Search results for: 177Lu labeled somatostatin analogues

Authors: Martin Cvek, Miroslav Mrlik, Michal Sedlacik, Tomas Plachy

Abstract:

Magnetorheological elastomers (MREs) are multi-phase composite materials containing micron-sized ferromagnetic particles dispersed in an elastomeric matrix. Their properties such as modulus, damping, magneto-striction, and electrical conductivity can be controlled by an external magnetic field and/or pressure. These features of the MREs are used in the development of damping devices, shock attenuators, artificial muscles, sensors or active elements of electric circuits. However, imperfections on the particle/matrix interfaces result in the lower performance of the MREs when compared with theoretical values. Moreover, magnetic particles are susceptible to corrosion agents such as acid rains or sea humidity. Therefore, the modification of particles is an effective tool for the improvement of MRE performance due to enhanced compatibility between particles and matrix as well as improvements of their thermo-oxidation and chemical stability. In this study, the carbonyl iron (CI) particles were controllably modified with poly(trimethylsilyloxyethyl methacrylate) (PHEMATMS) nano-grafts to develop magnetic core–shell structures exhibiting proper wetting with various elastomeric matrices resulting in improved performance within a frame of rheological, magneto-piezoresistance, pressure-piezoresistance, or radio-absorbing properties. The desired molecular weight of PHEMATMS nano-grafts was precisely tailored using surface-initiated atom transfer radical polymerization (ATRP). The CI particles were firstly functionalized using a 3-aminopropyltriethoxysilane agent, followed by esterification reaction with α-bromoisobutyryl bromide. The ATRP was performed in the anisole medium using ethyl α-bromoisobutyrate as a macroinitiator, N, N´, N´´, N´´-pentamethyldiethylenetriamine as a ligand, and copper bromide as an initiator. To explore the effect PHEMATMS molecular weights on final properties, two variants of core-shell structures with different nano-graft lengths were synthesized, while the reaction kinetics were designed through proper reactant feed ratios and polymerization times. The PHEMATMS nano-grafts were characterized by nuclear magnetic resonance and gel permeation chromatography proving information to their monomer conversions, molecular chain lengths, and low polydispersity indexes (1.28 and 1.35) as the results of the executed ATRP. The successful modifications were confirmed via Fourier transform infrared- and energy-dispersive spectroscopies while expected wavenumber outputs and element presences, respectively, of constituted PHEMATMS nano-grafts, were occurring in the spectra. The surface morphology of bare CI and their PHEMATMS-grafted analogues was further studied by scanning electron microscopy, and the thicknesses of grafted polymeric layers were directly observed by transmission electron microscopy. The contact angles as a measure of particle/matrix compatibility were investigated employing the static sessile drop method. The PHEMATMS nano-grafts enhanced compatibility of hydrophilic CI with low-surface-energy hydrophobic polymer matrix in terms of their wettability and dispersibility in an elastomeric matrix. Thus, the presence of possible defects at the particle/matrix interface is reduced, and higher performance of modified MREs is expected.

Keywords: atom transfer radical polymerization, core-shell, particle modification, wettability

Procedia PDF Downloads 200

Authors: Tsega Asresa, Getahun Tigistu, Melaku Bayih

Abstract:

Computer vision is a subfield of artificial intelligence that allows computers and systems to extract meaning from digital images and video. It is used in a wide range of fields of study, including self-driving cars, video surveillance, medical diagnosis, manufacturing, law, agriculture, quality control, health care, facial recognition, and military applications. Aromatic medicinal plants are botanical raw materials used in cosmetics, medicines, health foods, essential oils, decoration, cleaning, and other natural health products for therapeutic and Aromatic culinary purposes. These plants and their products not only serve as a valuable source of income for farmers and entrepreneurs but also going to export for valuable foreign currency exchange. In Ethiopia, there is a lack of technologies for the classification and identification of Aromatic medicinal plant parts and disease type cured by aromatic medicinal plants. Farmers, industry personnel, academicians, and pharmacists find it difficult to identify plant parts and disease types cured by plants before ingredient extraction in the laboratory. Manual plant identification is a time-consuming, labor-intensive, and lengthy process. To alleviate these challenges, few studies have been conducted in the area to address these issues. One way to overcome these problems is to develop a deep learning model for efficient identification of Aromatic medicinal plant parts with their corresponding disease type. The objective of the proposed study is to identify the aromatic medicinal plant parts and their disease type classification using computer vision technology. Therefore, this research initiated a model for the classification of aromatic medicinal plant parts and their disease type by exploring computer vision technology. Morphological characteristics are still the most important tools for the identification of plants. Leaves are the most widely used parts of plants besides roots, flowers, fruits, and latex. For this study, the researcher used RGB leaf images with a size of 128x128 x3. In this study, the researchers trained five cutting-edge models: convolutional neural network, Inception V3, Residual Neural Network, Mobile Network, and Visual Geometry Group. Those models were chosen after a comprehensive review of the best-performing models. The 80/20 percentage split is used to evaluate the model, and classification metrics are used to compare models. The pre-trained Inception V3 model outperforms well, with training and validation accuracy of 99.8% and 98.7%, respectively.

