HDR syndrome: Large cohort and systematic review.

HDR syndrome is a rare disease characterized by hypoparathyroidism, deafness, and renal dysplasia. An autosomal dominant disease caused by heterozygous pathogenic GATA3 variants, the penetrance of each associated condition is variable. Literature reviews have provided some answers, but many questions remain, in particular what the relationship is between genotype and phenotype. The current study examines 28 patients with HDR syndrome combined with an exhaustive review of the literature. Some conditions such as hearing loss are almost always present, while others described as rare initially, do not seem to be so rare after all (genital malformations and basal ganglia calcifications). By modeling pathogenic GATA3 variants found in HDR syndrome, we found that missense variations appear to always be located in the same area (close to the two Zinc Finger domain). We describe new pathogenic GATA3 variants, of which some seem to always be associated with certain conditions. Many audiograms were studied to establish a typical audiometric profile associated with a phenotype in HDR. As mentioned in the literature, hearing function should always be assessed as early as possible and follow up of patients with HDR syndrome should include monitoring of parathyroid function and vesicoureteral reflux in order to prevent complications.

Modeling the neurocognitive dynamics of language across the lifespan.

Healthy aging is associated with a heterogeneous decline across cognitive functions, typically observed between language comprehension and language production (LP). Examining resting-state fMRI and neuropsychological data from 628 healthy adults (age 18-88) from the CamCAN cohort, we performed state-of-the-art graph theoretical analysis to uncover the neural mechanisms underlying this variability. At the cognitive level, our findings suggest that LP is not an isolated function but is modulated throughout the lifespan by the extent of inter-cognitive synergy between semantic and domain-general processes. At the cerebral level, we show that default mode network (DMN) suppression coupled with fronto-parietal network (FPN) integration is the way for the brain to compensate for the effects of dedifferentiation at a minimal cost, efficiently mitigating the age-related decline in LP. Relatedly, reduced DMN suppression in midlife could compromise the ability to manage the cost of FPN integration. This may prompt older adults to adopt a more cost-efficient compensatory strategy that maintains global homeostasis at the expense of LP performances. Taken together, we propose that midlife represents a critical neurocognitive juncture that signifies the onset of LP decline, as older adults gradually lose control over semantic representations. We summarize our findings in a novel synergistic, economical, nonlinear, emergent, cognitive aging model, integrating connectomic and cognitive dimensions within a complex system perspective.

Acute and repeated exposures of normal human bronchial epithelial (NHBE) cells culture to particles from a coloured pyrotechnic smoke.

Coloured pyrotechnic smokes are frequently used in the military field and occasionally by civilians, but their health hazards have been little studied. The main concern could rise from inhalation of smoke particles. Our previous study showed that acute exposure to particles from a red signalling smoke (RSS) induced an antioxidant and inflammatory responses in small airway epithelial cells. The aim of this study was to further explore the toxicity of RSS particles at a more proximal level of the respiratory tract, using normal human bronchial epithelial cells grown at the Air-Liquid Interface. Acute exposure (24 h) induced an oxidative stress that persisted 24 h post-exposure, associated with particle internalization and epithelium morphological changes (cuboidal appearance and loss of cilia). Repeated exposures (4×16h) to RSS particles did not trigger oxidative stress but cell morphological changes occurred. Overall, this study provides a better overview of the toxic effects of coloured smoke particles.

Graph-based methods coupled with specific distributional distances for adversarial attack detection.

Artificial neural networks are prone to being fooled by carefully perturbed inputs which cause an egregious misclassification. These adversarial attacks have been the focus of extensive research. Likewise, there has been an abundance of research in ways to detect and defend against them. We introduce a novel approach of detection and interpretation of adversarial attacks from a graph perspective. For an input image, we compute an associated sparse graph using the layer-wise relevance propagation algorithm (Bach et al., 2015). Specifically, we only keep edges of the neural network with the highest relevance values. Three quantities are then computed from the graph which are then compared against those computed from the training set. The result of the comparison is a classification of the image as benign or adversarial. To make the comparison, two classification methods are introduced: (1) an explicit formula based on Wasserstein distance applied to the degree of node and (2) a logistic regression. Both classification methods produce strong results which lead us to believe that a graph-based interpretation of adversarial attacks is valuable.

RIPOR2: A new gene of non-syndromic cochleovestibular dysfunction, discrepancy between human pathology and animal models.

