Valeo (France)

Anger while driving negatively affects performance in both manual and supervised driving scenarios. While numerous studies have focused on detecting anger behind the wheel, research on its regulation remains limited and does not extend to automated driving. Moreover, previous work used alone emotion regulation strategies while combined methods work best in emotion modulation. This study explores a multimodal approach to regulate anger in autonomous driving by integrating auditory and visual modalities. Specifically, we combined low-arousal music, unconscious biofeedback via green ambient lighting, and participants’ photos. In our experiment, 36 participants were split into three groups. Anger was induced in two groups with only one benefiting from the regulation. The third is a control group with neutral induction and no regulation. Our findings highlight that the multimodal intervention effectively improves subjective and physiological emotional responses without specific impact on driving performances. Additionally, participants reported high acceptance and perceived relaxation, highlighting the potential of this approach for enhancing driver well-being in autonomous vehicles.

ABSTRACTSeveral cancer potency estimates have been proposed by regulatory agencies to characterize the dose response of cobalt and/or cobalt compounds. The objective of this research is to investigate whether these proposed cancer potency estimates for certain cobalt substances align with the available epidemiology literature. After review of the epidemiological literature, we identified a study appropriate for our analysis. We established whether our identified study was adequately powered to detect an elevated lung cancer risk. The power analysis assumed a Poisson distribution and used a one‐sided significance level of 0.05. Lung tumors in animals served as the basis for cancer potency estimates for several regulatory bodies. The study population from our identified study was used to calculate predicted excess lung cancer deaths using potency values reported by four regulatory organizations, which were then compared to observed lung cancer deaths. Monte Carlo methods were used to estimate sample size and cobalt exposure distribution of the highest exposure group. We determined that the our identified study has ≥ 98% statistical power to detect a 1.5‐fold or greater increase in lung cancer due to cobalt exposure in all but the lowest exposure group and all four exposure groups had 100% statistical power to detect a 2.0‐fold or greater increase in lung cancer due to cobalt exposure. Cobalt exposure at the estimated median of the highest exposure group resulted in hypothetical standardized mortality ratios (SMR) estimated from the regulatory potency values ranging from 3.54 to 8.61 compared to an observed SMR of 1.15 (95% CI: 0.92–1.43) in our identified study. On the basis of this analysis, the cancer potency estimates proposed by the included regulatory organizations are likely overestimations of excess lifetime human cancer risk after cobalt inhalation exposure.

Lifestyle interventions constitute first-line strategies for the management of metabolic dysfunction-associated steatotic liver disease (MASLD), with dietary modification and increased physical activity forming the core evidence-based approaches. In this framework, plant-based supplements are being investigated as potential adjunctive strategies. Totum-448, a polyphenol-rich plant formulation, has shown promising effects in preclinical studies. Here, we evaluated the effects of Totum-448, voluntary exercise (Vex), and their combination in a hamster model of MASLD. Fifty-four male golden Syrian hamsters were fed either a normal diet (ND) or a Western diet (WD) for six weeks. WD-fed animals were then randomized to receive WD alone, WD supplemented with Totum-448 (5% w/w), WD with Vex, or WD with both interventions for an additional five weeks. We hypothesized that only the combined intervention would elicit measurable benefits. Contrary to this hypothesis, distinct effects were observed for each intervention: Totum-448 reduced circulating lipid levels and downregulated hepatic markers of inflammation and fibrosis, whereas Vex improved body composition, increased energy expenditure, and reduced hepatic lipid content. No significant improvements were detected in histological markers of inflammation or fibrosis. The combined intervention did not provide additional benefits beyond those of the individual interventions. In conclusion, Totum-448 and Vex exert complementary effects in MASLD. Although no additive effects were observed, the combined intervention provided the broadest overall protection by preserving the benefits of each intervention alone. These findings support the relevance of combining exercise with plant-based nutritional strategies in the management of MASLD.

This article presents a state-of-the-art review of recent advances aimed at transforming traditional Failure Mode and Effects Analysis (FMEA) into a more intelligent, data-driven, and semantically enriched process. As engineered systems grow in complexity, conventional FMEA methods, which are largely manual, document-centric, and expert-dependent, have become increasingly inadequate for addressing the demands of modern systems engineering. We examine how techniques from Artificial Intelligence (AI), including machine learning and natural language processing, can transform FMEA into a more dynamic, data-driven, intelligent, and model-integrated process by automating failure prediction, prioritisation, and knowledge extraction from operational data. In parallel, we explore the role of ontologies in formalising system knowledge, supporting semantic reasoning, improving traceability, and enabling cross-domain interoperability. The review also synthesises emerging hybrid approaches, such as ontology-informed learning and large language model integration, which further enhance explainability and automation. These developments are discussed within the broader context of Model-Based Systems Engineering (MBSE) and function modelling, showing how AI and ontologies can support more adaptive and resilient FMEA workflows. We critically analyse a range of tools, case studies, and integration strategies, while identifying key challenges related to data quality, explainability, standardisation, and interdisciplinary adoption. By leveraging AI, systems engineering, and knowledge representation using ontologies, this review offers a structured roadmap for embedding FMEA within intelligent, knowledge-rich engineering environments.

