33 research outputs found
Optimisation algorithms in non-standard Banach spaces for inverse problems in imaging
No full textThis thesis focuses on the modelling, the theoretical analysis and the numerical implementation of advanced optimisation algorithms for imaging inverse problems (e.g,., image reconstruction in computed tomography, image deconvolution in microscopy imaging) in non-standard Banach spaces.
It is divided into two parts: in the former, the setting of Lebesgue spaces with a variable exponent map is considered to improve adaptivity of the solution with respect to standard Hilbert reconstructions; in the latter a modelling in the space of Radon measures is used to avoid the biases observed in sparse regularisation methods due to discretisation.
In more detail, the first part explores both smooth and non-smooth optimisation algorithms in reflexive spaces, which are Banach spaces endowed with the so-called Luxemburg norm. As a first result, we provide an expression of the duality maps in those spaces, which are an essential ingredient for the design of effective iterative algorithms.
To overcome the non-separability of the underlying norm and the consequent heavy computation times, we then study the class of modular functionals which directly extend the (non-homogeneous) -power of -norms to the general . In terms of the modular functions, we formulate handy analogues of duality maps, which are amenable for both smooth and non-smooth optimisation algorithms due to their separability. We thus study modular-based gradient descent (both in deterministic and in a stochastic setting) and modular-based proximal gradient algorithms in , and prove their convergence in function values. The spatial flexibility of such spaces proves to be particularly advantageous in addressing sparsity, edge-preserving and heterogeneous signal/noise statistics, while remaining efficient and stable from an optimisation perspective. We numerically validate this extensively on 1D/2D exemplar inverse problems (deconvolution, mixed denoising, CT reconstruction).
The second part of the thesis focuses on off-the-grid Poisson inverse problems formulated within the space of Radon measures. Our contribution consists in the modelling of a variational model which couples a Kullback-Leibler data term with the Total Variation regularisation of the desired measure (that is, a weighted sum of Diracs) together with a non-negativity constraint. A detailed study of the optimality conditions and of the corresponding dual problem is carried out and an improved version of the Sliding Frank-Wolfe algorithm is used for computing the numerical solution efficiently. To mitigate the dependence of the results on the choice of the regularisation parameter, an homotopy strategy is proposed for its automatic tuning, where, at each algorithmic iteration checks whether an informed stopping criterion defined in terms of the noise level is verified and updates the regularisation parameter accordingly. Several numerical experiments are reported on both simulated 2D and real 3D fluorescence microscopy data
A continuous, non-convex & sparse super-resolution approach for fluorescence microscopy data with Poisson noise
No full textWe propose a non-convex sparsity-promoting variational
model for the problem of super-resolution in Single
Molecule Localization Microscopy (SMLM). Namely, we study
a continuous non-convex relaxation of a non-continuous and
non-convex variational model where a weighted-L2 data fidelity
modeling signal-dependent Poisson noise is combined with an
L0-regularization to promote signal sparsity. The proposed relaxation
is obtained by adapting the Continuous Exact L0 (CEL0)
relaxation of the analogous `2`0 problem with Gaussian noise
to the Poisson scenario, which is more realistic in fluorescence
microscopy applications. The associated optimization problem is
then solved by an iterative reweighted L1 (IRL1) algorithm. The
weighted-L2 data fidelity leads to a challenging estimation of the
algorithmic parameters for which efficient computation strategies
are detailed. To validate our approach, we report qualitative
and quantitative localization results for a simulated dataset,
showing that the proposed weighted-CEL0 (WCEL0) model is
well suited and capable to deal with Poisson measurements with
high accuracy and precision
Banaras jyotirliṅgas: constitution and transformations of a transposed divine group and its pilgrimage