Cochleovestibular dysfunctions are rare conditions misrecognized. A homozygous pathogenic variation c.1561C > T (p.Arg521*) in RIPOR2 (RHO family interacting cell polarization regulator 2) has been identified by WES in Tunisian siblings suffering from congenital bilateral profound hearing and vestibular dysfunctions. In contrast to the vestibular areflexia observed in our patients, deaf Ripor2 KO mouse model and our zebrafish model have normal vestibular function.

Inter-regional correlation estimators for functional magnetic resonance imaging.

Functional magnetic resonance imaging (fMRI) functional connectivity between brain regions is often computed using parcellations defined by functional or structural atlases. Typically, some kind of voxel averaging is performed to obtain a single temporal correlation estimate per region pair. However, several estimators can be defined for this task, with various assumptions and degrees of robustness to local noise, global noise, and region size. In this paper, we systematically present and study the properties of 9 different functional connectivity estimators taking into account the spatial structure of fMRI data, based on a simple fMRI data spatial model. These include 3 existing estimators and 6 novel estimators. We demonstrate the empirical properties of the estimators using synthetic, animal, and human data, in terms of graph structure, repeatability and reproducibility, discriminability, dependence on region size, as well as local and global noise robustness. We prove analytically the link between regional intra-correlation and inter-region correlation, and show that the choice of estimator has a strong influence on inter-correlation values. Some estimators, including the commonly used correlation of averages (ca), are positively biased, and have more dependence to region size and intra-correlation than robust alternatives, resulting in spatially-dependent bias. We define the new local correlation of averages estimator with better theoretical guarantees, lower bias, significantly lower dependence on region size (Spearman correlation 0.40 vs 0.55, paired t-test T=27.2, p=1.1e-47), at negligible cost to discriminative power, compared to the ca estimator. The difference in connectivity pattern between the estimators is not distributed uniformly throughout the brain, but rather shows a clear ventral-dorsal gradient, suggesting that region size and intra-correlation plays an important role in shaping functional networks defined using the ca estimator, and leading to non-trivial differences in their connectivity structure. We provide an open source R package and equivalent Python implementation to facilitate the use of the new estimators, together with preprocessed rat time-series.

Functional brain networks reflect spatial and temporal autocorrelation.

High-throughput experimental methods in neuroscience have led to an explosion of techniques for measuring complex interactions and multi-dimensional patterns. However, whether sophisticated measures of emergent phenomena can be traced back to simpler, low-dimensional statistics is largely unknown. To explore this question, we examined resting-state functional magnetic resonance imaging (rs-fMRI) data using complex topology measures from network neuroscience. Here we show that spatial and temporal autocorrelation are reliable statistics that explain numerous measures of network topology. Surrogate time series with subject-matched spatial and temporal autocorrelation capture nearly all reliable individual and regional variation in these topology measures. Network topology changes during aging are driven by spatial autocorrelation, and multiple serotonergic drugs causally induce the same topographic change in temporal autocorrelation. This reductionistic interpretation of widely used complexity measures may help link them to neurobiology.

Nodal-statistics-based equivalence relation for graph collections.

Node role explainability in complex networks is very difficult yet is crucial in different application domains such as social science, neurosciences, or computer science. Many efforts have been made on the quantification of hubs revealing particular nodes in a network using a given structural property. Yet, in several applications, when multiple instances of networks are available and several structural properties appear to be relevant, the identification of node roles remains largely unexplored. Inspired by the node automorphically equivalence relation, we define an equivalence relation on graph nodes associated with any collection of nodal statistics (i.e., any functions on the node set). This allows us to define new graph global measures: the power coefficient and the orthogonality score to evaluate the parsimony and heterogeneity of a given nodal statistics collection. In addition, we introduce a new method based on structural patterns to compare graphs that have the same vertices set. This method assigns a value to a node to determine its role distinctiveness in a graph family. Extensive numerical results of our method are conducted on both generative graph models and real data concerning human brain functional connectivity. The differences in nodal statistics are shown to be dependent on the underlying graph structure. Comparisons between generative models and real networks combining two different nodal statistics reveal the complexity of human brain functional connectivity with differences at both global and nodal levels. Using a group of 200 healthy controls connectivity networks, our method computes high correspondence scores among the whole population to detect homotopy and finally quantify differences between comatose patients and healthy controls.

Recurrent Benign Paroxysmal Positional Vertigo in DFNB16 Patients with Biallelic STRC Gene Deletions.