Abstract Multi-access edge computing (MEC) has emerged as a promising paradigm to reduce communication overhead and computational burden on mobile users (MUs) by offloading resource-intensive applications to nearby multi-access edge servers (MESs). This strategy significantly lowers energy consumption and execution latency. However, most existing studies on MEC assume an ideal scenario in which MESs are already optimally deployed, typically co-located with every base station (BS) in fifth-generation (5G) networks. While deploying an MES at each BS ensures low-latency services and high quality of service (QoS), it introduces several practical limitations. These include high capital and operational expenditures for service providers, underutilised MESs in sparsely populated regions, and elevated service costs for end users due to continuous MES activation regardless of demand. To address these challenges, this work proposes a novel Delay-based Categorisation and Adaptive MES-Activation (DCAMA) approach aimed at optimising MES activation in beyond-5G (B5G) wireless networks. Rather than assuming static and full MES deployment, the DCAMA strategy dynamically determines the minimum number of MESs required based on current network workload and task urgency. This energy-aware and cost-efficient method enables scalable MES activation without compromising service performance .

Facial nerve palsy is the most serious complication to be aware of in surgery for parotid gland tumors. It occurs due to various factors including direct intraoperative injury, traction, thermal injury, ischemia, and anatomical variations. While most cases occur immediately after surgery, some may develop within the first 24 to 72 hours due to postoperative edema or hematoma. Delayed facial nerve paralysis occurring more than 72 hours after surgery is extremely rare. We report a case of a 60-year-old woman who developed delayed facial nerve palsy 16 days after parotid gland surgery. Mild paralysis of the lower lip was observed postoperatively, but no abnormalities were detected in other facial muscles. The postoperative course was uneventful, and the patient was discharged on the eighth day. However, on the 10th day after the surgery, pain appeared around her ear, and severe facial nerve paralysis developed on the 16th day. Pain was present from the ear to the scalp of the temporal region, but no objective findings such as postoperative infection, edema, hematoma, or tumor recurrence were observed. The patient received combined therapy with steroids and antiviral agents based on the treatment for idiopathic facial nerve palsy, and the paralysis fully recovered after four months. Delayed facial nerve palsy following parotid gland tumor surgery is extremely rare, and it is important to investigate causative factors before initiating treatment.

Functional electrical stimulation (FES)-assisted cycling may reduce perceived effort by lowering the required motor command compared to voluntary cycling. While benefits on effort perception have been shown during walking in stroke and multiple sclerosis patients, its effectiveness during cycling in stroke rehabilitation remains unproven. Thus, this work aimed to test the effect of functional electrical stimulation-assisted cycling on stroke patients’ perception of effort (primary aim) in a randomized controlled study design. In an exploratory way, this work also aimed to examine the effect of FES-assisted cycling on cerebral blood flow and cognitive performance (exploratory aims) in a subsample. Fifteen post-stroke patients completed functional electrical stimulation-assisted (Kurage, Lyon, France) and traditional cycling sessions separated by 72 h. Perceived effort, cardio, and cerebrovascular parameters were monitored during exercise. Cognitive performance was assessed before and after each session. Qualitative data were reported after both sessions. Patients reported a lower perceived effort during functional electrical stimulation-assisted cycling than traditional cycling. Both sessions increased heart rate, end-tidal CO2, cardiac output, and cerebral artery blood flow velocity, with higher blood lactate levels after functional electrical stimulation-assisted cycling. Traditional and functional electrical stimulation-assisted cycling positively impacted cognitive performance, with more pronounced improvements observed in the FES condition. Traditional and functional electrical stimulation-assisted cycling induced similar increased cardio and cerebrovascular responses. However, patients perceived functional electrical stimulation-assisted cycling as less effortful than traditional cycling. As effort is a barrier to regular exercise engagement and adherence, these results are promising for implementing functional electrical stimulation-assisted cycling in stroke patients’ rehabilitation.

The increasing insistence on the environmentally friendly urban infrastructure requires intelligent structures that can take into account the heterogeneous sensor data, predictive analytics, and computational modeling. Conventional solutions can be ineffective in controlling the large-scale smart systems, leading to poor energy consumption and high carbon emission. In this paper, AI-SF2, a multisensor data fusion, AI-based predictive analytics, and computational modeling are proposed as an integrated framework to optimize smart infrastructures and green energy systems. Structural, environmental and energy sensor information are combined to offer strong situational awareness, and deep learning models predict energy demand, identify anomalies, and project carbon emissions. Real-time decision making and proactive resource allocation is simulated using computational models that provide dynamic interaction. Large scale simulations and testbed evaluations show up to 21.6 % energy savings, 18.9% renewable integration, and 24.2 % faster anomaly detection than current baselines. AI-SF2 provides a challenge-scalable, adaptive, and realistic solution to the sustainable smart cities. It also focuses on safe, privacy-capable sensor integration needed to have reliable smart infrastructure functioning.