340 research outputs found
Lenguaje inclusivo. Dra. María Marta García Negroni
No full textConversamos con la Dra. María Marta García Negroni, investigadora del CONICET, sobre el lenguaje y sus posibilidades de transformación
Krill oil, vitamin D and Lactobacillus reuteri cooperate to reduce gut inflammation
No full textCurrent research into original therapies to treat intestinal inflammation is focusing on no-drug therapies. KLD is a mixture of krill oil (KO), probiotic Lactobacillus reuteri (LR), and vitamin D (VitD3). The aim of this study was to assess in vitro and in vivo the potential cooperative effects of KLD in reducing gut inflammation. Colorectal adenocarcinoma cell lines, CACO2 and HT29, and C57BL/6 mice were used for in vitro and in vivo analyses, respectively. Cells were exposed to cytomix (interferon gamma + tumour necrosis factor alpha (TNF-a)) to induce inflammation or co-exposed to cytomix and KO, LR and VitD3 alone or to cytomix and KLD. Animals were treated for 7 days with dextran sodium sulphate (DSS) to induce colitis or with DSS and KLD. In vitro assays: F-actin expression was analysed by immunofluorescence; scratch test and trans-epithelial electric resistance test were performed to measure wound healing; adhesion/invasion assays of adhesive and invasive Escherichia coli (AIEC) bacteria were made; mRNA expression of TNF-α, interleukin (IL)-8 and vitamin D receptor (VDR) was detected by quantitative PCR. In vivo assays: body weight, clinical score, histological score and large intestine weight and length were estimated; mRNA expression of TNF-α, IL-1ß, IL-6, IL-10 by quantitative PCR; VDR expression was detected by quantitative PCR and immunohistochemistry. In vitro: KLD restores epithelial cell-cell adhesion and mucosal healing during inflammation, while decreases the adhesiveness and invasiveness of AIEC bacteria and TNF-α and IL-8 mRNA expression and increases VDR expression. In vivo: KLD significantly improves body weight, clinical score, histological score and large intestine length of mice with DSS-induced colitis and reduces TNF-α, IL-1ß and IL-6 mRNA levels, while increases IL-10 mRNA and VDR levels. KLD has significant effects on the intestinal mucosa, strongly decreasing inflammation, increasing epithelial restitution and reducing pathogenicity of harmful commensal bacteria. © 2018 Wageningen Academic Publishers
Un edificio residenziale di epoca imperiale nei pressi dell'Aurelia Vetus
No full textIn 2009, during archaeological excavations along via del Fontanile Arenato, a modern street
tracing out an ancient back road from via Aurelia Vetus and situated in the Roman suburb,
some remains of an ancient building were brought to light, dating from a large span. In particular
were found a sandstone quarry and a well, in use between the end of the Republican age and the
beginning of the Imperial period, and some rooms of a large building with different phases going on
at least to V century AD. The quality of the decoration (an opus sectile "oor, a lot of decorated
plaster and pottery fragments, etc.) suggests that the structures were part of a villa with a quite
high aesthetic value, belonged to a rich owner
Evaluation of the efficiency of an assay procedure for gangliosides in human serum
No full textAn efficiency assessment of a ganglioside assay procedure was carried out on human serum gangliosides from healthy subjects of different sex and age. The analysis of the gangliosides, extracted with chloroform/methanol and purified by lipid partitioning, ion exchange column chromatographic separation and desalting procedures as described by Senn et al. (1989) Eur J Biochem 181: 657-62, was performed by HPTLC followed by densitometric quantification. The yield of the procedure, expressed as radioactivity recovery, was determined by adding GM3 ganglioside, tritium labelled at the sialic acid acetyl group and at the C3 position of sphingosine, to the lyophilized serum or by associating it with the serum lipoproteins. Tn spite of the fact that the extraction and purification procedures were performed exactly as described we found the radioactivity recovery to be variable (25-50%) and much lower than that proposed. Much of the radioactivity was found in the organic phase after lipid partitioning, whilst all the ganglioside purification steps led to some further loss. After the introduction of some modifications to the procedure the recovery improved, reaching 67-79%. The analyses on 33 samples of 5 ml showed a human serum ganglioside content of about 10 nmol ml(-1) (as corrected for the recovery), and confined that GM3 ganglioside is the main component of the total serum ganglioside mixture
Analyzing the impact of splicing artifacts in partially fake speech signals
No full textSpeech deepfake detection has recently gained significant attention within the multimedia forensics community. Related issues have also been explored, such as the identification of partially fake signals, i.e., tracks that include both real and fake speech segments. However, generating high-quality spliced audio is not as straightforward as it may appear. Spliced signals are typically created through basic signal concatenation. This process could introduce noticeable artifacts that can make the generated data easier to detect. We analyze spliced audio tracks resulting from signal concatenation and investigate their artifacts, study the causes behind their presence, and assess whether such artifacts introduce any bias in existing datasets. Our findings reveal that by analyzing splicing artifacts, we can achieve a detection EER of 6.16% and 7.36% on PartialSpoof and HAD datasets, respectively, without needing to train any detector. These results underscore the complexities of generating reliable spliced audio data and lead to discussions that can help improve future research in this area
Activity of 3-ketosphinganine synthase during differentiation and aging of neuronal cells in culture
