Producción de Investigación

Permanent URI for this community

https://repositorio.uai.edu.ar/handle/123456789/14

Browse

Browsing Producción de Investigación by Title

  • Item
    A baseline underwater soundscape of an intensely human-exploited estuarine and the effects of vessel traffic sound
    (Sociedad Argentina de Informática e Investigación Operativa (SADIO), 2024-8-28) Pons, Juan ; Uibrig, Román ; Molina, Juan ; Pons, Claudia Fabiana
    In this article we studied the anthropically impacted natural environ mental sound in the port of Bahía Blanca, located in the southern province of Buenos Aires, Argentina. To acquire the acoustic signals, an omni-directional passive hydrophone was used. The acoustic signals were analysed using scripts implemented in the R programming language. Temporal series without maritime traffic were used as a baseline to describe the soundscape in the harbour area by estimating its power spectral density (PSD). Subsequently, the acoustic environ ment was analysed with the presence of two man-made acoustic sources “boat” and “ship” in the vicinity. Finally, the calculated normal soundscape level in the harbour has a magnitude of 116.25 dB re 1 µPa.
  • Item
    A cloud powered relaxed heterogeneous distributed shared memory system
    (Universidad Nacional del Centro de la Provincia de Buenos Aires, 2018-10) Teragni, Matías Iván ; Zabala, Gonzalo Esteban ; Blanco, Sebastián Gabriel
    Distributed systems allow the existence of impressive pieces of software, but usually impose strict restrictions on the implementation language and model. We propose a distribution system model that enables the incorporation of any hardware device connected to the internet as its nodes, and places no restriction on the execution engine, allowing the transparent incorporation of any existing codebase into a Distributed Shared Memory.Cloud Computing,
  • Item
    A Co-Training Model Based in Learning Transfer for the Classification of Research Papers
    (IEEE, 2024-10-9) Cevallos Culqui, Alex ; Pons, Claudia Fabiana ; Rodríguez, Gustavo
    A multitude of scholarly papers can be accessed online, and their continual growth poses challenges in categorization. In diverse academic fields, organizing these documents is important, as it assists institutions, journals, and scholars in structuring their content to improve the visibility of research. In this study, we propose a co-training model based on transfer learning to classify papers according to institutional research lines. We utilize cotraining text processing techniques to enhance model learning through transformers, enabling the identification of trends and patterns in document texts. The model is structured with two views (titles and abstracts) for data preprocessing and training. Each input employs different document representation techniques that augment its training using BERT's pre-trained scheme. For evaluating the proposed model, a dataset comprising 898 institutional papers is compiled. These documents undergo classification prediction in five or eleven classes, and the model performance is compared with individually trained models from each view using the BART pre-trained scheme and combined models. The best precision level of 0,87 has been achieved, compared to BERT pre-trained model's metric of 0,78 (five classes). These findings suggest that co-training models can be a valuable approach to improve the predictive performance of text classification.
  • Item
    A flexible and expressive formalism to specify metamorphic properties for BIG DATA systems validation
    (Universidad Nacional de La Rioja - EUDELAR, 2023-1) Asteasuain, Fernando
    BIG DATA systems represent a huge challenge for software engineering validations tasks since they have been classified as "non testable". Metamorphic Relationships (MR) have been proposed as a technique to overcome this problem. These relationships establish interactions between data that can be used to validate the expected behavior of the system. However, the process of exploring and defining MRs is a very arduous one, and an expressive and flexible specification language is needed to denote them. In this work we show how the Feather Weight Visual Scenarios (FVS) framework can be seen as an appealing tool to specify MRs. We exploit FVS features to model complex MR interactions and analysis, allowing the possibility to perform non trivial operations between MRs such as refinement and consistency checking. FVS is shown in action by introducing a proof of concept example focused on a machine learning system over biology cell images.