OBJECTIVE: Deletions of STRC gene (DFNB16) account for 12% of isolated congenital mild to moderate hearing loss (HL). In mice, the stereocilin protein, encoded by STRC , is present in the vestibular kinocilium embedded in the otoconial membrane of the utricular macula. Despite this, effects on vestibular function have not been widely investigated. The aim of this study was to investigate the prevalence of benign paroxysmal positional vertigo (BPPV) in a cohort of DFNB16 patients. STUDY DESIGN: Observational descriptive epidemiological study. SETTING: Single-center study, in a tertiary referral center. PATIENTS: Older than 5 years, with a genetic diagnosis of HL related to biallelic STRC gene deletions, diagnosed between 2015 and 2021. INTERVENTION: Patients or their parents were interviewed to determine whether they had experienced vertigo or episodes of BPPV. MAIN OUTCOME MEASURE: Criteria were at least five acute episodes of rotatory vertigo, each lasting less than 1 minute, episodes triggered by changes in specific head position, and an absence of neurological symptoms. RESULTS: Sixty-four patients having mild (33%) to moderate (66%) HL were included. Median age was 15 years (range, 6-48 yr). Prevalence of BPPV was 39% (25 of 64). Median age of first onset was 13 years (range, 3-18 yr). CONCLUSIONS: This study showed recurrent BPPV and early age of onset in patients with biallelic STRC gene deletions. BPPV may be associated with the HL phenotype in patients with STRC gene deletions. It is important to inform patients and families of this potential risk such that appropriate management can be proposed.

Innovative graph analysis method to assess gene expression modulation after fine particles exposures of 3D human airway epithelia.

Environmental particles have dramatic consequences for health, especially for the most vulnerable people, such as asthmatics. To better understand the impact on gene expression modulation of fine particles (PM2.5-0.3) from different emission sources, a 3D-airway model, a human bronchial epithelium (MucilAir-HF™) reconstructed from primary cells from healthy (EpiH) or asthmatic (EpiA) donors, was used. Repeated air-liquid exposures were performed, and epithelia were sacrificed to extract RNAs and assess gene expression. Data were analyzed according to the emission sources, physiological status, and exposure doses using a recent model consisting in a graph analysis on pairwise expression ratio. The results were compared with those from the classical ΔΔCt method. The graph analysis method proved to have better statistical properties than the classical ΔΔCt method and demonstrated that repeated PM2.5-0.3 exposures induced a dose-dependent up-regulation of the metabolic gene (CYP1B1) and a down-regulation of the inflammation gene (CXCL10). These modulations were greater for "industrial" than for "urban traffic" fine particles, and the effects were found to be greater after exposure of EpiA than EpiH, thus emphasizing the importance of the epithelium's physiological status in sensitivity to particles. Our study is original in terms of the experimental conditions and the graphical statistical analysis model established. The results highlight the importance of particle chemistry on the modulation of cellular and molecular responses, which may vary according to the individual's vulnerability.

Functional hub disruption emphasizes consciousness recovery in severe traumatic brain injury.

Severe traumatic brain injury can lead to transient or even chronic disorder of consciousness. To increase diagnosis and prognosis accuracy of disorder of consciousness, functional neuroimaging is recommended 1 month post-injury. Here, we investigated brain networks remodelling on longitudinal data between 1 and 3 months post severe traumatic brain injury related to change of consciousness. Thirty-four severe traumatic brain-injured patients were included in a cross-sectional and longitudinal clinical study, and their MRI data were compared to those of 20 healthy subjects. Long duration resting-state functional MRI were acquired in minimally conscious and conscious patients at two time points after their brain injury. The first time corresponds to the exit from intensive care unit and the second one to the discharge from post-intensive care rehabilitation ward. Brain networks data were extracted using graph analysis and metrics at each node quantifying local (clustering) and global (degree) connectivity characteristics. Comparison with brain networks of healthy subjects revealed patterns of hyper- and hypo-connectivity that characterize brain networks reorganization through the hub disruption index, a value quantifying the functional disruption in each individual severe traumatic brain injury graph. At discharge from intensive care unit, 24 patients' graphs (9 minimally conscious and 15 conscious) were fully analysed and demonstrated significant network disruption. Clustering and degree nodal metrics, respectively, related to segregation and integration properties of the network, were relevant to distinguish minimally conscious and conscious groups. At discharge from post-intensive care rehabilitation unit, 15 patients' graphs (2 minimally conscious, 13 conscious) were fully analysed. The conscious group still presented a significant difference with healthy subjects. Using mixed effects models, we showed that consciousness state, rather than time, explained the hub disruption index differences between minimally conscious and conscious groups. While severe traumatic brain-injured patients recovered full consciousness, regional functional connectivity evolved towards a healthy pattern. More specifically, the restoration of a healthy brain functional segregation could be necessary for consciousness recovery after severe traumatic brain injury. For the first time, extracting the hub disruption index directly from each patient's graph, we were able to track the clinical alteration and subsequent recovery of consciousness during the first 3 months following a severe traumatic brain injury.

