HAL Portal UPHF (Université Polytechnique Hauts-de-France)
Not a member yet
28473 research outputs found
Sort by
Gaussian Process-driven Hidden Markov Models for Early Diagnosis of Infant Gait Anomalies
No full textThis work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.Gait analysis is critical in the early detection and intervention of motor neurological disorders in infants. Despite its importance, traditional methods often struggle to model the high variability and rapid developmental changes inherent to infant gait. To address these challenges, we propose a probabilistic Gaussian Process (GP)-driven Hidden Markov Model (HMM) to capture the complex temporal dynamics of infant gait cycles and enable automatic recognition of gait anomalies. We use a Multi-Output GP (MoGP) framework to model interdependencies between multiple gait signals, with a composite kernel designed to account for smooth, non-smooth, and periodic behaviors exhibited in gait cycles. The HMM segments gait phases into normal and abnormal states, facilitating the precise identification of pathological movement patterns in stance and swing phases. The proposed model is trained and assessed using a dataset of infants with and without motor neurological disorders via leave-one-subject-out cross-validation. Results demonstrate that the MoGP outperforms Long Short-Term Memory (LSTM) based neural networks in modeling gait dynamics, offering improved accuracy, variance explanation, and temporal alignment. Further, the predictive performance of MoGP provides a principled framework for uncertainty quantification, allowing confidence estimation in gait trajectory predictions. Additionally, the HMM enhances interpretability by explicitly modeling gait phase transitions, improving the detection of subtle anomalies across multiple gait cycles. These findings highlight the MoGP-HMM framework as a robust automatic gait analysis tool, allowing early diagnosis and intervention strategies for infants with neurological motor disorders
Selective area molecular beam epitaxy of InSb on InP(111) : from thin films to quantum nanostructures
No full textInternational audienceInSb is a material of choice for infrared as well as spintronic devices but its integration on large lattice mismatched semi-insulating III–V substrates has so far altered its exceptional properties. Here, we investigate the direct growth of InSb on InP(111) B substrates with molecular beam epitaxy. Despite the lack of a thick metamorphic buffer layer for accommodation, we show that quasi-continuous thin films can be grown using a very high Sb/In flux ratio. The quality of the films is further studied with Hall measurements on large-scale devices to assess the impact of the InSb surface and InSb/InP interface on the electronic properties. Taking advantage of the optimized growth conditions for the formation of thin films, the selective area molecular beam epitaxial growth of nanostructures is subsequently investigated. Based on cross-sectional transmission electron microscopy and scanning near-field optical microscopy in the middle-wave infrared, ultra-thin and very long in-plane InSb nanowires as well as more complex nanostructures such as nano-rings and crosses are achieved with a good structural quality
Platelet extracellular vesicles-loaded hydrogel bandages for personalized wound care
No full textInternational audienceAutologous or allogeneic platelet-derived extracellular vesicles (pEVs) show potential in enhancing tissue recovery and healing chronic wounds. pEVs promote neovascularization and cell migration while reducing inflammation, oxidative stress, and scarring. However, their efficacy in clinical settings is challenged by their susceptibility to washout by wound exudate. Hydrogel-based bandages are effective carriers that stabilize pEVs for optimal personalized wound care. These bandages can be tailored for easy removal to minimize damage to regenerated tissue and can incorporate antibacterial or moisture-retaining properties. Furthermore, the possibility of integrating sensors in the wound bed will enable a theragnostic approach to healing. This review explores advancements in pEV-loaded hydrogels and their potential for personalized clinical applications
Effect of High Temperature RF Stress on the Trapping Behavior of Carbon Doped AlN/GaN/AlGaN HEMTs
No full textInternational audienceThis study investigates the impact of short-term hightemperature operating life (HTOL) reliability testing on the trapping behavior and power performance of carbon-doped AlN/GaN/AlGaN High Electron Mobility Transistors (HEMTs). HTOL stress can lead to significant degradation in key performance metrics, including output power (Pout) and poweradded efficiency (PAE), primarily due to increased current collapse and knee walkout. Transmission Electron Microscopy (TEM) analysis revealed the formation of voids at the gate foot on the drain side, indicating physical damage. Drain current transient spectroscopy (DCT) revealed post-stress changes in trap dynamics, including the emergence of a secondary emission process (E2). Further analysis attributed E2 to the overlap of enhanced emission from a field-dependent deep-level trap (E1, activation energy ~0.7 eV) and a stress-induced increased density of traps with distributed energy level
Holistic Bowtie Model of AI-Based Technology in Defense Systems: How to Think, Explore, and Implement Efficient and Meaningful Chains of Abilities, Control, Responsibility, and Accountability Between Humans, AI, Organization, and the Environment?