  • Item
    A functional analysis of the cyclophilin repertoire in the protozoan parasite Trypanosoma cruzi
    (MDPI, 2018-10-31) Fuchs, Alicia Graciela ; Perrone, Alina E ; Milduberger, Natalia A. ; Bustos, Patricia L. ; Bua, Jaqueline
    Trypanosoma cruzi is the etiological agent of Chagas disease. It affects eight million people worldwide and can be spread by several routes, such as vectorborne transmission in endemic areas and congenitally, and is also important in non-endemic regions such as the United States and Europe due to migration from Latin America. Cyclophilins (CyPs) are proteins with enzymatic peptidyl-prolyl isomerase activity (PPIase), essential for protein folding in vivo. Cyclosporin A (CsA) has a high binding affinity for CyPs and inhibits their PPIase activity. CsA has proved to be a parasiticidal drug on some protozoa, including T. cruzi. In this review, we describe the T. cruzi cyclophilin gene family, that comprises 15 paralogues. Among the proteins isolated by CsA-affinity chromatography, we found orthologues of mammalian CyPs. TcCyP19, as the human CyPA, is secreted to the extracellular environment by all parasite stages and could be part of a complex interplay involving the parasite and the host cell. TcCyP22, an orthologue of mitochondrial CyPD, is involved in the regulation of parasite cell death. Our findings on T. cruzi cyclophilins will allow further characterization of these processes, leading to new insights into the biology, the evolution of metabolic pathways, and novel targets for anti-T. cruzi contro
  • Item
    A homolog of cyclophilin D is expressed in Trypanosoma cruzi and is involved in the oxidative stress–damage response
    (Cell Death Differentiation Association (ADMC), 2017-2-6) Bustos, Patricia L. ; Volta, Viviana J. ; Perrone, Alina E ; Milduberger, Natalia A. ; Bua, Jaqueline
    Mitochondria have an important role in energy production, homeostasis and cell death. The opening of the mitochondrial permeability transition pore (mPTP) is considered one of the key events in apoptosis and necrosis, modulated by cyclophilin D (CyPD), a crucial component of this protein complex. In Trypanosoma cruzi, the protozoan parasite that causes Chagas disease, we have previously described that mitochondrial permeability transition occurs after oxidative stress induction in a cyclosporin A-dependent manner, a well-known cyclophilin inhibitor. In the present work, a mitochondrial parasite cyclophilin, named TcCyP22, which is homolog to the mammalian CyPD was identified. TcCyP22-overexpressing parasites showed an enhanced loss of mitochondrial membrane potential and loss of cell viability when exposed to a hydrogen peroxide stimulus compared with control parasites. Our results describe for the first time in a protozoan parasite that a mitochondrial cyclophilin is a component of the permeability transition pore and is involved in regulated cell death induced by oxidative stress
  • Item
    A Machine Learning Approach for Atrial Fibrillation Detection in Telemonitored Patients
    (Springer, Cham, 2024-5-31) Barrera, Pedro ; Vecino Schandy, Lorenza ; Bonomini, María Paula ; Mateos, Cristian ; Hirsch, Matías ; Grana, Lucas ; Liberczuk, Sergio Javier
    Atrial fibrillation (AF) is the most common type of cardiac arrhythmia. As it is typically asymptomatic, it often goes undiagnosed until major complications arise, such as stroke. Therefore, the development of rapid, economical, and widely accessible diagnostic tools for detecting AF at an early stage is crucial. Telemonitoring with machine learning-assisted devices shows promise in achieving this goal. This paper presents an algorithm that automatically detects AF in signals obtained by portable electrocardiographs connected to a telemonitoring platform via smartphones. The algorithm consists of three stages: a noise detection, ectopic beat removal and an AF detection. The noise detection involves analyzing the ECG signals using 5-s windows with a 1-s shift. A K-nearest neighbors (KNN) classifier predicts the presence or absence of noise in each window, allowing for the detection of noisy and non-noisy segments of the signal. The non-noisy segments are processed using a Pan-Tompkins algorithm to find the R peaks of the signal, and the corresponding RR interval series. Then ectopic beats are removed using an XGBoost classifier, generating the NN series. In the AF detection stage, X features are obtained from this series, which serve as input features of an XGBoost classifier that predicts the presence or absence of AF in the ECG signal. The algorithm was trained and tested using the Physionet Short Single-Lead AF Database (SSLAFDB) and achieved an accuracy of 90.87% and an F1-score of 90.91%. Further validation was performed by an external partner using two other databases, reporting an accuracy of 90.41% and 89.61% respectively.
