Anatole Chessel

Polarization resolved second-harmonic-generation (pSHG) microscopy is increasingly used for mapping organized arrays of non-centrosymmetric proteins such as collagen, myosin and tubulin, and holds potential for probing their molecular structure and supramolecular organization in intact tissues. However, the contrast mechanism of pSHG is complex, and the development of applications in the life sciences is hampered by the lack of models accurately relating the observed pSHG signals to the underlying molecular and macromolecular organization. In this work, we establish a general multi-scale numerical framework relating the micrometer-scale SHG measurements to the atomic-scale and molecular structure of the proteins under study and their supramolecular arrangement. We first develop a new method to automatically analyze pSHG signals independently of the protein type and fiber orientation. We then …

BiaPy, a unified open-source bioimage analysis library, offers a comprehensive suite of deep learning-powered workflows. Tailored for users of all levels, BiaPy features an intuitive interface, zero-code notebooks, and Docker integration. With support for 2D and 3D image data, it addresses existing gaps by providing multi-GPU capabilities, memory optimization, and compatibility with large datasets. As a collaborative and accessible solution, BiaPy aims to empower researchers by democratizing the use of sophisticated and efficient bioimage analysis workflows.

One central question in neuroscience and multiple sclerosis (MS) is how the metabolic coupling between microglia, oligodendrocytes (OLs) and neurons modulate myelin formation, demyelination, remyelination as well as neuronal energetic status and supply failure. The metabolic interplay between different cells has not been studied before at the single-cell level because of the lack of appropriate microscopy techniques. Here we develop advanced multimodal label-free and non-invasive techniques based on non-linear optics and intrinsic biomarkers to image myelin and metabolism with subcellular resolution. We implemented third-harmonic generation (THG) microscopy to probe myelin organization with single fiber resolution and two-photon excitation fluorescence lifetime microscopy (2P-FLIM) of the metabolic biomarkers NAD (P) H and FAD for measurements of cellular redox states and glycolysis and …

BioImageLoader (BIL) is a python library that handles bioimage datasets for machine learning applications, easing simple workflows and enabling complex ones. BIL attempts to wrap the numerous and varied bioimages datasets in unified interfaces, to easily concatenate, perform image augmentation, and batch-load them. By acting at a per experimental dataset level, it enables both a high level of customization and a comparison across experiments. Here we present the library and show some application it enables, including retraining published deep learning architectures and evaluating their versatility in a leave-one-dataset-out fashion.

While supervised deep neural networks have become the dominant method for image analysis tasks in bioimages, truly versatile methods are not available yet because of the diversity of modalities and conditions and the cost of retraining. In practice, day-to-day biological image analysis still largely relies on ad hoc workflows often using classical linear filters. We propose NU-Net, a convolutional neural network filter selectively enhancing cells and nuclei, as a drop-in replacement of chains of classical linear filters in bioimage analysis pipelines. Using a style transfer architecture, a novel perceptual loss implicitly learns a soft separation of background and foreground. We used self-supervised training using 25 datasets covering diverse modalities of nuclear and cellular images. We show its ability to selectively improve contrast, remove background and enhance objects across a wide range of datasets and workflow while keeping image content. The pre-trained models are light and practical, and published as free and open-source software for the community. NU-Net is also available as a plugin for Napari.

Calcific uremic arteriolopathy (CUA) is a severely morbid disease, affecting mostly dialyzed end-stage renal disease (ESRD) patients, associated with calcium deposits in the skin. Calcifications have been identified in ESRD patients without CUA, indicating that their presence is not specific to the disease. The objective of this retrospective multicenter study was to compare elastic fiber structure and skin calcifications in ESRD patients with CUA to those without CUA using innovative structural techniques. Fourteen ESRD patients with CUA were compared to 12 ESRD patients without CUA. Analyses of elastic fiber structure and skin calcifications using multiphoton microscopy followed by machine-learning analysis and field-emission scanning electron microscopy coupled with energy dispersive X-ray were performed. Elastic fibers specifically appeared fragmented in CUA. Quantitative analyses of multiphoton images …

A key property of the human cornea is to maintain its curvature and consequently its refraction capability despite daily changes in intraocular pressure. This is closely related to the multiscale structure of the corneal stroma, which consists of 1–3 µm-thick stacked lamellae made of thin collagen fibrils. Nevertheless, the distribution, size, and orientation of these lamellae along the depth of the cornea are poorly characterized up to now. In this study, we use second harmonic generation (SHG) microscopy to visualize the collagen distribution over the full depth of 10 intact and unstained human corneas (500–600 µm thick). We take advantage of the small coherence length in epi-detection to axially resolve the lamellae while maintaining the corneal physiological curvature. Moreover, as raw epi-detected SHG images are spatially homogenous because of the sub-wavelength size of stromal collagen fibrils, we use a …

Photorefractive Keratectomy (PRK) is a widely used laser-assisted refractive surgical technique. While generally safe, in some cases it leads to subepithelial inflammation or fibrosis. We here present a robust, machine learning based algorithm for the detection of fibrosis based on spectral domain optical coherence tomography (SD-OCT) images recorded in vivo on standard clinical devices.