Keywords: aromatic medicinal plant, computer vision, convolutional neural network, deep learning, plant classification, residual neural network

Procedia PDF Downloads 186

Authors: Hila Levi

Abstract:

Respect for human rights is a fundamental element of the European Union's (EU) identity and law. Surprisingly, however, the protection of human rights has been significantly restricted in the austerity programs ordered by the International Monetary Fund (IMF), the European Central Bank (ECB) and the European Commission (EC) (often labeled 'the Troika') in return for financial aid to the crisis-hit countries. This paper focuses on the role of the ECB in the crisis management. While other international financial institutions, such as the IMF or the World Bank, may opt to neglect human rights obligations, one might expect a greater respect of human rights from the ECB, which is bound by the EU Charter of Fundamental Rights. However, this paper argues that ECB officials made no significant effort to protect human rights or strike an adequate balance between competing financial and human rights needs while coping with the crisis. ECB officials were preoccupied with the need to stabilize the economy and prevent a collapse of the Eurozone, and paid only marginal attention to human rights concerns in the design and implementation of Troikas' programs. This paper explores the role of Organizational Culture (OC) in explaining this marginalization. While International Relations (IR) research on Intergovernmental Organizations (IGOs) behavior has traditionally focused on external interests of powerful member states, and on national and economic considerations, this study focuses on particular institutions' internal factors and independent processes. OC characteristics have been identified in OC literature as an important determinant of organizational behavior. This paper suggests that cultural characteristics are also vital for the examination of IGOs, and particularly for understanding the ECB's behavior during the crisis. In order to assess the OC of the ECB and the impact it had on its policies and decisions during the Eurozone crisis, the paper uses the results of numerous qualitative interviews conducted with high-ranking officials and staff members of the ECB involved in the crisis management. It further reviews primary sources of the ECB (such as ECB statutes, and the Memoranda of Understanding signed between the crisis countries and the Troika), and secondary sources (such as the report of the UN High Commissioner for Human Rights on Austerity measures and economic, social, and cultural rights). It thus analyzes the interaction between the ECBs culture and the almost complete absence of human rights considerations in the Eurozone crisis resolution scheme. This paper highlights the importance and influence of internal ideational factors on IGOs behavior. From a more practical perspective, this paper may contribute to understanding one of the obstacles in the process of human rights implementation in international organizations, and provide instruments for better protection of social and economic rights.

Keywords: European central bank, eurozone crisis, intergovernmental organizations, organizational culture

Procedia PDF Downloads 154

Authors: Mohammad H. Taheri, David J. Brown, Nasser Sherkat

Abstract:

Engagement is one of the most important factors in determining successful outcomes and deep learning in students. Existing approaches to detect student engagement involve periodic human observations that are subject to inter-rater reliability. Our solution uses real-time multimodal multisensor data labeled by objective performance outcomes to infer the engagement of students. The study involves four students with a combined diagnosis of cerebral palsy and a learning disability who took part in a 3-month trial over 59 sessions. Multimodal multisensor data were collected while they participated in a continuous performance test. Eye gaze, electroencephalogram, body pose, and interaction data were used to create a model of student engagement through objective labeling from the continuous performance test outcomes. In order to achieve this, a type of continuous performance test is introduced, the Seek-X type. Nine features were extracted including high-level handpicked compound features. Using leave-one-out cross-validation, a series of different machine learning approaches were evaluated. Overall, the random forest classification approach achieved the best classification results. Using random forest, 93.3% classification for engagement and 42.9% accuracy for disengagement were achieved. We compared these results to outcomes from different models: AdaBoost, decision tree, k-Nearest Neighbor, naïve Bayes, neural network, and support vector machine. We showed that using a multisensor approach achieved higher accuracy than using features from any reduced set of sensors. We found that using high-level handpicked features can improve the classification accuracy in every sensor mode. Our approach is robust to both sensor fallout and occlusions. The single most important sensor feature to the classification of engagement and distraction was shown to be eye gaze. It has been shown that we can accurately predict the level of engagement of students with learning disabilities in a real-time approach that is not subject to inter-rater reliability, human observation or reliant on a single mode of sensor input. This will help teachers design interventions for a heterogeneous group of students, where teachers cannot possibly attend to each of their individual needs. Our approach can be used to identify those with the greatest learning challenges so that all students are supported to reach their full potential.

Keywords: affective computing in education, affect detection, continuous performance test, engagement, flow, HCI, interaction, learning disabilities, machine learning, multimodal, multisensor, physiological sensors, student engagement

Procedia PDF Downloads 94

Authors: Ivana Loncarevic, Biljana Pajin, Jovana Petrovic, Danica Zaric, Vesna Tumbas Saponjac, Aleksandar Fistes

Abstract:

Chocolate is a food that activates pleasure centers in the human brain. In comparison to black and milk chocolate, white chocolate does not contain fat-free cocoa solids and thus lacks bioactive components. The aim of this study was to examine the sensory characteristics of enriched white chocolate with the addition of 10% of raspberry juice encapsulated in maltodextrins (denoted as encapsulate). Chocolate is primarily intended for enjoyment, and therefore, the sensory expectation is a critical factor for consumers when selecting a new type of chocolate. Consumer acceptance of chocolate depends primarily on the appearance and taste, but also very much on the mouthfeel, which mainly depends on the particle size of chocolate. Chocolate samples were evaluated by a panel of 8 trained panelists, food technologists, trained according to ISO 8586 (2012). Panelists developed the list of attributes to be used in this study: intensity of red color (light to dark); glow on the surface (mat to shiny); texture on snap (appearance of cavities or holes on the snap surface that are seen - even to gritty); hardness (hardness felt during the first bite of chocolate sample in half by incisors - soft to hard); melting (the time needed to convert solid chocolate into a liquid state – slowly to quickly); smoothness (perception of evenness of chocolate during melting - very even to very granular); fruitiness (impression of fruity taste - light fruity notes to distinct fruity notes); sweetness (organoleptic characteristic of pure substance or mixture giving sweet taste - lightly sweet to very sweet). The chocolate evaluation was carried out 24 h after sample preparation in the sensory laboratory, in partitioned booths, which were illuminated with fluorescent lights (ISO 8589, 2007). Samples were served in white plastic plates labeled with three-digit codes from a random number table. Panelist scored the perceived intensity of each attribute using a 7-point scale (1 = the least intensity and 7 = the most intensity) (ISO 4121, 2002). The addition of 10% of encapsulate had a big influence on chocolate color, where enriched chocolate got a nice reddish color. At the same time, the enriched chocolate sample had less intensity of gloss on the surface. The panelists noticed that addition of encapsulate reduced the time needed to convert solid chocolate into a liquid state, increasing its hardness. The addition of encapsulate had a significant impact on chocolate flavor. It reduced the sweetness of white chocolate and contributed to the fruity raspberry flavor.

Keywords: white chocolate, encapsulated raspberry juice, color, sensory characteristics

Procedia PDF Downloads 160

Authors: David Oluigbo, Erik Hemberg, Nathan Shwatal, Wenqi Ding, Yin Yuan, Susanna Mierau

Abstract:

Introduction: Calcium imaging is an established technique in neuroscience research for detecting activity in neural networks. Bursts of action potentials in neurons lead to transient increases in intracellular calcium visualized with fluorescent indicators. Manual identification of cell bodies and their contours by experts typically takes 10-20 minutes per calcium imaging recording. Our aim, therefore, was to design an automated pipeline to facilitate and optimize calcium imaging data analysis. Our pipeline aims to accelerate cell body and contour identification and production of graphical representations reflecting changes in neuronal calcium-based fluorescence. Methods: We created a Python-based pipeline that uses OpenCV (a computer vision Python package) to accurately (1) detect neuron contours, (2) extract the mean fluorescence within the contour, and (3) identify transient changes in the fluorescence due to neuronal activity. The pipeline consisted of 3 Python scripts that could both be easily accessed through a Python Jupyter notebook. In total, we tested this pipeline on ten separate calcium imaging datasets from murine dissociate cortical cultures. We next compared our automated pipeline outputs with the outputs of manually labeled data for neuronal cell location and corresponding fluorescent times series generated by an expert neuroscientist. Results: Our results show that our automated pipeline efficiently pinpoints neuronal cell body location and neuronal contours and provides a graphical representation of neural network metrics accurately reflecting changes in neuronal calcium-based fluorescence. The pipeline detected the shape, area, and location of most neuronal cell body contours by using binary thresholding and grayscale image conversion to allow computer vision to better distinguish between cells and non-cells. Its results were also comparable to manually analyzed results but with significantly reduced result acquisition times of 2-5 minutes per recording versus 10-20 minutes per recording. Based on these findings, our next step is to precisely measure the specificity and sensitivity of the automated pipeline’s cell body and contour detection to extract more robust neural network metrics and dynamics. Conclusion: Our Python-based pipeline performed automated computer vision-based analysis of calcium image recordings from neuronal cell bodies in neuronal cell cultures. Our new goal is to improve cell body and contour detection to produce more robust, accurate neural network metrics and dynamic graphs.

Keywords: calcium imaging, computer vision, neural activity, neural networks

Procedia PDF Downloads 82

Authors: Norhayati Ayob, Shamsu Mohamad

Abstract:

This paper aims to provide an initial background of the process of making traditional ceramic pottery, focusing on the materials and the influence of culture heritage. Ceramic pottery is one of the hallmarks of Sabah’s heirloom, not only use as cooking and storage containers but also closely linked with folk cultures and heritage. The Bajau Laut ethnic community of Semporna or better known as the Sea Gypsies, mostly are boat dwellers and work as fishermen in the coast. This ethnic community is famous for their own artistic traditional heirloom, especially the traditional hand-made clay stove called Lapohan. It is found that in the daily life of Bajau Laut community, Lapohan (clay stove) is used to prepare the meal and as a food warmer while they are at the sea. Besides, Lapohan pottery conveys symbolic meaning of natural objects, which portrays the identity, and values of Bajau Laut community. It is acknowledged that the basic process of making potterywares was much the same for people all across the world, nevertheless, it is crucial to consider that different ethnic groups may have their own styles and choices of raw materials. Furthermore, it is still unknown why and how the Bajau Laut ethnic of Semporna get started making their own pottery and to survive until today by heavily depending on the raw materials available in Semporna. In addition, the emergent problem faced by the pottery maker in Sabah is the absence of young successor to continue the heirloom legacy. Therefore, this research aims to explore the traditional pottery making in Sabah, by investigating the background history of Lapohan pottery and to propose the classification of Lapohan based on design and motifs of traditional pottery that will be recognised throughout the study. It is postulated that different techniques and forms of making traditional pottery may produce different types of pottery in terms of surface decoration, shape, and size that portrays different cultures. This study will be conducted at Selakan Island, Semporna, which is the only location that still has Lapohan making. This study is also based on the chronological process of making pottery and taboos of the process of preparing the clay, forming, decoration technique, motif application and firing techniques. The relevant information for the study will be gathered from field study, including observation, in-depth interview and video recording. In-depth interviews will be conducted with several potters and the conversation and pottery making process will be recorded in order to understand the actual process of making Lapohan. The findings hope to provide several types of Lapohan based on different designs and cultures, for example, the one with flat-shape design or has round-shape on the top of clay stove will be labeled with suitable name based on their culture. In conclusion, it is hoped that this study will contribute to conservation for traditional pottery making in Sabah as well as to preserve their culture and heirloom for future generations.