No full textBanaras (Varanasi, Uttar Pradesh, India) is renowned as one of the more notable pilgrimage destinations of India. The ways of approaching and investigating the sacred landscape and the religious practices of this city varied throughout times and are still a matter of discussion among scholars. The author firstly addresses this debate in order to re-conceptualize the need and intents of writing (still) about Banaras and its mainstream religious traditions. The contribution addresses one common pattern of Indian sacred geography, that is the spatial transposition of gods. The article, in fact, goes through the formation path of a transposed group of pan-Indian deities, namely the jyotirliṅgas, in a city which is presented by eulogistic literature as a universal tīrtha, where all sacred centres and gods dwell. Through the analysis of textual and visual material the author shows how these divine forms have been produced in the city’s territory throughout time and projected spatially in the various shrines and, eventually, in a procession. The pilgrimage circuit connected with the twelve local jyotirliṅgas is investigated as a recent and evolving practice of Banaras religious life and its currently deviating path is shown as something to be constantly rephrased and negotiated. Ritual transformations appear as challenged by the need to adapt and survive in a developing urban context, where sacred space is shared, contested and cyclically re-written
Off-the-grid regularisation for Poisson inverse problems
No full textOff-the-grid regularisation has been extensively employed over the last decade in the context of ill-posed inverse problems formulated in the continuous setting of the space of Radon measures . These approaches enjoy convexity and counteract the discretisation biases as well the numerical instabilities typical of their discrete counterparts. In the framework of sparse reconstruction of discrete point measures (sum of weighted Diracs), a Total Variation regularisation norm in
is typically combined with an data term modelling additive Gaussian noise. To assess the framework of off-the-grid regularisation in the presence of signal-dependent Poisson noise, we consider in this work a variational model where Total Variation regularisation is coupled with a Kullback–Leibler data term under a non-negativity constraint. Analytically, we study the optimality conditions of the composite functional and analyse its dual problem. Then, we consider an homotopy strategy to select an optimal regularisation parameter and use it within a Sliding Frank-Wolfe algorithm. Several numerical experiments on both 1D/2D/3D simulated and real 3D fluorescent microscopy data are reported
Linear FDEM subsoil data inversion in Banach spaces
No full textThe applicative motivation of this paper is the reconstruction of some electromagnetic features of the earth superficial layer by measurements taken above the ground. We resort to frequency domain electromagnetic data inversion through a well-known linear integral model by considering three different collocation methods to approximate the solution of the continuous problem as a linear combination of linearly independent functions. The discretization leads to a strongly ill-conditioned linear system. To overcome this difficulty, an iterative regularization method based on Landweber iterations in Banach spaces is applied to reconstruct solutions which present discontinuities or have a low degree of smoothness. This kind of solutions are common in many imaging applications. Several numerical experiments show the good performance of the algorithm in comparison to other regularization techniques
Algorithmes d'optimisation dans des espaces de Banach non standard pour problèmes inverses en imagerie
No full textThis thesis focuses on the modelling, the theoretical analysis and the numerical implementation of advanced optimisation algorithms for imaging inverse problems (e.g,., image reconstruction in computed tomography, image deconvolution in microscopy imaging) in non-standard Banach spaces. It is divided into two parts: in the former, the setting of Lebesgue spaces with a variable exponent map L^{p(cdot)} is considered to improve adaptivity of the solution with respect to standard Hilbert reconstructions; in the latter a modelling in the space of Radon measures is used to avoid the biases observed in sparse regularisation methods due to discretisation.In more detail, the first part explores both smooth and non-smooth optimisation algorithms in reflexive L^{p(cdot)} spaces, which are Banach spaces endowed with the so-called Luxemburg norm. As a first result, we provide an expression of the duality maps in those spaces, which are an essential ingredient for the design of effective iterative algorithms.To overcome the non-separability of the underlying norm and the consequent