Oxidative potential and in vitro toxicity of particles generated by pyrotechnic smokes in human small airway epithelial cells.

Pyrotechnic smokes are widely used in civilian and military applications. The major issue arise from the release of particles after smoke combustion but the health risks related to their exposure are poorly documented whereas toxicity of airborne particles on the respiratory target are very well known. Therefore, this study aimed to explore the in vitro toxicity of the particle fraction of different pyrotechnic smokes. Particles from a red signalling smoke (RSS), an hexachloroethane-based obscuring smoke (HC-OS) and an anti-intrusion smoke (AIS) were collected from the cloud. RSS particles displayed the highest organic fraction (quinones and polycyclic aromatic hydrocarbons) of the three samples characterized. AIS particles contained K and cholesterol derivatives. HC-OS particles were mainly metallic with very high concentrations of Al, Fe and Ca. Intrinsic oxidative potential of smoke particles was measured with two assays. Depletions of DTT by RSS particles was greater than depletion obtained with AIS and HC-OS particles but depletion of acid ascorbic (AA) was only observed with HC-OS particles. In vitro toxicity was assessed by exposing human small airway epithelial cells (SAEC) to various concentrations of particles. After 24 h of exposure, cell viability was not affected but significant modifications of mRNA expression of antioxidant (SOD-1 and -2, catalase, HO-1, NQO-1) and inflammatory markers (IL-6, IL-8, TNF-α) were observed and were dependent on smoke type. Particles rich in metal, such as HC-OS, induced a greatest depletion of AA and a greatest inflammatory response, whereas particles rich in organic compounds, such as RSS, induced a greatest DTT depletion and a greatest antioxidant response. In conclusion, the three smoke particles have an intrinsic oxidative potential and triggered a cell adaptive response. Our study improved the knowledge of particle toxicity of pyrotechnic smokes and scientific approach developed here could be used to study other type of particles.

Human skin responses to environmental pollutants: A review of current scientific models.

Whatever the exposure route, chemical, physical and biological pollutants modify the whole organism response, leading to nerve, cardiac, respiratory, reproductive, and skin system pathologies. Skin acts as a barrier for preventing pollutant modifications. This review aims to present the available scientific models, which help investigate the impact of pollution on the skin. The research question was "Which experimental models illustrate the impact of pollution on the skin in humans?" The review covered a period of 10 years following a PECO statement on in vitro, ex vivo, in vivo and in silico models. Of 582 retrieved articles, 118 articles were eligible. In oral and inhalation routes, dermal exposure had an important impact at both local and systemic levels. Healthy skin models included primary cells, cell lines, co-cultures, reconstructed human epidermis, and skin explants. In silico models estimated skin exposure and permeability. All pollutants affected the skin by altering elasticity, thickness, the structure of epidermal barrier strength, and dermal extracellular integrity. Some specific models concerned wound healing or the skin aging process. Underlying mechanisms were an exacerbated inflammatory skin reaction with the modulation of several cytokines and oxidative stress responses, ending with apoptosis. Pathological skin models revealed the consequences of environmental pollutants on psoriasis, atopic dermatitis, and tumour development. Finally, scientific models were used for evaluating the safety and efficacy of potential skin formulations in preventing the skin aging process or skin irritation after repeated contact. The review gives an overview of scientific skin models used to assess the effects of pollutants. Chemical and physical pollutants were mainly represented while biological contaminants were little studied. In future developments, cell hypoxia and microbiota models may be considered as more representative of clinical situations. Models considering humidity and temperature variations may reflect the impact of these changes.

Functional Brain Dysconnectivity in Parkinson's Disease: A 5-Year Longitudinal Study.