  • Item
    A note on the well-posedness of control complex Ginzburg-Landau equations in Zhidkov spaces
    (Brazilian Society of Applied and Computational Mathematics (SBMAC), 2022) Besteiro, Agustín Tomás
    In this note, we consider the Complex Ginzburg-Landau equations with a bilinear control term in the real line. We prove well-posedness results concerned with the initial value problem for these equations in Zhidkov spaces using splitting methods.
  • Item
    A parallel tableau algorithm for BIG DATA verification
    (Universidad Nacional de La Matanza, 2020-10) Asteasuain, Fernando ; Rodríguez Caldeira, Luciana
    BIG DATA systems are becoming more and more present in our everyday life generating data and information that needs to be explored and analyzed. In this sense, formal verification tools and techniques must provide solutions to face with these new challenges since they been pointed out as one of the most needed software engineering activities to consolidate BIG DATA modern systems. In this work we present a parallel implementation of a tableau algorithm aiming to improve the performance of our formal verication scheme. The pursued objective behind this transformation is to adapt our framework to deal with BIG DATA systems.
  • Item
    A Sound and Correct Formalism to Specify, Verify and Synthesize Behavior in BIG DATA Systems
    (Springer, 2022-5-20) Asteasuain, Fernando ; Rodriguez Caldeira, Luciana
    In this work we consolidate our behavioral specification frame work based on the Feather Weight Visual Scenarios (FVS) language as a powerful tool to specify, verify and synthesize behavior for BIG DATA systems. We formally demonstrate that our approach is sound and cor rect end to end, including the latest extensions such as fluents and par tial specifications. In addition, our empirical validation is strengthen by adding new and complex case studies and incorporating, besides execu tion time, space exploration as a factor in the comparison with other approaches. We believe that the contributions introduced in this work aim to point up FVS as a solid tool to formally verify behavior in BIG DATA systems.
  • Item
    Aceptación e integración de ChatGPT en la educación universitaria: análisis de percepciones docentes en la carrera de Ingeniería Agronómica y estudiantiles en la asignatura Microbiología Agrícola y de los Bioinsumos de la UNNOBA
    (Facultad de Informática, Universidad Nacional de La Plata, 2025-12) De Benedetto, Juan Pablo ; Pons, Claudia Fabiana
    Este trabajo se inscribe en un perfil de investigación-acción, al implementar una intervención pedagógica en el aula y relevar sistemáticamente las percepciones de estudiantes y docentes sobre su desarrollo. En este marco, se analiza la aceptación e integración de ChatGPT como herramienta pedagógica en la educación universitaria, a partir de una experiencia concreta desarrollada en la Universidad Nacional del Noroeste de la Provincia de Buenos Aires (UNNOBA). Se aplicó una actividad práctica con estudiantes de Microbiología Agrícola y de los Bioinsumos, y una encuesta a docentes de la carrera de Ingeniería Agronómica. Los resultados indican que los estudiantes valoraron positivamente la utilidad, precisión y facilidad de uso de ChatGPT, destacando su aporte a la comprensión de conceptos, el análisis crítico y el trabajo en equipo. Por su parte, los docentes mostraron mayor cautela, con menor frecuencia de uso y preocupaciones vinculadas al plagio y a la dependencia tecnológica. Sin embargo, reconocen el potencial de la herramienta y expresaron interés en recibir capacitación. La investigación evidencia una brecha entre el entusiasmo estudiantil y la apropiación docente, y resalta la necesidad de acompañamiento institucional y formación crítica para una integración efectiva de la inteligencia artificial en el ámbito académico.