heavy computation times, we then study the class of modular functionals which directly extend the (non-homogeneous) p-power of L^p-norms to the general L^{p(cdot)}. In terms of the modular functions, we formulate handy analogues of duality maps, which are amenable for both smooth and non-smooth optimisation algorithms due to their separability. We thus study modular-based gradient descent (both in deterministic and in a stochastic setting) and modular-based proximal gradient algorithms in L^{p(cdot)}, and prove their convergence in function values. The spatial flexibility of such spaces proves to be particularly advantageous in addressing sparsity, edge-preserving and heterogeneous signal/noise statistics, while remaining efficient and stable from an optimisation perspective. We numerically validate this extensively on 1D/2D exemplar inverse problems (deconvolution, mixed denoising, CT reconstruction). The second part of the thesis focuses on off-the-grid Poisson inverse problems formulated within the space of Radon measures. Our contribution consists in the modelling of a variational model which couples a Kullback-Leibler data term with the Total Variation regularisation of the desired measure (that is, a weighted sum of Diracs) together with a non-negativity constraint. A detailed study of the optimality conditions and of the corresponding dual problem is carried out and an improved version of the Sliding Franke-Wolfe algorithm is used for computing the numerical solution efficiently. To mitigate the dependence of the results on the choice of the regularisation parameter, an homotopy strategy is proposed for its automatic tuning, where, at each algorithmic iteration checks whether an informed stopping criterion defined in terms of the noise level is verified and update the regularisation parameter accordingly. Several numerical experiments are reported on both simulated 2D and real 3D fluorescence microscopy data.Cette thèse porte sur la modélisation, l'analyse théorique et l'implémentation numérique d'algorithmes d'optimisation pour la résolution de problèmes inverses d'imagerie (par exemple, la reconstruction d'images en tomographie et la déconvolution d'images en microscopie) dans des espaces de Banach non standard. Elle est divisée en deux parties: dans la première, nous considérons le cadre des espaces de Lebesgue à exposant variable L^{p(cdot)} afin d'améliorer l'adaptabilité de la solution par rapport aux reconstructions obtenues dans le cas standard d'espaces d'Hilbert; dans la deuxième partie, nous considérons une modélisation dans l'espace des mesures de Radon pour éviter les biais dus à la discrétisation observés dans les méthodes de régularisation parcimonieuse. Plus en détail, la première partie explore à la fois des algorithmes d'optimisation lisse et non lisse dans les espaces L^{p(cdot)} réflexifs, qui sont des espaces de Banach dotés de la norme dite de Luxemburg. Comme premier résultat, nous fournissons une expression des cartes de dualité dans ces espaces, qui sont un ingrédient essentiel pour la conception d'algorithmes itératifs efficaces. Pour surmonter la non-séparabilité de la norme sous-jacente et les temps de calcul conséquents, nous étudions ensuite la classe des fonctions modulaires qui étendent directement la puissance (non homogène) p > 1 des normes L^p au cadre L^{p(cdot)}. En termes de fonctions modulaires, nous formulons des analogues des cartes duales qui sont plus adaptées aux algorithmes d'optimisation lisse et non lisse en raison de leur séparabilité. Nous étudions alors des algorithmes de descente de gradient (à la fois déterministes et stochastiques) basés sur les fonctions modulaires, ainsi que des algorithmes modulaires de gradient proximal dans L^{p(cdot)}, dont nous prouvons la convergence en termes des valeurs de la fonctionnelle. La flexibilité de ces espaces s'avère particulièrement avantageuse pour la modélisation de la parcimonie et les statistiques hétérogènes du signal/bruit, tout en restant efficace et stable d'un point de vue de l'optimisation. Nous validons cela numériquement de manière approfondie sur des problèmes inverses exemplaires en une/deux dimension(s) (déconvolution, débruitage mixte, tomographie). La deuxième partie de la thèse se concentre sur la formulation des problèmes inverses avec un bruit de Poisson formulés dans l'espace des mesures de Radon. Notre contribution consiste en la modélisation d'un modèle variationnel qui couple un terme de données de divergence de Kullback-Leibler avec la régularisation de la Variation Totale de la mesure souhaitée (une somme pondérée de Diracs) et une contrainte de non-négativité. Nous proposons