BACKGROUND: Tracking longitudinal functional brain dysconnectivity in Parkinson's disease (PD) is a key element to decoding the underlying physiopathology and understanding PD progression. OBJECTIVES: The objectives of this follow-up study were to explore, for the first time, the longitudinal changes in the functional brain networks of PD patients over 5 years and to associate them with their cognitive performance and the lateralization of motor symptoms. METHODS: We used a 5-year longitudinal cohort of PD patients (n = 35) who completed motor and non-motor assessments and sequent resting state (RS) high-density electroencephalography (HD-EEG) recordings at three timepoints: baseline (BL), 3 years follow-up (3YFU) and 5 years follow-up (5YFU). We assessed disruptions in frequency-dependent functional networks over the course of the disease and explored their relation to clinical symptomatology. RESULTS: In contrast with HC (n = 32), PD patients showed a gradual connectivity impairment in α2 (10-13 Hz) and ß (13-30 Hz) frequency bands. The deterioration in the global cognitive assessment was strongly correlated with the disconnected networks. These disconnected networks were also associated with the lateralization of motor symptoms, revealing a dominance of the right hemisphere in terms of impaired connections in the left-affected PD patients in contrast to dominance of the left hemisphere in the right-affected PD patients. CONCLUSIONS: Taken together, our findings suggest that with disease progression, dysconnectivity in the brain networks in PD can reflect the deterioration of global cognitive deficits and the lateralization of motor symptoms. RS HD-EEG may be an early biomarker of PD motor and non-motor progression. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Spatiotemporal reorganization of corticostriatal networks encodes motor skill learning.

Motor skill learning requires the activity of the dorsal striatum, with a differential global implication of the dorsomedial and dorsolateral territories. We investigate here whether and how specific striatal neurons encode the acquisition and consolidation of a motor skill. Using ex vivo two-photon calcium imaging after rotarod training, we report that highly active (HA) striatal populations arise from distinct spatiotemporal reorganization in the dorsomedial (DMS) and dorsolateral (DLS) striatum networks and are correlated with learning performance. The DMS overall activity decreases in early training, with few and sparsely distributed HA cells, while the DLS shows a progressive and long-lasting formation of HA cell clusters. These reorganizations result from reinforcement of synaptic connections to the DMS and anatomical rearrangements to the DLS. Targeted silencing of DMS or DLS HA cells with the cFos-TRAP strategy strongly impairs individual performance. Our data reveal that discrete domains of striatal populations encode acquisition and long-lasting retention of a motor skill.

Congenital abnormalities associated with microtia: A 10-YEARS retrospective study.

OBJECTIVE: Microtia is a congenital auricular malformation, often part of a syndromic form (35%-55% of cases). The accurate prevalence of associated malformations remains to be determined with regard to the heterogeneous results of the previous studies. This study aims to describe in a large population cohort the abnormalities associated with microtia and to determine the most suitable assessment for these children. METHODS: This is a retrospective and observational cohort study collecting data from the medical records of children affected by microtia, diagnosed or followed-up between 2007 and 2017. Data were collected via a computer database. Clinical data, as well as imaging or genetic results, were noted. RESULTS: Six hundred ninety four children were included, 587 (84.6%) with unilateral and 107 (15.4%) with bilateral microtia. Inner ear malformations were observed in 14.1% of the ears. The main associated anomalies were hemifacial microsomia (29%), velopharyngeal insufficiency (9%), ophthalmologic (6.2%), vertebral (5.9%), cardiac (5.5%) and kidney (3%) abnormalities. Main identified entities were Goldenhar, Treacher-Collins and Guion-Almeida syndromes. CONCLUSION: A comprehensive clinical assessment must be completed when microtia is diagnosed. Besides screening well-known oculo-auriculo-vertebral spectrum malformations, velopharyngeal insufficiency should be systematically sought. Specialized care must be provided to the very frequently associated hemifacial macrosomia. Mild forms of this last malformation may correspond to Guion-Almeida syndrome, especially in cases of learning disability.

A covariate-constraint method to map brain feature space into lower dimensional manifolds.

Human brain connectome studies aim to both explore healthy brains, and extract and analyze relevant features associated with pathologies of interest. Usually this consists of modeling the brain connectome as a graph and using graph metrics as features. A fine brain description requires graph metrics computation at the node level. Given the relatively reduced number of patients in standard cohorts, such data analysis problems fall in the high-dimension, low-sample-size framework. In this context, our goal is to provide a machine learning technique that exhibits flexibility, gives the investigator an understanding of the features and covariates, allows visualization and exploration, and yields insight into the data and the biological phenomena at stake. The retained approach is dimension reduction in a manifold learning methodology; the originality is that the investigator chooses one (or several) reduced variables. The proposed method is illustrated in two studies. The first one addresses comatose patients; the second one compares young and elderly populations. The method sheds light on the differences between brain connectivity graphs using graph metrics and potential clinical interpretations of these differences.

