[Delirium superimposed on dementia. Clinical practice and perspectives]. / État confusionnel aigu chez les personnes atteintes de démence. Pratique clinique et perspectives.

Delirium is a commonly encountered syndrome in clinical practice. Older people with cognitive disorders are most at risk of developing delirium during the hospital stay. However, the risk of underdiagnosis is significant, particularly because of the overlap of behavioral symptoms and cognitive impairment between delirium and dementia. With an aging population and an ever-increasing prevalence of cognitive disorders, this article exposes the bilateral relationship maintained between dementia and delirium and bring certain diagnostic tools and clinical management clues to support the clinician. Finally, the perspectives of ongoing clinical research that could respond to current challenges are discussed.

L'état confusionnel aigu (ECA) est un syndrome fréquemment rencontré dans la pratique clinique. Les personnes âgées atteintes de troubles cognitifs sont les plus à risque de développer un ECA durant le séjour hospitalier. Le risque de sous-diagnostic est important, les symptômes comportementaux et de l'atteinte cognitive en lien avec la démence étant notamment parfois difficiles à distinguer de ceux de l'état confusionnel. Devant une population vieillissante et une prévalence de troubles cognitifs toujours grandissante, cet article aborde la relation bilatérale entretenue entre la démence et l'ECA, et propose certains outils diagnostiques et aides à la prise en charge pour le clinicien. Enfin, les perspectives de la recherche clinique en cours qui pourraient répondre aux défis actuels sont évoquées.

CRAFT: Concept Recursive Activation FacTorization for Explainability.

Attribution methods, which employ heatmaps to identify the most influential regions of an image that impact model decisions, have gained widespread popularity as a type of explainability method. However, recent research has exposed the limited practical value of these methods, attributed in part to their narrow focus on the most prominent regions of an image - revealing "where" the model looks, but failing to elucidate "what" the model sees in those areas. In this work, we try to fill in this gap with CRAFT - a novel approach to identify both "what" and "where" by generating concept-based explanations. We introduce 3 new ingredients to the automatic concept extraction literature: (i) a recursive strategy to detect and decompose concepts across layers, (ii) a novel method for a more faithful estimation of concept importance using Sobol indices, and (iii) the use of implicit differentiation to unlock Concept Attribution Maps. We conduct both human and computer vision experiments to demonstrate the benefits of the proposed approach. We show that the proposed concept importance estimation technique is more faithful to the model than previous methods. When evaluating the usefulness of the method for human experimenters on a human-centered utility benchmark, we find that our approach significantly improves on two of the three test scenarios. Our code is freely available: github.com/deel-ai/Craft.

A Benchmark for Compositional Visual Reasoning.

A fundamental component of human vision is our ability to parse complex visual scenes and judge the relations between their constituent objects. AI benchmarks for visual reasoning have driven rapid progress in recent years with state-of-the-art systems now reaching human accuracy on some of these benchmarks. Yet, there remains a major gap between humans and AI systems in terms of the sample efficiency with which they learn new visual reasoning tasks. Humans' remarkable efficiency at learning has been at least partially attributed to their ability to harness compositionality - allowing them to efficiently take advantage of previously gained knowledge when learning new tasks. Here, we introduce a novel visual reasoning benchmark, Compositional Visual Relations (CVR), to drive progress towards the development of more data-efficient learning algorithms. We take inspiration from fluid intelligence and non-verbal reasoning tests and describe a novel method for creating compositions of abstract rules and generating image datasets corresponding to these rules at scale. Our proposed benchmark includes measures of sample efficiency, generalization, compositionality, and transfer across task rules. We systematically evaluate modern neural architectures and find that convolutional architectures surpass transformer-based architectures across all performance measures in most data regimes. However, all computational models are much less data efficient than humans, even after learning informative visual representations using self-supervision. Overall, we hope our challenge will spur interest in developing neural architectures that can learn to harness compositionality for more efficient learning.

What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methods.