une étude détaillée des conditions d'optimalité et du problème dual correspondant. Nous considérons une version améliorée de l'algorithme de Sliding Franke-Wolfe pour calculer la solution numérique du problème de manière efficace. Pour limiter la dépendance des résultats du choix du paramètre de régularisation, nous considérons une stratégie d'homotopie pour son ajustement automatique où à chaque itération algorithmique, on vérifie si un critère d'arrêt défini en termes du niveau de bruit est vérifié et on met à jour le paramètre de régularisation en conséquence. Plusieurs expériences numériques sont rapportées à la fois sur des données de microscopie de fluorescence simulées en 1D/2D et réelles en 3D
Algorithmes d'optimisation dans des espaces de Banach non standard pour problèmes inverses en imagerie
No full textThis thesis focuses on the modelling, the theoretical analysis and the numerical implementation of advanced optimisation algorithms for imaging inverse problems (e.g,., image reconstruction in computed tomography, image deconvolution in microscopy imaging) in non-standard Banach spaces. It is divided into two parts: in the former, the setting of Lebesgue spaces with a variable exponent map L^{p(cdot)} is considered to improve adaptivity of the solution with respect to standard Hilbert reconstructions; in the latter a modelling in the space of Radon measures is used to avoid the biases observed in sparse regularisation methods due to discretisation.In more detail, the first part explores both smooth and non-smooth optimisation algorithms in reflexive L^{p(cdot)} spaces, which are Banach spaces endowed with the so-called Luxemburg norm. As a first result, we provide an expression of the duality maps in those spaces, which are an essential ingredient for the design of effective iterative algorithms.To overcome the non-separability of the underlying norm and the consequent heavy computation times, we then study the class of modular functionals which directly extend the (non-homogeneous) p-power of L^p-norms to the general L^{p(cdot)}. In terms of the modular functions, we formulate handy analogues of duality maps, which are amenable for both smooth and non-smooth optimisation algorithms due to their separability. We thus study modular-based gradient descent (both in deterministic and in a stochastic setting) and modular-based proximal gradient algorithms in L^{p(cdot)}, and prove their convergence in function values. The spatial flexibility of such spaces proves to be particularly advantageous in addressing sparsity, edge-preserving and heterogeneous signal/noise statistics, while remaining efficient and stable from an optimisation perspective. We numerically validate this extensively on 1D/2D exemplar inverse problems (deconvolution, mixed denoising, CT reconstruction). The second part of the thesis focuses on off-the-grid Poisson inverse problems formulated within the space of Radon measures. Our contribution consists in the modelling of a variational model which couples a Kullback-Leibler data term with the Total Variation regularisation of the desired measure (that is, a weighted sum of Diracs) together with a non-negativity constraint. A detailed study of the optimality conditions and of the corresponding dual problem is carried out and an improved version of the Sliding Franke-Wolfe algorithm is used for computing the numerical solution efficiently. To mitigate the dependence of the results on the choice of the regularisation parameter, an homotopy strategy is proposed for its automatic tuning, where, at each algorithmic iteration checks whether an informed stopping criterion defined in terms of the noise level is verified and update the regularisation parameter accordingly. Several numerical experiments are reported on both simulated 2D and real 3D fluorescence microscopy data.Cette thèse porte sur la modélisation, l'analyse théorique et l'implémentation numérique d'algorithmes d'optimisation pour la résolution de problèmes inverses d'imagerie (par exemple, la reconstruction d'images en tomographie et la déconvolution d'images en microscopie) dans des espaces de Banach non standard. Elle est divisée en deux parties: dans la première, nous considérons le cadre des espaces de Lebesgue à exposant variable L^{p(cdot)} afin d'améliorer l'adaptabilité de la solution par rapport aux reconstructions obtenues dans le cas standard d'espaces d'Hilbert; dans la deuxième partie, nous considérons une modélisation dans l'espace des mesures de Radon pour éviter les biais dus à la discrétisation observés dans les méthodes de régularisation parcimonieuse. Plus en détail, la première partie explore