes it possible to create a spatial model of an engineering structure. It is concluded that the use of the information model greatly simplifies the work with the projected object, providing a number of advantages over traditional methods of designing bridge structures. This approach to design increases efficiency, reduces the like lihood of errors,and promotes more effective collaboration between project participants. As a result, the information model provides a more holistic approach to design, contributing to the creation of a highquality and completed project.

<bold>Objective:</bold> Obstructive sleep apnea (OSA) is treated with continuous positive airway pressure (CPAP). Some patients have residual excessive daytime sleepiness (EDS) under CPAP. Pitolisant reduce significantly EDS. A meta-analysis assessed efficacy and safety of pitolisant (P) vs placebo (pl) to treat EDS in OSA patients using CPAP. <bold>Method:</bold> An individual patient meta-analytical approach was used to analyse to CPAP-treated patients in pivotal Pitolisant trials; a 2-level (study-patient) hierarchical model assuming a random treatment effect was used. Epworth Sleepiness Scale (ESS) and Oxford Sleep Resistance test (OSLER) were co-primary endpoints. Secondary endpoints included EDS-Z scores, fatigue, clinical global impression, and quality of life (QoL). <bold>Results:</bold> 3 randomized controlled trials were identified. Individual patient data came from 423 CPAP-treated patients (pl n=120, P20mg n=183, P40mg n=120). Treatment effects on ESS were -3.20 (95% CI: -4.37, -2.00; P<0.001) and ‑3.57 (95% CI: -4.87, -2.80); P<0.001) for 20mg and 40mg doses. Treatment effects (minutes) for OSLER were 1.24 (95% CI: 0.60, 1.10); P=0.001) and 1.21 (95% CI: 0.06, 1.38); P=0.006). Efficacy was homogeneous across subgroups; P40mg was superior to P20mg for older age (≥50 years) and higher baseline apnea-hypopnea index values (≥15). No significant differences were observed between the two doses across various safety parameters. <bold>Conclusion:</bold> Pitolisant 20 mg and 40 mg were significantly therapeutically superior to placebo in treating residual EDS in patients with OSA who received CPAP on the outcomes for ESS, OSLER, and QoL.

<bold>Introduction</bold> Obstructive sleep apnoea (OSA) is a chronic respiratory disease associated with sleepiness and reduced quality of life. Despite continuous positive airway pressure (CPAP), many patients experience residual EDS. Pharmacological treatments authorised in Europe and/or the USA are modafinil/armodafinil(M), solriamfetol(S), and pitolisant(P). In the absence of head-to-head trials, the relative efficacy and safety of these agents is largely unknown. <bold>Method</bold> Randomised controlled trials (RCTs) that compared the efficacy and safety of authorised medications were analysed using network meta-analysis. The primary efficacy endpoint combined Epworth Sleepiness Scale (ESS) and OSLER/MWT Z-scores. Quality of life (QoL), overall and cardiovascular safety, and benefit-risk ratios were calculated. <bold>Results</bold> 20 RCTs with 4015 subjects were included. Analysis of combined subjective (ESS) and objective (OSLER/MWT) efficacy outcome Z-scores showed that S: effect size [ES]=0.66[95% CI: 0.36, 0.96], P: ES=0.66[95% CI: 0.44, 0.88], M200mg: ES=0.53[95% CI: 0.33, 0.73]; M400mg: ES=0.53[95% CI: 0.42, 0.64] had a clinically meaningful improvement in efficacy. P-scores ranked placebo, then P, M200mg, M400mg and S for overall safety; and P, then S, M400mg and M200mg for benefit-risk ratio. <bold>Conclusion</bold> Pitolisant, solriamfetol and modafinil had comparable efficacy for maintaining wakefulness in OSA patients. Pitolisant had a better safety profile and benefit-risk ratio compared with solriamfetol and modafinil. The overall and cardiovascular safety risk ratios suggest that pitolisant might be the best candidate for OSA patients with cardiovascular comorbidities.

SummaryObstructive sleep apnea (OSA) syndrome commonly leads to excessive daytime sleepiness (EDS). Pitolisant, a selective histamine‐3 receptor antagonist, is efficacious at doses up to 20 mg once daily in OSA treated or not with continuous positive airway pressure (CPAP). We assessed the efficacy and safety of pitolisant at doses up to 40 mg once daily in patients with moderate to severe OSA treated or not with CPAP therapy. In this phase 3, multicentre, randomised, double‐blind, placebo‐controlled clinical trial, patients with OSA were assigned 2:1 to receive pitolisant (according to an individual up‐titration scheme, 10, 20 or 40 mg once daily) or placebo for 12 weeks. The primary endpoint was a change in the Epworth Sleepiness Scale (ESS) score from baseline to week 12. Secondary endpoints included a change in reaction time using the Oxford Sleep Resistance test (OSleR), Clinical Global Impression of Change (CGI‐C), and Patient's Global Opinion of the Effect (PGOE) of study treatment. Overall, 361 patients (mean age 52.4 years, 77.3% male; mean apnea‐hypopnea index [AHI] 27.0 events/h) were randomised to receive pitolisant (n = 242; 50% received CPAP) or placebo (n = 119; 48.7% CPAP). After the dose‐adjustment phase (week 3), 88.8% of patients received pitolisant 40 mg. Compared with placebo, pitolisant produced a significant reduction in the ESS score at week 12 (least square mean difference −2.6 (95% CI: −3.4; −1.8; p < 0.001)) irrespective of CPAP use; and improved the reaction time on OSleR, CGI‐C, and PGOE at week 12. Pitolisant was well tolerated; no new safety signals were identified. In conclusion, pitolisant up to 40 mg once daily was an effective treatment for EDS in patients with moderate to severe OSA irrespective of CPAP use.