  • Item
    Adaptation and validation of the texas revised inventory of grief—present scale
    (Open Medical Publishing, 2022) Gruppi, Rocío Luana ; Yaccarini, Cecilia ; Freiberg Hoffmann, Agustín ; Futterman, Andrew ; Simkin, Hugo
    This study aims to evaluate the psychometric properties and internal consistency of the Spanish version of the Texas Revised Inventory of Grief- Present [TRIG-Present] in Buenos Aires, Argentina, which assesses a series of thoughts, emotions and behaviors in losses related to the present. A total of 285 adults participated in the study with ages between 18 and 80 years (M = 55.09, SD = 15.27) and both sexes (Men = 42.8%, Women = 57.2%). The three-factor model resulted in acceptable fit indices (TLI = .970; CFI = .976; SRMR = .064). The results indicated an acceptable internal consistency for Emotional Response (? = .850), Not Acceptance (? = .816) and Thought (? = .837). The spanish adaptation of the TRIG-Present presents 13 items proposed by the original authors.
  • Item
    Addressing the State Explosion Problem for Big Data Systems Formal Verication
    (Universidad Nacional de Luján, 2023-10-12) Asteasuain, Fernando
    The formal verification of BIG DATA systems remains as a challenging task to be addressed since a very large and complex state space describing the behavior of the system must be explored and verified. In particular, the state explosion problem arises as one of the most problematic issues to be faced against. Some approaches have leveraged on some architectural patterns used in BIG DATA system development, especially those focused on the MAP-REDUCE architecture. Taking this into consideration in this work we present VG-FVS, a new version of our framework FVS (Feather weight Visual Scenarios), which is specially developed to address the state explosion problem. This is achieved by integrating FVS with MaRDiGraS, a generic library which eases the state space exploration using a MAP-REDUCE software architecture. Empirical validation analyzing BIG DATA systems was carried on, showing promising results for our approach.
  • Item
    Adversarial image generation using genetic algorithms with black-box technique
    (Sociedad Argentina de Informática (SADIO), 2023-10-20) Pons, Claudia Fabiana ; Pérez, Gabriela
    Abstract. Convolutional neural networks are a technique that has demonstrated great success in computer vision tasks, such as image classification and object detection. Like any machine learning model, they have limitations and vulnerabilities that must be carefully considered for safe and effective use. One of the main limitations lies in their complexity and the difficulty of interpreting their internal workings, which can be exploited for malicious purposes. The goal of these attacks is to make deliberate changes to the input data in order to deceive the model and cause it to make incorrect decisions. These attacks are known as adversarial attacks. This work focuses on the generation of adversarial images using genetic algorithms for a convolutional neural network trained on the MNIST dataset. Several strategies are employed, including targeted and untargeted attacks, as well as the presentation of interpretable and non-interpretable images that are unrecognizable to humans but are misidentified and confidently classified by the network. The experiment demonstrates the ability to generate adversarial images in a relatively short time, highlighting the vulnerability of neural networks and the ease with which they can be deceived. These results underscore the importance of developing more secure and reliable artificial intelligence systems capable of resisting such attacks. .