A multitude of explainability methods has been described to try to help users better understand how modern AI systems make decisions. However, most performance metrics developed to evaluate these methods have remained largely theoretical - without much consideration for the human end-user. In particular, it is not yet clear (1) how useful current explainability methods are in real-world scenarios; and (2) whether current performance metrics accurately reflect the usefulness of explanation methods for the end user. To fill this gap, we conducted psychophysics experiments at scale (n = 1,150) to evaluate the usefulness of representative attribution methods in three real-world scenarios. Our results demonstrate that the degree to which individual attribution methods help human participants better understand an AI system varies widely across these scenarios. This suggests the need to move beyond quantitative improvements of current attribution methods, towards the development of complementary approaches that provide qualitatively different sources of information to human end-users.

A validated Distributed Activation Energy Model (DAEM) to predict the chemical degradation of biomass as a function of hydrothermal treatment conditions.

This study proposes a DAEM (Distributed Activation Energy Model) approach to predict the chemical alterations of lignocellulosic biomass as a function of hydrothermal treatment conditions. The model is first tuned by an original device allowing the sample shrinkage to be continuously assessed during hydrothermal treatment in saturated water vapor up to 190 °C. The shrinkage dynamic is supplied in the DAEM model as an indicator of the degree of biomass conversion. A set of chemical analyses was performed at selected residence times and treatment temperatures to correlate this degree of conversion with the resulting chemical molecules. A set of functions was then derived from this database to correlate the degree of conversion with the components concentrations. Finally, a validation database was built with different combinations of temperature levels and residence times. The model was proved to be predictive on this new dataset.

Electrophysiological correlates of improved executive function following EEG neurofeedback in adult attention deficit hyperactivity disorder.

OBJECTIVE: Event-related potentials (ERPs) are reported to be altered in relation to cognitive processing deficits in attention deficit hyperactivity disorder (ADHD). However, this evidence is mostly limited to cross-sectional data. The current study utilized neurofeedback (NFB) as a neuromodulatory tool to examine the ERP correlates of attentional and inhibitory processes in adult ADHD using a single-session, within-subject design. METHODS: We recorded high-density EEG in 25 adult ADHD patients and 22 neurotypical controls during a Go/NoGo task, before and after a 30-minute NFB session designed to down-regulate the alpha (8-12 Hz) rhythm. RESULTS: At baseline, ADHD patients demonstrated impaired Go/NoGo performance compared to controls, while Go-P3 amplitude inversely correlated with ADHD-associated symptomatology in childhood. Post NFB, task performance improved in both groups, significantly enhancing stimulus detectability (d-prime) and reducing reaction time variability, while increasing N1 and P3 ERP component amplitudes. Specifically for ADHD patients, the pre-to-post enhancement in Go-P3 amplitude correlated with measures of improved executive function, i.e., enhanced d-prime, reduced omission errors and reduced reaction time variability. CONCLUSIONS: A single-session of alpha down-regulation NFB was able to reverse the abnormal neurocognitive signatures of adult ADHD during a Go/NoGo task. SIGNIFICANCE: The study demonstrates for the first time the beneficial neurobehavioral effect of a single NFB session in adult ADHD, and reinforces the notion that ERPs could serve as useful diagnostic/prognostic markers of executive dysfunction.

Multiscale investigation on the chemical and anatomical changes of lignocellulosic biomass for different severities of hydrothermal treatment.

The chemical changes sustained by lignocellulosic biomass during hydrothermal treatment are reflected at multiple scales. This study proposes to benefit from this multiscale nature in order to provide a global understanding of biomass alterations during hydrothermal treatment. For this purpose, complementary imaging techniques-confocal Raman microscopy and X-ray nano-tomography-analysed by image processing and coupled to chemical measurements were used. This unique combination of analyses provided valuable information on topochemical and morphological changes of poplar samples, without the artefacts of sample preparation. At the cell wall level, holocellulose hydrolysis and lignin modifications were observed, which corresponded to anatomical modifications observed at higher scales. Overall, after treatment, samples shrank and had thinner cell walls. When subjected to more severe pre-treatments, cells were disrupted and detached from adjacent cells. Anatomical changes were then used to obtain quantitative indicators of the treatment severity. The effects of treatment at different scales can thus be quantitatively connected in both directions, from micro to macro and from macro to micro.