Solar ultraviolet longwave UVA1 exposure of human skin has short-term consequences at cellular and molecular level, leading at long-term to photoaging. Following exposure, reactive oxygen species (ROS) are generated, inducing oxidative stress that might impair cellular metabolic activity. However, the dynamic of UVA1 impact on cellular metabolism remains unknown because of lacking adequate live imaging techniques. Here we assess overtime the UVA1- induced metabolic stress response in reconstructed human skin with multicolor two-photon fluorescence lifetime microscopy (FLIM). Simultaneous imaging of the two endogenous biomarkers nicotinamide adenine dinucleotide (NAD(P)H) and flavin adenine dinucleotide (FAD) by wavelength mixing allows quantifying cellular metabolism in function of NAD(P)+/NAD(P)H and FAD/FADH2 redox ratios We measure NAD(P)H and FAD fluorescence lifetime and …

Background. Laminins (LM) are heterotrimeric glycoproteins comprised of one α, one β, and one γchain that play a fundamental role in basement membrane architecture and function in human skin. The five C-terminal laminin globular (LG) modules of several α chains are modified by proteolysis to generate a shortened LG1-3 containing LM α chain and a secreted LG4-5 tandem which harbours binding sites for heparan sulphate proteoglycans including those carried by the cell surface receptor syndecan-1. The LG4-5 module within the full-length α3 chain of LM-332 (α3β3γ2) or released from it, was reported to promote the migration of epidermal keratinocytes during the epithelialization phase of wound repair. Here, we have investigated the effect of a peptide mimicking the syndecan-1 binding site within the α3 chain LG4-5 module on wound healing.Materials & Methods. The efficacy of the LM peptide was evaluated in vitro using primary human keratinocytes and in vivo in a model of partial thickness wounds in pigs. Hyaluronic acid based formulations were created as dressings to carry the peptide to the wounds. The follow-up and evaluation of wound closure included macroscopic observations and microscopic analysis of biopsies taken at different times. Imaging tissue sections stained or labelled with antibodies assessed characterization of both the regenerated epidermis and the repaired dermis. Second Harmonic Generation (SHG) Microscopy revealed the 3-dimensional architecture of the collagen fibrils in the repaired dermis.Results.A peptide-induced acceleration of epidermal closure was found in vitro. In vivo, the examination of the …

Tridimensional microscopy and algorithms for automated segmentation and tracing are revolutionizing neuroscience through the generation of growing libraries of neuron reconstructions. Innovative computational methods are needed to analyze these neuronal traces. In particular, means to characterize the geometric properties of traced neurites along their trajectory have been lacking. Here, we propose a local tridimensional (3D) scale metric derived from differential geometry, measuring for each point of a curve the characteristic length where it is fully 3D as opposed to being embedded in a 2D plane or 1D line. The larger this metric is and the more complex the local 3D loops and turns of the curve are. Available through the GeNePy3D open-source Python quantitative geometry library (https://genepy3d.gitlab.io), this approach termed nAdder offers new means of describing and comparing axonal and dendritic arbors. We validate this metric on simulated and real traces. By reanalysing a published zebrafish larva whole brain dataset, we show its ability to characterize different population of commissural axons, distinguish afferent connections to a target region and differentiate portions of axons and dendrites according to their behavior, shedding new light on the stereotypical nature of neurites’ local geometry.

Classification de cornées humaines saines et pathologiques par la quantification d’images SD-OCT cliniques - École polytechnique Accéder directement au contenu Documentation FR Français (FR) Anglais (EN) Se connecter Portail HAL Polytechnique Recherche Loading... Recherche avancée Information de documents Titres Titres Sous-titre Titre de l'ouvrage Titre du volume (Série) Champ de recherche par défaut (multicritères) + texte intégral des PDF Résumé Texte intégral indexé des documents PDF Mots-clés Type de document Sous-type de document Tous les identifiants du document Identifiant HAL du dépôt Langue du document (texte) Pays (Texte) Ville À paraître (true ou false) Ajouter Auteur Auteur (multicritères) Auteur (multicritères) Auteur : Nom complet Auteur : Nom de famille Auteur : Prénom Auteur : Complément de nom, deuxième prénom Auteur : Organisme payeur Auteur : IdHal (chaîne de …