  • Item
    Alcohol hangover induces nitric oxide metabolism changes by impairing NMDA receptor-PSD95-nNOS pathway
    (Elsevier, 2021-5-5) Karadayian, Analía G. ; Bustamante, Juanita ; Lores-Arnaiz, Silvia
    Alcohol hangover is defined as the combination of mental and physical symptoms experienced the day after a single episode of heavy drinking, starting when blood alcohol concentration approaches zero. We previously evidenced increments in free radical generation and an imbalance in antioxidant defences in non-synaptic mitochondria and synaptosomes during hangover. It is widely known that acute alcohol exposure induces changes in nitric oxide (NO) production and blocks the binding of glutamate to NMDAR in central nervous system. Our aim was to evaluate the residual effect of acute ethanol exposure (hangover) on NO metabolism and the role of NMDA receptor-PSD95-nNOS pathway in non-synaptic mitochondria and synaptosomes from mouse brain cortex. Results obtained for the synaptosomes fraction showed a 37% decrease in NO total content, a 36% decrease in NOS activity and a 19% decrease in nNOS protein expression. The in vitro addition of glutamate to synaptosomes produced a concentration-dependent enhancement of NO production which was significantly lower in samples from hangover mice than in controls for all the glutamate concentrations tested. A similar patter of response was observed for nNOS activity being decreased both in basal conditions and after glutamate addition. In addition, synaptosomes exhibited a 64% and 15% reduction in NMDA receptor subunit GluN2B and PSD-95 protein expression, respectively. Together with this, glutamate-induced calcium entry was significant decreased in synaptosomes from alcohol-treated mice. On the other hand, in non-synaptic mitochondria, no significant differences were observed in NO content, NOS activity or nNOS protein expression. The expression of iNOS remained unaltered in synaptosomes and non-synaptic mitochondria. Here we demonstrated that hangover effects on NO metabolism are strongly evidenced in synaptosomes probably due to a disruption in NMDAR/PSD- 95/nNOS pathway.
  • Item
    Alcohol hangover: impairments in behavior and bioenergetics in central nervous system
    (Biocell, 2016-4-16) Karadayian, Analía G. ; Bustamante, Juanita ; Lores-Arnaiz, Silvia
    Alcohol hangover (AH) is defined as the temporary state after alcohol binge-like drinking, starting when EtOH is absent in plasma. Results from our laboratory have shown behavioral impairments and mitochondrial dysfunction in an experimental model of AH in mice. Our model consisted in a single i.p. injection of EtOH (3.8 g/kg BW) or saline solution in male and female mice, sacrificing the animals 6 hours after injection. Motor and affective behavior together with mitochondrial function and free radical production were evaluated in brain cortex and cerebellum during AH. Results showed that hangover animals exhibited a significant reduction in neuromuscular coordination, motor strength and locomotion together with a loss of gait stability and walking deficiencies. Moreover, an increment in anxiety-like behavior together with fear-related phenotype and depression signs were observed. In relation to bioenergetics metabolism, AH induced a reduction in oxygen uptake, inhibition of respiratory complexes, changes in mitochondrial membrane permeability, decrease in transmembrane potential, increase in O2•- and H2O2 production and impairment in nitric oxide metabolism. All together our data suggest that the physiopathological state of AH involves behavioral impairments and mitochondrial dysfunction in mouse brain cortex and cerebellum showing the long lasting effects of acute EtOH exposure in CNS.
  • Item
    Algoritmos de Inteligencia Artificial para la Detección de Patologías Relacionadas con el Cáncer de Pulmón a través del Análisis de Imágenes utilizando Redes Neuronales Convolucionales y Data Augmentation: un mapeo sistemático de la literatura
    ( 2023-10-26) Ramirez Amador, Pablo
    El cáncer de pulmón es una de las principales causas de muerte en el mundo y su diagnóstico temprano es crucial para mejorar el pronóstico y la calidad de vida de los pacientes. Sin embargo, el proceso de interpretación de imágenes médicas para la detección del cáncer de pulmón es complejo y requiere de expertos capacitados. En este contexto, la inteligencia artificial (IA) y el aprendizaje profundo (DL) surgen como herramientas potenciales para automatizar y optimizar el análisis de imágenes. El objetivo de este trabajo es revisar las aplicaciones más recientes y relevantes de la IA y el DL en el campo de la radiología para la detección del cáncer de pulmón. Para ello, se realizó una búsqueda exhaustiva en bases de datos científicas como PubMed, IEEEXPLORE, Scopus y Web of Science y se seleccionaron 96 artículos publicados desde el año 2015 hasta la actualidad que abordan el uso de IA y DL en la ingeniería biomédica. Se enfatiza el uso de redes neuronales convolucionales (CNN) con transferencia de conocimiento y Data Augmentation como técnicas prometedoras para mejorar la precisión y la eficiencia del proceso de interpretación de imágenes. Los resultados muestran que el uso de IA y DL puede ofrecer una alternativa efectiva para el diagnóstico temprano del cáncer de pulmón, con una alta sensibilidad y especificidad. Sin embargo, también se identifican limitaciones y desafíos actuales que deben abordarse para garantizar su aplicación responsable y segura en la práctica clínica, tales como la falta de datos estandarizados, la explicabilidad de los modelos, la privacidad de los pacientes y las implicaciones éticas y sociales. Se concluye que el uso de IA y DL puede tener un impacto positivo en la atención al paciente con cáncer de pulmón, pero se requiere más investigación y regulación para asegurar su calidad y confiabilidad.