No full textInternational audienceAs AI creates a strong technological push on our societies and defense systems, increasing requirements to counteract hybrid warfare pull the demands for new methods to think defense in a systemic and interconnected way. The holistic bowtie model offers a method for understanding the connections between our defense systems and surrounding organizations and societies on an interconnected planet. It is derived from the class of bowtie models already in use for decades in hazard and accident analysis and is inspired by and a potential answer to the Chinese principle of 天下 Tianxia. The holistic bowtie model focuses on critical situations and is gradually expanded to include other system levels, their causal influences, and control loops. The article starts with the historical example of the spears of Schoeningen, introduces the concept step by step, puts it into the military context of modern (cognitive) warfare, discusses the role of cognition, emotions, truth, sensemaking, trust and mistrust, and finally agility as essential building blocks in a strong alliance
2015 - Récits et fictions du terrorisme: Textes réunis par Alexandre Gefen en collaboration avec le Programme 13-Novembre
No full textCe numéro présente une analyse approfondie des récits et fictions engendrés par les attentats de 2015 en France. Il examine comment la culture française a utilisé divers supports narratifs - témoignages, romans, documentaires, bandes dessinées - pour représenter et comprendre ces événements traumatiques
La construction sur un chemin en indivision forcée
No full textInternational audience(Civ. 3e, 7 mai 2025, no 24-15.027, D. actu. 5 juin 2025, obs. A. Gaillard ; D. 2025. 889
Strain Decoupling in Ge/Si via Selective Misfit-Network Etching.
No full textInternational audienceCMOS-compatible Ge/Si heterostructures enable high-performance electroabsorption modulators, photodetectors, photonic interconnects, and multijunction photovoltaics, yet their scalability is hindered by the 4.2% lattice and thermal mismatch that generates dense dislocations. Here, we report a selective electrochemical process that forms a sub-nanometric (~15 Å) two-dimensional free surface (2DFS) at the Ge/Si interface by etching along the misfit dislocation network. This interfacial gap elastically decouples the Ge epilayer, achieving full strain relaxation, eliminating observable threading dislocations, and enhancing room-temperature photoluminescence while maintaining a smooth, device-compatible surface. Finite-element simulations, porosity–rate calibration, and HAADF-STEM analysis collectively establish the etching conditions and process window for wafer-scale realization of defect-free Ge/Si integration
Integration of Millimeter-Wave Air Filled Cavities and Filters Using Vertically Aligned Carbon Nanotubes
No full textInternational audienceThis work is a step towards the 3D integration of millimeter-wave and THz circuits and subsystems with a nano-packaging technology using vertically aligned carbon nanotubes (VACNTs). In doing so, we have developed and optimized a fabrication process based on a low temperature transfer of VACNT, allowing to successfully implement millimeter-wave air filled cavities and filters towards 75 GHz. The circuits were designed and optimized using a finite element modeling tool. We particularly studied the impact of the VACNT density, which allowed to bridge the gap between device design and fabrication. The measured insertion loss values for the cavity and the filter are 3.78 dB and 4.38 dB, respectively. These experimental values obtained are in excellent agreement with the calculated ones, which paves the way for the 3D integration of more complex and higher frequency functions using this vertically aligned carbon nanotube technology
Luneburg lens for surface water waves
No full textInternational audienceFocusing of surface water waves is investigated using an analog of a Luneburg lens constructed by submerged pillars. Efficient focusing is observed on a wide range of frequencies. The results may find application in the ocean-human interaction, such as coastal protection