The advent of large-scale fluorescence and electronic microscopy techniques along with maturing image analysis is giving life sciences a deluge of geometrical objects in 2D/3D (+ t) to deal with. These objects take the form of large scale, localised, precise, single cell, quantitative data such as cells’ positions, shapes, trajectories or lineages, axon traces in whole brains atlases or varied intracellular protein localisations, often in multiple experimental conditions. The data mining of those geometrical objects requires a variety of mathematical and computational tools of diverse accessibility and complexity. Here we present a new Python library for quantitative 3D geometry called GeNePy3D which helps handle and mine information and knowledge from geometric data, providing a unified application programming interface (API) to methods from several domains including computational geometry, scale space methods or spatial statistics. By framing this library as generically as possible, and by linking it to as many state-of-the-art reference algorithms and projects as needed, we help render those often specialist methods accessible to a larger community. We exemplify the usefulness of the GeNePy3D toolbox by re-analysing a recently published whole-brain zebrafish neuronal atlas, with other applications and examples available online. Along with an open source, documented and exemplified code, we release reusable containers to allow for convenient and wide usability and increased reproducibility.

In skeletal muscle, long noncoding RNAs (lncRNAs) are involved in dystrophin protein stabilization but also in the regulation of myocytes proliferation and differentiation. Hence, they could represent promising therapeutic targets and/or biomarkers for Duchenne and Becker muscular dystrophy (DMD/BMD). DMD and BMD are X-linked myopathies characterized by a progressive muscular dystrophy with or without dilatative cardiomyopathy. Two-thirds of DMD gene mutations are represented by deletions, and 63% of patients carrying DMD deletions are eligible for 45 to 55 multi-exons skipping (MES), becoming BMD patients (BMDΔ45-55). We analyzed the genomic lncRNA presence in 38 BMDΔ45-55 patients and characterized the lncRNA localized in introns 44 and 55 of the DMD gene. We highlighted that all four lncRNA are differentially expressed during myogenesis in immortalized and primary human myoblasts. In addition, the lncRNA44s2 was pointed out as a possible accelerator of differentiation. Interestingly, lncRNA44s expression was associated with a favorable clinical phenotype. These findings suggest that lncRNA44s2 could be involved in muscle differentiation process and become a potential disease progression biomarker. Based on these results, we support MES45-55 therapy and propose that the design of the CRISPR/Cas9 MES45-55 assay consider the lncRNA sequences bordering the exonic 45 to 55 deletion.

EDAM is a well-established ontology of operations, topics, types of data, and data formats that are used in bioinformatics and its neighbouring fields [1,2] . EDAM-bioimaging is an extension of EDAM dedicated to bioimage analysis, bioimage informatics, and bioimaging. It is being developed in collaboration between the ELIXIR research infrastructure and the NEUBIAS and COMULIS COST Actions, in close contact with the Euro-BioImaging research infrastructure and the Global BioImaging network. EDAM-bioimaging contains an inter-related hierarchy of concepts including bioimage analysis and related operations, bioimaging topics and technologies, and bioimage data and their formats. The modelled concepts enable interoperable descriptions of software, publications, data, workflows, and training materials, fostering open science and "reproducible" bioimage analysis. New developments in EDAM-bioimaging at the time of publication [3] include among others: A concise but relatively comprehensive ontology of Machine learning, Artificial intelligence, and Clustering (to the level relevant in particular in bioimaging, biosciences, and also scientific data analysis in general) Added and refined topics and synonyms within Sample preparation and Tomography, and finalised coverage of imaging techniques (all of these to the high-level extent that influences choices of downstream analysis, i.e. the scope of EDAM) EDAM-bioimaging continues being under active development, with a growing and diversifying community of contributors. It is used in BIII.eu, the registry of bioimage analysis tools, workflows, and training materials, and emerging also in …

AbstractThe advent of large-scale microscopy along with advances in automated image analysis algorithms is currently revolutionizing neuroscience. These approaches result in rapidly increasing libraries of neuron reconstructions requiring innovative computational methods to draw biological insight from. Here, we propose a framework from differential geometry based on scale-space representation to extract a quantitative structural readout of neurite traces seen as tridimensional (3D) curves within the anatomical space. We define and propose algorithms to compute a multiscale hierarchical decomposition of traced neurites according to their intrinsic dimensionality, from which we deduce a local 3D scale, i.e. the scale in microns at which the curve is fully 3D as opposed to being embedded in a 2D plane or a 1D line. We applied our scale-space analysis to recently published data including zebrafish whole brain traces to demonstrate the importance of the computed local 3D scale for description and comparison at the single arbor levels and as a local spatialized information characterizing axons populations at the whole brain level. The use of this broadly applicable approach is facilitated through an open-source implementation in Python available through GeNePy3D, a quantitative geometry library.