No full textChanges in the enzyme 3-ketosphinganine synthase activity in rat cerebellar granule cells in culture were studied during differentiation and aging. The enzyme activity with palmitoyl-CoA and stearoyl-CoA, precursors of, respectively, C18-sphinganine and C20-sphinganine, was studied on the total cell homogenate using radioactive serine. The enzyme assay was performed by thin-layer chromatography (TLC) separation of the enzyme reaction mixture, and the resultant radioactive 3-ketosphinganine was identified by chromatographic comparison with a chemically synthesized 3-ketosphinganine, and quantified by determination of the TLC radioactivity distribution on the basis of the radioactivity content of cell lipid extract that was determined by scintillation counting. Using palmitoyl-CoA, the enzyme activity progressively increased from 40 to 54 pmol of 3-ketosphinganine/mg cell DNA per min in the first 8 days and then progressively decreased, and was 39 pmol of C18-(3-ketosphinganine)/mg cell DNA per mill at day 22 in culture. For stearoyl-Coh the enzyme activity was very low at day one and then increased to a constant value of about 15 pmol of C20-(3-ketosphinganine) /mg cell DNA per min. These results are in good agreement with the finding that the ganglioside species that contain C18-sphingosine increase during cell differentiation and remain constant during cell aging, while the ganglioside species that contain C20-sphingosine continuously increase during both cell differentiation and aging
Localization of Synthetic Manipulations in Western Blot Images
No full textRecent breakthroughs in deep learning and generative systems have significantly fostered the creation of synthetic media, as well as the local alteration of real content via the insertion of highly realistic synthetic manipulations. Local image manipulation, in particular, poses serious challenges to the integrity of digital content and societal trust. This problem is not only confined to multimedia data, but also extends to biological images included in scientific publications, like images depicting Western blots. In this work, we address the task of localizing synthetic manipulations in Western blot images. To discriminate between pristine and synthetic pixels of an analyzed image, we propose a synthetic detector that operates on small patches extracted from the image. We aggregate patch contributions to estimate a tampering heatmap, highlighting synthetic pixels out of pristine ones. Our methodology proves effective when tested over two manipulated Western blot image datasets, one altered automatically and the other manually by exploiting advanced AI-based image manipulation tools that are unknown at our training stage. We also explore the robustness of our method over an external dataset of other scientific images depicting different semantics, manipulated through unseen generation techniques. We release our experimental code and the manipulated datasets at https://github.com/polimiispl/western-blot-synthetic-manipulation-localization
Atlas portátil de la infancia: Cartas extraordinarias de María Negroni
No full textArgentinian writer María Negroni has been working in recent years on a series of books which are simultaneously coherent and made of splitting fragments. One of the latest, Cartas extraordinarias (Extraordinary Letters) (2013), presents some challenges with respect to genre definitions. It could be read following Florencia Garramuño’s ideas and what she describes as the new forms of “no specificity” in art. The simplest definition about Negroni’s book is that it is an anthology of apocryphal letters: the author presents false letters of writers we used to read in our childhood. But that does not explain the kind of genre we are dealing with. We probably should read this book as a volume of short stories, even if it is not presented as such. But that is not the only challenge. The book works with notions such as collection and montage, anachronism and false adscriptions. Childhood is alluded through the subject but also through the form and materiality of the volume. Negroni works together, hand in hand, with the Uruguayan painter Fidel Sclavo. They reach a dialogue between image and written text, which is disconcerting and ambiguous, beautiful and nostalgic at the same time.La autora argentina María Negroni viene hilvanado una obra que es a la vez fragmentaria y coherente. Uno de sus textos más recientes es Cartas extraordinarias (2013) que propone algunos desafíos a la hora de las definiciones genéricas. Plantea la cuestión de lo que Florencia Garramunño estudia como lo “no específico” en los nuevos modos de organización de lo sensible. Lo más simple que se puede decir de este libro es que se presenta como una antología de cartas apócrifas, pero eso no termina de definirlo. ¿Es un volumen de cuentos o una novela epistolar? Este no es el único juego que propone el libro, que por más de una razón alude al universo de la infancia. Se trata de una colección que parafrasea otra colección de los autores que leíamos en la niñez. A su vez, el libro llama la atención porque está ilustrado, en un trabajo que hace uso del montaje. Es el resultado del “encuentro feliz” entre la obra de Negroni y la del artista uruguayo Fidel Sclavo. Ambos logran un diálogo inquietante y complejo, mediante el cual se retorna a las topografías de la infancia para hablar de nuevas modalidades estéticas en las que se borronean las fronteras genéricas
Freeze and learn: continual learning with selective freezing for speech deepfake detection
No full textIn speech deepfake detection, one of the critical aspects is developing detectors able to generalize on unseen data and distinguish fake signals across different datasets. Common approaches to this challenge involve incorporating diverse data into the training process or fine-tuning models on unseen datasets. However, these solutions can be computationally demanding and may lead to the loss of knowledge acquired from previously learned data. Continual learning techniques offer a potential solution to this problem, allowing the models to learn from unseen data without losing what they have already learned. Still, the optimal way to apply these algorithms for speech deepfake detection remains unclear, and we do not know which is the best way to apply these algorithms to the developed models. In this paper we address this aspect and investigate whether, when retraining a speech deepfake detector, it is more effective to apply continual learning across the entire model or to update only some of its layers while freezing others. Our findings, validated across multiple models, indicate that the most effective approach among the analyzed ones is to update only the weights of the initial layers, which are responsible for processing the input features of the detector
- …