Auditory processing disorder in children: the value of a multidisciplinary assessment.

PURPOSE: Auditory processing disorder (APD) may affect 0.2-5% of the paediatric population. The diagnosis of APD remains difficult because of polymorphic symptoms possibly entangled with other difficulties. The purpose of this study was to evaluate a new multi-disciplinary assessment in the French language. METHODS: The battery of tests was composed of: (a) APD targeted speech assessment: speech perception in noise, a dichotic test, temporal processing tests (patterns); (b) Psychometric assessment: sustained auditory attention, sustained visual attention, evaluation of cognitive functions; (c) phonemic identification and discrimination; (d) ENT examination, tonal and vocal audiometry and ABR recordings. The diagnosis was made if two of the targeted speech tests were 2 standard deviations (SDs) below the mean or if only one of the tests was 3 SDs below. The auditory attention tests, as well as the phonemic identification and discrimination tests were complementary to the diagnostic battery. However, they did not allow for the diagnosis of APD. RESULTS: 50 children suspected of APD benefited from this protocol, and 12 were excluded from the study. A diagnosis of APD was confirmed in 17 children (45%). 59% of the patients had associated disorders. The most effective tests for diagnosing APD were dichotic testing (p = 0.001) and pattern recognition (frequency, p = 0.001). The sustained auditory attention test (p = 0.01) and the phonemic identification and discrimination test reinforced the diagnosis of APD. CONCLUSION: It seems important to evaluate children suspected of APD with a multi-disciplinary protocol. It makes it possible to diagnose APD children, but also to identify attentional difficulties and cognitive disorders that may be associated.

Influence of the environmental relative humidity on the inflammatory response of skin model after exposure to various environmental pollutants.

The skin is an essential barrier, protecting the body against the environment and its numerous pollutants. Several environmental pollutants are known to affect the skin, inducing premature aging through mechanisms including oxidative stress, inflammation, and impairment of skin functions. Even climate conditions can impact the skin. Therefore, using a Reconstructed Human Epidermis (RHE), we tested the effect of two samples of fine particulate matters (PM0.3-2.5 - one metals-rich sample and the other organic compounds-rich), two Volatile Organic Compounds mixtures (VOCs - from a solvent-based paint and a water-based paint) and Tobacco Smoke (TS). All pollutants affected cellular functionality, but to a lesser extent for the water-based paint VOC. This effect was enhanced when RHE were preconditioned for 2 h by a semi-dry airflow (45% relative humidity) before pollutants application, compared to preconditioning by a humid airflow (90% relative humidity). In the absence of preconditioning, IL-1α, IL-6, IL-8, and RANTES were almost systematically induced by pollutants. When RHE were preconditioned by a semi-dry or humid airflow before being subjected to pollutants, the increase of IL-1α, IL-8, and RANTES falls into two groups. Similarly to RHE not treated with pollutants, RHE treated with VOCs after preconditioning by a semi-dry airflow showed increased IL-1α, IL-8, and RANTES release. On the contrary, RHE treated with PM or TS after preconditioning by a semi-dry airflow show a lower increase in IL-1α, IL-8, and RANTES compared to preconditioning by a humid airflow. The effect of real environmental relative humidity conditions of the air, combined with acute exposure to various environmental pollutants, seemed to relate mainly to structural changes of the skin, determining the outcome of the inflammatory response depending on the physicochemical characteristics of pollutants.

U-Net Neural Network for Heartbeat Detection in Ballistocardiography.

Monitoring vital signs of neonates can be harmful and lead to developmental troubles. Ballistocardiography, a contactless heart rate monitoring method, has the potential to reduce this monitoring pain. However, signal processing is uneasy due to noise, inherent physiological variability and artifacts (e.g. respiratory amplitude modulation and body position shifts). We propose a new heartbeat detection method using neural networks to learn this variability. A U-Net model takes thirty-second-long records as inputs and acts like a nonlinear filter. For each record, it outputs the samples probabilities of belonging to IJK segments. A heartbeat detection algorithm finally detects heartbeats from those segments, based on a distance criterion. The U-Net has been trained on 30 healthy subjects and tested on 10 healthy subjects, from 8 to 74 years old. Heartbeats have been detected with 92% precision and 80% recall, with possible optimization in the future to achieve better performance.