In situ measurements of viscoelastic properties of biomass during hydrothermal treatment to assess the kinetics of chemical alterations.

This work aimed to use continuous measurements of viscoelastic properties to evaluate the effect of hydrothermal treatment on poplar samples. Different conditions (temperature and pre-soaking liquid: acidic, neutral and alkaline) were tested on wood in both tangential and radial directions. Two viscoelastic properties were determined: the modulus of elasticity and the stress relaxation. The applicability of these properties as indicators of the kinetics of biomass deconstruction was also evaluated, thanks to the chemical analyses performed on the treated solid and the recovered liquid phase. The ultimate goal is to build a macroscopic indicator capable of establishing rules to optimize the hydrothermal treatment before the explosion stage. The joint use of the two parameters succeeded in revealing the effects of chemical degradation, including the coexistence of cleavage and re-condensation and the impact of process conditions (temperature, residence time, and pre-soaking liquid). The monotonous behavior of stress relaxation is a major asset as a possible macroscopic indicator of biomass deconstruction.

Synergistic Influences of Stearic Acid Coating and Recycled PET Microfibers on the Enhanced Properties of Composite Materials.

This study aims to produce novel composite artificial marble materials by bulk molding compound processes, and improve their thermal and mechanical properties. We employed stearic acid as an efficient surface modifying agent for CaCO3 particles, and for the first time, a pretreated, recycled, polyethylene terephthalate (PET) fibers mat is used to reinforce the artificial marble materials. The innovative aspects of the study are the surface treatment of CaCO3 particles by stearic acid. Stearic acid forms a monolayer shell, coating the CaCO3 particles, which enhances the compatibility between the CaCO3 particles and the matrix of the composite. The morphology of the composites, observed by scanning electron microscopy, revealed that the CaCO3 phase was homogeneously dispersed in the epoxy matrix under the support of stearic acid. A single layer of a recycled PET fibers mat was pretreated and designed in the core of the composite. As expected, these results indicated that the fibers could enhance flexural properties, and impact strength along with thermal stability for the composites. This combination of a pretreated, recycled, PET fibers mat and epoxy/CaCO3-stearic acid could produce novel artificial marble materials for construction applications able to meet environmental requirements.

Linking alpha oscillations, attention and inhibitory control in adult ADHD with EEG neurofeedback.

Abnormal patterns of electrical oscillatory activity have been repeatedly described in adult ADHD. In particular, the alpha rhythm (8-12 Hz), known to be modulated during attention, has previously been considered as candidate biomarker for ADHD. In the present study, we asked adult ADHD patients to self-regulate their own alpha rhythm using neurofeedback (NFB), in order to examine the modulation of alpha oscillations on attentional performance and brain plasticity. Twenty-five adult ADHD patients and 22 healthy controls underwent a 64-channel EEG-recording at resting-state and during a Go/NoGo task, before and after a 30 min-NFB session designed to reduce (desynchronize) the power of the alpha rhythm. Alpha power was compared across conditions and groups, and the effects of NFB were statistically assessed by comparing behavioral and EEG measures pre-to-post NFB. Firstly, we found that relative alpha power was attenuated in our ADHD cohort compared to control subjects at baseline and across experimental conditions, suggesting a signature of cortical hyper-activation. Both groups demonstrated a significant and targeted reduction of alpha power during NFB. Interestingly, we observed a post-NFB increase in resting-state alpha (i.e. rebound) in the ADHD group, which restored alpha power towards levels of the normal population. Importantly, the degree of post-NFB alpha normalization during the Go/NoGo task correlated with individual improvements in motor inhibition (i.e. reduced commission errors) only in the ADHD group. Overall, our findings offer novel supporting evidence implicating alpha oscillations in inhibitory control, as well as their potential role in the homeostatic regulation of cortical excitatory/inhibitory balance.