à la fois des algorithmes d'optimisation lisse et non lisse dans les espaces L^{p(cdot)} réflexifs, qui sont des espaces de Banach dotés de la norme dite de Luxemburg. Comme premier résultat, nous fournissons une expression des cartes de dualité dans ces espaces, qui sont un ingrédient essentiel pour la conception d'algorithmes itératifs efficaces. Pour surmonter la non-séparabilité de la norme sous-jacente et les temps de calcul conséquents, nous étudions ensuite la classe des fonctions modulaires qui étendent directement la puissance (non homogène) p > 1 des normes L^p au cadre L^{p(cdot)}. En termes de fonctions modulaires, nous formulons des analogues des cartes duales qui sont plus adaptées aux algorithmes d'optimisation lisse et non lisse en raison de leur séparabilité. Nous étudions alors des algorithmes de descente de gradient (à la fois déterministes et stochastiques) basés sur les fonctions modulaires, ainsi que des algorithmes modulaires de gradient proximal dans L^{p(cdot)}, dont nous prouvons la convergence en termes des valeurs de la fonctionnelle. La flexibilité de ces espaces s'avère particulièrement avantageuse pour la modélisation de la parcimonie et les statistiques hétérogènes du signal/bruit, tout en restant efficace et stable d'un point de vue de l'optimisation. Nous validons cela numériquement de manière approfondie sur des problèmes inverses exemplaires en une/deux dimension(s) (déconvolution, débruitage mixte, tomographie). La deuxième partie de la thèse se concentre sur la formulation des problèmes inverses avec un bruit de Poisson formulés dans l'espace des mesures de Radon. Notre contribution consiste en la modélisation d'un modèle variationnel qui couple un terme de données de divergence de Kullback-Leibler avec la régularisation de la Variation Totale de la mesure souhaitée (une somme pondérée de Diracs) et une contrainte de non-négativité. Nous proposons une étude détaillée des conditions d'optimalité et du problème dual correspondant. Nous considérons une version améliorée de l'algorithme de Sliding Franke-Wolfe pour calculer la solution numérique du problème de manière efficace. Pour limiter la dépendance des résultats du choix du paramètre de régularisation, nous considérons une stratégie d'homotopie pour son ajustement automatique où à chaque itération algorithmique, on vérifie si un critère d'arrêt défini en termes du niveau de bruit est vérifié et on met à jour le paramètre de régularisation en conséquence. Plusieurs expériences numériques sont rapportées à la fois sur des données de microscopie de fluorescence simulées en 1D/2D et réelles en 3D
Optimisation algorithms in non-standard Banach spaces for inverse problems in imaging
No full textCette thèse porte sur la modélisation, l'analyse théorique et l'implémentation numérique d'algorithmes d'optimisation pour la résolution de problèmes inverses d'imagerie (par exemple, la reconstruction d'images en tomographie et la déconvolution d'images en microscopie) dans des espaces de Banach non standard. Elle est divisée en deux parties: dans la première, nous considérons le cadre des espaces de Lebesgue à exposant variable L^{p(cdot)} afin d'améliorer l'adaptabilité de la solution par rapport aux reconstructions obtenues dans le cas standard d'espaces d'Hilbert; dans la deuxième partie, nous considérons une modélisation dans l'espace des mesures de Radon pour éviter les biais dus à la discrétisation observés dans les méthodes de régularisation parcimonieuse. Plus en détail, la première partie explore à la fois des algorithmes d'optimisation lisse et non lisse dans les espaces L^{p(cdot)} réflexifs, qui sont des espaces de Banach dotés de la norme dite de Luxemburg. Comme premier résultat, nous fournissons une expression des cartes de dualité dans ces espaces, qui sont un ingrédient essentiel pour la conception d'algorithmes itératifs efficaces. Pour surmonter la non-séparabilité de la norme sous-jacente et les temps de calcul conséquents, nous étudions ensuite la classe des fonctions modulaires qui étendent directement la puissance (non homogène) p > 1 des normes L^p au cadre L^{p(cdot)}. En termes de fonctions modulaires, nous formulons des analogues des cartes duales qui sont plus adaptées aux algorithmes d'optimisation lisse et non lisse en raison de leur séparabilité. Nous étudions alors des algorithmes de descente de gradient (à la fois déterministes et stochastiques) basés sur les fonctions modulaires, ainsi que des algorithmes modulaires de gradient proximal dans L^{p(cdot)}, dont nous prouvons la convergence en termes des valeurs de la fonctionnelle. La flexibilité de ces espaces s'avère particulièrement