Keywords: Bajau Laut, culture, Lapohan, traditional pottery

Procedia PDF Downloads 187

Authors: Agnieszka E. Laudy, Karolina Stępien, Sergiusz Lulinski, Krzysztof Durka, Stefan Tyski

Abstract:

1,3-dihydro-1-hydroxy-2,1-benzoxaborole and its derivatives are a particularly interesting group of synthetic agents and were successfully employed in supramolecular chemistry medicine. The first important compounds, 5-fluoro-1,3-dihydro-1-hydroxy-2,1-benzoxaborole and 5-chloro-1,3-dihydro-1-hydroxy-2,1-benzoxaborole were identified as potent antifungal agents. In contrast, (S)-3-(aminomethyl)-7-(3-hydroxypropoxy)-1-hydroxy-1,3-dihydro-2,1-benzoxaborole hydrochloride is in the second phase of clinical trials as a drug for the treatment of Gram-negative bacterial infections of the Enterobacteriaceae family and Pseudomonas aeruginosa. Equally important and difficult task is to search for compounds active against Gram-negative bacilli, which have multi-drug-resistance efflux pumps actively removing many of the antibiotics from bacterial cells. We have examined whether halogen-substituted benzoxaborole-based derivatives and their analogues possess antibacterial activity and are substrates for multi-drug-resistance efflux pumps. The antibacterial activity of 1,3-dihydro-3-hydroxy-1,1-dimethyl-1,2,3-benzosiloxaborole and 10 halogen-substituted its derivatives, as well as 1,2-phenylenediboronic acid and 3 synthesised fluoro-substituted its analogs, were evaluated. The activity against the reference strains of Gram-positive (n=5) and Gram-negative bacteria (n=10) was screened by the disc-diffusion test (0.4 mg of tested compounds was applied onto paper disc). The minimal inhibitory concentration values and the minimal bactericidal concentration values were estimated according to The Clinical and Laboratory Standards Institute and The European Committee on Antimicrobial Susceptibility Testing recommendations. During the minimal inhibitory concentration values determination with or without phenylalanine-arginine beta-naphthylamide (50 mg/L) efflux pump inhibitor, the concentrations of tested compounds ranged 0.39-400 mg/L in the broth medium supplemented with 1 mM magnesium sulfate. Generally, the studied benzosiloxaboroles and benzoxadiboroles showed a higher activity against Gram-positive cocci than against Gram-negative rods. Moreover, benzosiloxaboroles have the higher activity than benzoxadiboroles compounds. In this study, we demonstrated that substitution (mono-, di- or tetra-) of 1,3-dihydro-3-hydroxy-1,1-dimethyl-1,2,3-benzosiloxaborole with halogen groups resulted in an increase in antimicrobial activity as compared to the parent substance. Interestingly, the 6,7-dichloro-substituted parent substance was found to be the most potent against Gram-positive cocci: Staphylococcus sp. (minimal inhibitory concentration 6.25 mg/L) and Enterococcus sp. (minimal inhibitory concentration 25 mg/L). On the other hand, mono- and dichloro-substituted compounds were the most actively removed by efflux pumps present in Gram-negative bacteria mainly from Enterobacteriaceae family. In the presence of efflux pump inhibitor the minimal inhibitory concentration values of chloro-substituted benzosiloxaboroles decreased from 400 mg/L to 3.12 mg/L. Of note, the highest increase in bacterial susceptibility to tested compounds in the presence of phenylalanine-arginine beta-naphthylamide was observed for 6-chloro-, 6,7-dichloro- and 6,7-difluoro-substituted benzosiloxaboroles. In the case of Escherichia coli, Enterobacter cloacae and P. aeruginosa strains at least a 32-fold decrease in the minimal inhibitory concentration values of these agents were observed. These data demonstrate structure-activity relationships of the tested derivatives and highlight the need for further search for benzoxaboroles and related compounds with significant antimicrobial properties. Moreover, the influence of phenylalanine-arginine beta-naphthylamide on the susceptibility of Gram-negative rods to studied benzosiloxaboroles indicate that some tested agents are substrates for efflux pumps in Gram-negative rods.

Keywords: antibacterial activity, benzosiloxaboroles, efflux pumps, phenylalanine-arginine beta-naphthylamide

Procedia PDF Downloads 271

Authors: Soowon Park, Min-Ji Kim, Jun-Young Lee

Abstract:

Motivation: The stigma of mental illness has been studied in many disciplines, including social psychology, counseling psychology, sociology, psychiatry, public health care, and related areas, because individuals labeled as ‘mentally ill’ are often deprived of their rights and their life opportunities. To understand the factors that deepen the stigma of mental illness, it is important to understand the influencing factors of the stigma. Problem statement: Depression is a common disorder in adults, but the incidence of help-seeking is low. Researchers have believed that this poor help-seeking behavior is related to the stigma of mental illness, which results from low mental health literacy. However, it is uncertain that increasing mental health literacy decreases mental health stigmatization. Furthermore, even though decreasing stigmatization is important, the stigma of mental illness is still a stable and long-lasting phenomenon. Thus, factors other than knowledge about mental disorders have the power to maintain the stigma. Investigating the influencing factors that facilitate the stigma of psychiatric disease could help lower the social stigmatization. Approach: Face-to-face interviews were conducted with a multi-clustering sample. A total of 700 Korean participants (38% male), ranging in age from 18 to 78 (M(SD)age= 48.5(15.7)) answered demographical questions, Korean version of Link’s Perceived Devaluation and Discrimination (PDD) scale for the assessment of social stigmatization against depression, and the Korean version of the WHO-Composite International Diagnostic Interview for the assessment of mental disorders. Multiple-regression was conducted to find the predicting factors of social stigmatization against depression. Ages, sex, years of education, income, living location, and experience of mental illness were used as the predictors. Results: Predictors accounted for 14% of the variance in the stigma of depressive disorders (F(6, 693) = 20.27, p < .001). Among those, only age, years of education, and experience of mental illness significantly predicted social stigmatization against depression. The standardized regression coefficient of age had a negative association with stigmatization (β = -.20, p < .001), but years of education (β = .20, p < .001) and experience of mental illness (β = .08, p < .05) positively predicted depression stigmatization. Conclusions: The present study clearly demonstrates the association between personal factors and depressive disorder stigmatization. Younger age, more education, and self-stigma appeared to increase the stigmatization. Young, highly educated, and mentally ill people tend to reject patients with depressive disorder as friends, teachers, or babysitters; they also tend to think that those patients have lower intelligence and abilities. These results suggest the possibility that people from a high social class, or highly educated people, who have the power to make decisions, help maintain the social stigma against mental illness patients. To increase the awareness that people from high social classes have more stigmatization against depressive disorders will help decrease the biased attitudes against mentally ill patients.

Keywords: depressive disorder stigmatization, age, education, self-stigma

Procedia PDF Downloads 405

Authors: Marcos Silva Tavares, Jamile Raquel Regazzo, Edson José de Souza Sardinha, Murilo Mesquita Baesso

Abstract:

Common beans are a widely cultivated and consumed legume globally, serving as a staple food for humans, especially in developing countries, due to their nutritional characteristics. Nitrogen (N) is the most limiting nutrient for productivity, and foliar analysis is crucial to ensure balanced nitrogen fertilization. Excessive N applications can cause, either isolated or cumulatively, soil and water contamination, plant toxicity, and increase their susceptibility to diseases and pests. However, the quantification of N using conventional methods is time-consuming and costly, demanding new technologies to optimize the adequate supply of N to plants. Thus, it becomes necessary to establish constant monitoring of the foliar content of this macronutrient in plants, mainly at the V4 stage, aiming at precision management of nitrogen fertilization. In this work, the objective was to evaluate the performance of a deep learning model, Resnet-50, in the classification of foliar nitrogen in common beans using RGB images. The BRS Estilo cultivar was sown in a greenhouse in a completely randomized design with four nitrogen doses (T1 = 0 kg N ha-1, T2 = 25 kg N ha-1, T3 = 75 kg N ha-1, and T4 = 100 kg N ha-1) and 12 replications. Pots with 5L capacity were used with a substrate composed of 43% soil (Neossolo Quartzarênico), 28.5% crushed sugarcane bagasse, and 28.5% cured bovine manure. The water supply of the plants was done with 5mm of water per day. The application of urea (45% N) and the acquisition of images occurred 14 and 32 days after sowing, respectively. A code developed in Matlab© R2022b was used to cut the original images into smaller blocks, originating an image bank composed of 4 folders representing the four classes and labeled as T1, T2, T3, and T4, each containing 500 images of 224x224 pixels obtained from plants cultivated under different N doses. The Matlab© R2022b software was used for the implementation and performance analysis of the model. The evaluation of the efficiency was done by a set of metrics, including accuracy (AC), F1-score (F1), specificity (SP), area under the curve (AUC), and precision (P). The ResNet-50 showed high performance in the classification of foliar N levels in common beans, with AC values of 85.6%. The F1 for classes T1, T2, T3, and T4 was 76, 72, 74, and 77%, respectively. This study revealed that the use of RGB images combined with deep learning can be a promising alternative to slow laboratory analyses, capable of optimizing the estimation of foliar N. This can allow rapid intervention by the producer to achieve higher productivity and less fertilizer waste. Future approaches are encouraged to develop mobile devices capable of handling images using deep learning for the classification of the nutritional status of plants in situ.

Keywords: convolutional neural network, residual network 50, nutritional status, artificial intelligence

Procedia PDF Downloads 19

Authors: Sherif Fakher

Abstract:

Hydrogen has been gaining a lot of global attention as an uprising contributor in the energy sector. Labeled as an energy carrier, hydrogen is used in many industries and can be used to generate electricity via fuel cells. Blue hydrogen involves the production of hydrogen from hydrocarbons using different processes that emit CO₂. However, the CO₂ is captured and stored. Hence, very little environmental damage occurs during the hydrogen production process. This research investigates the ability to use different catalysts for the production of hydrogen from different hydrocarbon sources, including coal, oil, and gas, using a two-step Aquathermolysis reaction. The research presents the results of experiments conducted to evaluate different catalysts and also highlights the main advantages of this process over other blue hydrogen production methods, including methane steam reforming, autothermal reforming, and oxidation. Two methods of hydrogen generation were investigated including partial oxidation and aquathermolysis. For those two reactions, the reaction kinetics, thermodynamics, and medium were all investigated. Following this, experiments were conducted to test the hydrogen generation potential from both methods. The porous media tested were sandstone, ash, and prozzolanic material. The spent oils used were spent motor oil and spent vegetable oil from cooking. Experiments were conducted at temperatures up to 250 C and pressures up to 3000 psi. Based on the experimental results, mathematical models were developed to predict the hydrogen generation potential at higher thermodynamic conditions. Since both partial oxidation and aquathermolysis require relatively high temperatures to undergo, it was important to devise a method by which these high temperatures can be generated at a low cost. This was done by investigating two factors, including the porous media used and the reliance on the spent oil. Of all the porous media used, the ash had the highest thermal conductivity. The second step was the partial combustion of part of the spent oil to generate the heat needed to reach the high temperatures. This reduced the cost of the heat generation significantly. For the partial oxidation reaction, the spent oil was burned in the presence of a limited oxygen concentration to generate carbon monoxide. The main drawback of this process was the need for burning. This resulted in the generation of other harmful and environmentally damaging gases. Aquathermolysis does not rely on burning, which makes it the cleaner alternative. However, it needs much higher temperatures to run the reaction. When comparing the hydrogen generation potential for both using gas chromatography, aquathermolysis generated 23% more hydrogen using the same volume of spent oil compared to partial oxidation. This research introduces the concept of using spent oil for hydrogen production. This can be a very promising method to produce a clean source of energy using a waste product. This can also help reduce the reliance on freshwater for hydrogen generation which can divert the usage of freshwater to other more important applications.

Keywords: blue hydrogen production, catalytic aquathermolysis, direct carbon dioxide capture, CCUS

Procedia PDF Downloads 31

Authors: G. C. Santini, C. Potrich, L. Lunelli, L. Vanzetti, S. Marasso, M. Cocuzza, C. Pederzolli

Abstract:

The relevance of circulating miRNAs as non-invasive biomarkers for several pathologies is nowadays undoubtedly clear, as they have been found to have both diagnostic and prognostic value able to add fundamental information to patients’ clinical picture. The availability of these data, however, relies on a time-consuming process spanning from the sample collection and processing to the data analysis. In light of this, strategies which are able to ease this procedure are in high demand and considerable effort have been made in developing Lab-on-a-chip (LOC) devices able to speed up and standardise the bench work. In this context, a very promising polydimethylsiloxane (PDMS)-based microdevice which integrates the processing of the biological sample, i.e. purification of extracellular miRNAs, and reverse transcription was previously developed in our lab. In this study, we aimed at the improvement of the miRNA extraction performances of this micro device by increasing the ability of its surface to absorb extracellular miRNAs from biological samples. For this purpose, we focused on the modulation of two properties of the material: roughness and charge. PDMS surface roughness was modulated by casting with several templates (terminated with silicon oxide coated by a thin anti-adhesion aluminum layer), followed by a panel of curing conditions. Atomic force microscopy (AFM) was employed to estimate changes at the nanometric scale. To introduce modifications in surface charge we functionalized PDMS with different mixes of positively charged 3-aminopropyltrimethoxysilanes (APTMS) and neutral poly(ethylene glycol) silane (PEG). The surface chemical composition was characterized by X-ray photoelectron spectroscopy (XPS) and the number of exposed primary amines was quantified with the reagent sulfosuccinimidyl-4-o-(4,4-dimethoxytrityl) butyrate (s-SDTB). As our final end point, the adsorption rate of all these different conditions was assessed by fluorescence microscopy by incubating a synthetic fluorescently-labeled miRNA. Our preliminary analysis identified casting on thermally grown silicon oxide, followed by a curing step at 85°C for 1 hour, as the most efficient technique to obtain a PDMS surface roughness in the nanometric scaleable to trap miRNA. In addition, functionalisation with 0.1% APTMS and 0.9% PEG was found to be a necessary step to significantly increase the amount of microRNA adsorbed on the surface, therefore, available for further steps as on-chip reverse transcription. These findings show a substantial improvement in the extraction efficiency of our PDMS microdevice, ultimately leading to an important step forward in the development of an innovative, easy-to-use and integrated system for the direct purification of less abundant circulating microRNAs.

Keywords: circulating miRNAs, diagnostics, Lab-on-a-chip, polydimethylsiloxane (PDMS)

Procedia PDF Downloads 318

Authors: Noga Bregman

Abstract:

Accurate and efficient earthquake detection and phase picking are crucial for seismic hazard assessment and emergency response. This study introduces EQMamba, a deep-learning method that combines the strengths of the Earthquake Transformer and the Mamba model for simultaneous earthquake detection and phase picking. EQMamba leverages the computational efficiency of Mamba layers to process longer seismic sequences while maintaining a manageable model size. The proposed architecture integrates convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM) networks, and Mamba blocks. The model employs an encoder composed of convolutional layers and max pooling operations, followed by residual CNN blocks for feature extraction. Mamba blocks are applied to the outputs of BiLSTM blocks, efficiently capturing long-range dependencies in seismic data. Separate decoders are used for earthquake detection, P-wave picking, and S-wave picking. We trained and evaluated EQMamba using a subset of the STEAD dataset, a comprehensive collection of labeled seismic waveforms. The model was trained using a weighted combination of binary cross-entropy loss functions for each task, with the Adam optimizer and a scheduled learning rate. Data augmentation techniques were employed to enhance the model's robustness. Performance comparisons were conducted between EQMamba and the EQTransformer over 20 epochs on this modest-sized STEAD subset. Results demonstrate that EQMamba achieves superior performance, with higher F1 scores and faster convergence compared to EQTransformer. EQMamba reached F1 scores of 0.8 by epoch 5 and maintained higher scores throughout training. The model also exhibited more stable validation performance, indicating good generalization capabilities. While both models showed lower accuracy in phase-picking tasks compared to detection, EQMamba's overall performance suggests significant potential for improving seismic data analysis. The rapid convergence and superior F1 scores of EQMamba, even on a modest-sized dataset, indicate promising scalability for larger datasets. This study contributes to the field of earthquake engineering by presenting a computationally efficient and accurate method for simultaneous earthquake detection and phase picking. Future work will focus on incorporating Mamba layers into the P and S pickers and further optimizing the architecture for seismic data specifics. The EQMamba method holds the potential for enhancing real-time earthquake monitoring systems and improving our understanding of seismic events.

Keywords: earthquake, detection, phase picking, s waves, p waves, transformer, deep learning, seismic waves

Procedia PDF Downloads 52

Authors: Yumei Han, Naiqing Song, Xiaoping Yang, Yuping Han

Abstract:

This study explores the conceptualization of key competencies that children are expected to develop in three year preschools (age 3-6) and the assessment practices of such key competencies in China. Assessment of children development has been put into the central place of early childhood education quality evaluation system in China. In the context of students key competencies development centered education reform in China, defining and selecting key competencies of children in preschools are of great significance in that they would lay a solid foundation for children’s lifelong learning path, and they would lead to curriculum and instruction reform, teacher development reform as well as quality evaluation reform in the early childhood education area. Based on sense making theory and framework, this study adopted multiple stakeholders’ (early childhood educators, parents, evaluation administrators, scholars in the early childhood education field) perspectives and grass root voices to conceptualize and operationalize key competencies for children in preschools in Southwest China. On the ground of children development theories, Chinese and international literature related to children development and key competencies, and key competencies frameworks by UNESCO, OECD and other nations, the authors designed a two-phase sequential mixed method study to address three main questions: (a) How is early childhood key competency defined or labeled from literature and from different stakeholders’ views? (b) Based on the definitions explicated in the literature and the surveys on different stakeholders, what domains and components are regarded to constitute the key competency framework of children in three-year preschools in China? (c) How have early childhood key competencies been assessed and measured, and how such assessment and measurement contribute to enhancing early childhood development quality? On the first phase, a series of focus group surveys were conducted among different types of stakeholders around the research questions. Moreover, on the second phase, based on the coding of the participants’ answers, together with literature synthesis findings, a questionnaire survey was designed and conducted to select most commonly expected components of preschool children’s key competencies. Semi-structured open questions were also included in the questionnaire for the participants to add on competencies beyond the checklist. Rudimentary findings show agreeable concerns on the significance and necessity of conceptualization and assessment of key competencies for children in preschools, and a key competencies framework composed of 7 domains and 25 indicators was constructed. Meanwhile, the findings also show issues in the current assessment practices of children’s competencies, such as lack of effective assessment tools, lack of teacher capacity in applying the tools to evaluating children and advancing children development accordingly. Finally, the authors put forth suggestions and implications for China and international communities in terms of restructuring early childhood key competencies framework, and promoting child development centered reform in early childhood education quality evaluation and development.

Keywords: assessment, conceptualization, early childhood education quality in China, key competencies

Procedia PDF Downloads 249

Authors: Yana Posmitnaya, Galina Rudnitskaya, Tatyana Lukashenko, Anton Bukatin, Anatoly Evstrapov

Abstract:

An important area of biological and medical research is the study of genetic mutations and polymorphisms that can alter gene function and cause inherited diseases and other diseases. The following methods to analyse DNA fragments are used: capillary electrophoresis and electrophoresis on microfluidic chip (MFC), mass spectrometry with electrophoresis on MFC, hybridization assay on microarray. Electrophoresis on MFC allows to analyse small volumes of samples with high speed and throughput. A soft lithography in polydimethylsiloxane (PDMS) was chosen for operative fabrication of MFCs. A master-form from silicon and photoresist SU-8 2025 (MicroChem Corp.) was created for the formation of micro-sized structures in PDMS. A universal topology which combines T-injector and simple cross was selected for the electrophoretic separation of the sample. Glass K8 and PDMS Sylgard® 184 (Dow Corning Corp.) were used for fabrication of MFCs. Electroosmotic flow (EOF) plays an important role in the electrophoretic separation of the sample. Therefore, the estimate of the quantity of EOF and the ways of its regulation are of interest for the development of the new methods of the electrophoretic separation of biomolecules. The following methods of surface modification were chosen to change EOF: high-frequency (13.56 MHz) plasma treatment in oxygen and argon at low pressure (1 mbar); 1% aqueous solution of polyvinyl alcohol; 3% aqueous solution of Kolliphor® P 188 (Sigma-Aldrich Corp.). The electroosmotic mobility was evaluated by the method of Huang X. et al., wherein the borate buffer was used. The influence of physical and chemical methods of treatment on the wetting properties of the PDMS surface was controlled by the sessile drop method. The most effective way of surface modification of MFCs, from the standpoint of obtaining the smallest value of the contact angle and the smallest value of the EOF, was the processing with aqueous solution of Kolliphor® P 188. This method of modification has been selected for the treatment of channels of MFCs, which are used for the separation of mixture of oligonucleotides fluorescently labeled with the length of chain with 10, 20, 30, 40 and 50 nucleotides. Electrophoresis was performed on the device MFAS-01 (IAI RAS, Russia) at the separation voltage of 1500 V. 6% solution of polydimethylacrylamide with the addition of 7M carbamide was used as the separation medium. The separation time of components of the mixture was determined from electropherograms. The time for untreated MFC was ~275 s, and for the ones treated with solution of Kolliphor® P 188 – ~ 220 s. Research of physical-chemical methods of surface modification of MFCs allowed to choose the most effective way for reducing EOF – the modification with aqueous solution of Kolliphor® P 188. In this case, the separation time of the mixture of oligonucleotides decreased about 20%. The further optimization of method of modification of channels of MFCs will allow decreasing the separation time of sample and increasing the throughput of analysis.