  • Item
    Alineación de glosarios específicos de dominio
    (Even3, Brazil., 2024-8-9) Grijalva, Paola ; Cornejo, Galo ; Antonelli, Ruben Leandro ; Thomas, Pablo
    Los glosarios son una parte importante de cualquier documento de requerimientos de software, ya que hace explícitos los términos técnicos en un dominio y proporciona definiciones, ayudando a mitigar la imprecisión y la ambigüedad. El Léxico Extendido del Lenguaje (LEL) es un glosario que describe el vocabulario del dominio caracterizando a cada expresión a través de dos atributos (a diferencia de un glosario tradicional que tiene una sola descripción). En la actualidad, es muy común la interoperación de los sistemas informáticos, en donde cierto sistema brinda servicios para que consuman otros sistemas. En este marco, es crucial comprender el lenguaje de ambos sistemas, para que la interacción sea efectiva. Este artículo propone un método de alineación de glosarios LEL del dominio. El método de alineación propuesto se basa en identificación de discrepancias y similitudes, para luego acordar el significado de los términos similares. Este articulo presenta una evaluación preliminar con resultados promisorios utilizando como caso de estudio un modelo de calidad para instituciones de educación superior.
  • Item
    An approach for Reverse Engineering from Web Applications into the Language of the Domain using the LEL Glossary
    (Even3, Brazil., 2024-8-9) Granizo Rodríguez, Angela Verónica ; Antonelli, Ruben Leandro ; Firmenich, Sergio ; Firmenich, Diego
    Requirement engineering plays a crucial role in the software lifecycle, since errors made in the requirements require significant effort to be corrected in later stages. The main source of requirements is people; however, it is common to analyze existing applications when developing new software. This is particularly the case in the process of reengineering. On the other hand, the language of the domain is essential to understanding the domain and thus comprehending the requirements. Language Extended Lexicon (LEL) is a structured glossary designed to capture this language. This paper proposes an approach for obtaining the language of an application domain from a web application using the LEL glossary. The process comprises three main activities: general analysis of the web application, domain language capture, and the verification of the generated domain language. Additionally, this paper describes a web browser extension tool designed to support the process. Finally, the paper presents the results of a preliminary evaluation with promising outcomes regarding the applicability of the approach.
  • Item
    An expressive and enriched specification language to synthezise behavior in BIG DATA systems
    (Universidad Nacional de Salta, 2021) Asteasuain, Fernando ; Rodríguez Caldeira, Luciana
    In this work we extend our behavioral speci_cation and controller synthesis framework FVS to deal with BIG DATA requirements. For one side, we enriched FVS expressive power by exhibiting how our language can handle uents and partial speci_cations. For the other side, we combined FVS with a parallel model checker in order to automatically obtain a controller given the behavior speci_cation. In this way, FVS can be presented as an attractive tool to formally verify and synthesize behavior for BIG DATA systems. Our approach is compared to other well known parallel tool analyzing a complex big data system.