avantageuse pour la modélisation de la parcimonie et les statistiques hétérogènes du signal/bruit, tout en restant efficace et stable d'un point de vue de l'optimisation. Nous validons cela numériquement de manière approfondie sur des problèmes inverses exemplaires en une/deux dimension(s) (déconvolution, débruitage mixte, tomographie). La deuxième partie de la thèse se concentre sur la formulation des problèmes inverses avec un bruit de Poisson formulés dans l'espace des mesures de Radon. Notre contribution consiste en la modélisation d'un modèle variationnel qui couple un terme de données de divergence de Kullback-Leibler avec la régularisation de la Variation Totale de la mesure souhaitée (une somme pondérée de Diracs) et une contrainte de non-négativité. Nous proposons une étude détaillée des conditions d'optimalité et du problème dual correspondant. Nous considérons une version améliorée de l'algorithme de Sliding Franke-Wolfe pour calculer la solution numérique du problème de manière efficace. Pour limiter la dépendance des résultats du choix du paramètre de régularisation, nous considérons une stratégie d'homotopie pour son ajustement automatique où à chaque itération algorithmique, on vérifie si un critère d'arrêt défini en termes du niveau de bruit est vérifié et on met à jour le paramètre de régularisation en conséquence. Plusieurs expériences numériques sont rapportées à la fois sur des données de microscopie de fluorescence simulées en 1D/2D et réelles en 3D.This thesis focuses on the modelling, the theoretical analysis and the numerical implementation of advanced optimisation algorithms for imaging inverse problems (e.g,., image reconstruction in computed tomography, image deconvolution in microscopy imaging) in non-standard Banach spaces. It is divided into two parts: in the former, the setting of Lebesgue spaces with a variable exponent map L^{p(cdot)} is considered to improve adaptivity of the solution with respect to standard Hilbert reconstructions; in the latter a modelling in the space of Radon measures is used to avoid the biases observed in sparse regularisation methods due to discretisation.In more detail, the first part explores both smooth and non-smooth optimisation algorithms in reflexive L^{p(cdot)} spaces, which are Banach spaces endowed with the so-called Luxemburg norm. As a first result, we provide an expression of the duality maps in those spaces, which are an essential ingredient for the design of effective iterative algorithms.To overcome the non-separability of the underlying norm and the consequent heavy computation times, we then study the class of modular functionals which directly extend the (non-homogeneous) p-power of L^p-norms to the general L^{p(cdot)}. In terms of the modular functions, we formulate handy analogues of duality maps, which are amenable for both smooth and non-smooth optimisation algorithms due to their separability. We thus study modular-based gradient descent (both in deterministic and in a stochastic setting) and modular-based proximal gradient algorithms in L^{p(cdot)}, and prove their convergence in function values. The spatial flexibility of such spaces proves to be particularly advantageous in addressing sparsity, edge-preserving and heterogeneous signal/noise statistics, while remaining efficient and stable from an optimisation perspective. We numerically validate this extensively on 1D/2D exemplar inverse problems (deconvolution, mixed denoising, CT reconstruction). The second part of the thesis focuses on off-the-grid Poisson inverse problems formulated within the space of Radon measures. Our contribution consists in the modelling of a variational model which couples a Kullback-Leibler data term with the Total Variation regularisation of the desired measure (that is, a weighted sum of Diracs) together with a non-negativity constraint. A detailed study of the optimality conditions and of the corresponding dual problem is carried out and an improved version of the Sliding Franke-Wolfe algorithm is used for computing the numerical solution efficiently. To mitigate the dependence of the results on the choice of the regularisation parameter, an homotopy strategy is proposed for its automatic tuning, where, at each algorithmic iteration checks whether an informed stopping criterion defined in terms of the noise level is verified and update the regularisation parameter accordingly. Several numerical experiments are reported on both simulated 2D and real 3D fluorescence microscopy data
Decreased osteocyte viability in multiple myeloma patients: osteolytic bone lesions, apoptosis and their potential role in bone remodeling alterations.
No full textOsteocytes seem to regulate bone remodelling by different manners including apoptosis. A reduction of osteocyte viability (OC-V) was shown in osteoporotic bone. Osteolysis/osteoporosis, induced by multiple myeloma (MM), are characterized by severely imbalanced uncoupled bone remodelling due to increased osteoclastogenesis and suppressed osteoblast differentiation occurring close to MM cell infiltration. The aim of this study is to investigate the eventual involvement of osteocytes in bone remodelling alterations occurring in MM patients. Iliac crest biopsies were taken from 34 patients with MM (52% of which displayed osteolytic bone lesions), 10 with monoclonal gammopathy of uncertain significance (MGUS) and 10 without haematological malignancies/osteoporosis/metabolic bone diseases. Viable osteocytes and degenerated or apoptotic osteocytes/empty lacunae were evaluated on 500 lacunae per histological section. Significant reductions of OC-V in MM patients were found compared to healthy controls, whereas not statistical significance in OC-V was observed between MM and MGUS. Death osteocytes/empty lacunae number was significantly increased in MM vs. controls but not as compared to MGUS. Concerning the skeletal involvement, in MM patients either OC-V percentage was significantly lower in osteolytic vs. non-osteolytic patients or the number of dead osteocytes/empty lacunae was higher in osteolytic vs. non-osteolytic patients. Monolayers were also performed of human preosteocytes incubated with/without conditioned media (CM) taken from human myeloma cell lines (HMCLs) or co-cultured with them, and TEM observations showed dead cells in those monolayers treated with HMCL-CM or co-cultured with HMCLs as compared to non treated cells. In CM of preosteocytes co-cultured with HMCLs significantly increased CD14+-derived osteoclastogenesis occurs, evaluated by TRAP staining and pit-forming assay. Our data demonstrate that MM bone is characterized by reduction of OC-V; the increase of osteocyte death (apoptosis/degeneration) in relation to the presence of bone lesions may represent a triggering event to osteoclast recruitment
O impacto da humanização da comunicação nas redes sociais e mídias sociais
No full textEsta pesquisa tem por objetivo identificar como a humanização da comunicação pode construir uma audiência conectada e engajada nas empresas, para além de uma estratégia de vendas nas redes sociais e mídias sociais. Para tanto, se vale de uma metodologia quantitativa de nível descritivo, procedimentos bibliográficos, uma amostragem não probabilística por conveniência com coleta de dados por meio de questionário estruturado e análise de dados por estatística descritiva. Os resultados evidenciam que a comunicação humanizada chama a atenção do consumidor, pois trata de uma filosofia de trabalho centrada nas motivações e objeções de ser humano, ou seja, a empresa pensa, age e desfruta, com pouco esforço, de conexão e lucratividade ao exercitar a cidadania corporativa no plano social-digital e sistêmico. [resumo fornecido pelo autor]This research aims to identify how the humanization of communication can build a connected and engaged audience in companies, in addition to a sales strategy on social networks and social media. For that, it uses a quantitative methodology of descriptive level, bibliographic procedures, a non-probabilistic organized by tolerant with data collection through secure standards and data analysis by descriptive statistics. The results show that humanized communication draws the attention of the consumer, as it deals with a work philosophy centered on the motivations and objections of human beings, that is, the company thinks, acts and enjoys, with little effort, connection and profitability at the corporate citizenship on the social-digital and systemic level. [resumo fornecido pelo autor