Keywords: electrophoresis, microfluidic chip, modification, nucleic acid, polydimethylsiloxane, soft lithography

Procedia PDF Downloads 413

Authors: Yiea Funk Te, Daniel Mueller, Irena Pletikosa Cvijikj

Abstract:

Small and medium enterprises (SMEs) play an important role in the economy of many countries. When the overall world economy is considered, SMEs represent 95% of all businesses in the world, accounting for 66% of the total employment. Existing studies show that the current business environment is characterized as highly turbulent and strongly influenced by modern information and communication technologies, thus forcing SMEs to experience more severe challenges in maintaining their existence and expanding their business. To support SMEs at improving their competitiveness, researchers recently turned their focus on applying data mining techniques to build risk and growth prediction models. However, data used to assess risk and growth indicators is primarily obtained via questionnaires, which is very laborious and time-consuming, or is provided by financial institutes, thus highly sensitive to privacy issues. Recently, web mining (WM) has emerged as a new approach towards obtaining valuable insights in the business world. WM enables automatic and large scale collection and analysis of potentially valuable data from various online platforms, including companies’ websites. While WM methods have been frequently studied to anticipate growth of sales volume for e-commerce platforms, their application for assessment of SME risk and growth indicators is still scarce. Considering that a vast proportion of SMEs own a website, WM bears a great potential in revealing valuable information hidden in SME websites, which can further be used to understand SME risk and growth indicators, as well as to enhance current SME risk and growth prediction models. This study aims at developing an automated system to collect business-relevant data from the Web and predict future growth trends of SMEs by means of WM and data mining techniques. The envisioned system should serve as an 'early recognition system' for future growth opportunities. In an initial step, we examine how structured and semi-structured Web data in governmental or SME websites can be used to explain the success of SMEs. WM methods are applied to extract Web data in a form of additional input features for the growth prediction model. The data on SMEs provided by a large Swiss insurance company is used as ground truth data (i.e. growth-labeled data) to train the growth prediction model. Different machine learning classification algorithms such as the Support Vector Machine, Random Forest and Artificial Neural Network are applied and compared, with the goal to optimize the prediction performance. The results are compared to those from previous studies, in order to assess the contribution of growth indicators retrieved from the Web for increasing the predictive power of the model.

Keywords: data mining, SME growth, success factors, web mining

Procedia PDF Downloads 267

Authors: Jazim Sohail, Filipe Teixeira-Dias

Abstract:

Mild traumatic brain injury (mTBI) within non-helmeted contact sports is a growing concern due to the serious risk of potential injury. Extensive research is being conducted looking into head kinematics in non-helmeted contact sports utilizing instrumented mouthguards that allow researchers to record accelerations and velocities of the head during and after an impact. This does not, however, allow the location of the impact on the head, and its magnitude and orientation, to be determined. This research proposes and validates two methods to quantify impact locations from instrumented mouthguard kinematic data, one using rigid body dynamics, the other utilizing machine learning. The rigid body dynamics technique focuses on establishing and matching moments from Euler’s and torque equations in order to find the impact location on the head. The methodology is validated with impact data collected from a lab test with the dummy head fitted with an instrumented mouthguard. Additionally, a Hybrid III Dummy head finite element model was utilized to create synthetic kinematic data sets for impacts from varying locations to validate the impact location algorithm. The algorithm calculates accurate impact locations; however, it will require preprocessing of live data, which is currently being done by cross-referencing data timestamps to video footage. The machine learning technique focuses on eliminating the preprocessing aspect by establishing trends within time-series signals from instrumented mouthguards to determine the impact location on the head. An unsupervised learning technique is used to cluster together impacts within similar regions from an entire time-series signal. The kinematic signals established from mouthguards are converted to the frequency domain before using a clustering algorithm to cluster together similar signals within a time series that may span the length of a game. Impacts are clustered within predetermined location bins. The same Hybrid III Dummy finite element model is used to create impacts that closely replicate on-field impacts in order to create synthetic time-series datasets consisting of impacts in varying locations. These time-series data sets are used to validate the machine learning technique. The rigid body dynamics technique provides a good method to establish accurate impact location of impact signals that have already been labeled as true impacts and filtered out of the entire time series. However, the machine learning technique provides a method that can be implemented with long time series signal data but will provide impact location within predetermined regions on the head. Additionally, the machine learning technique can be used to eliminate false impacts captured by sensors saving additional time for data scientists using instrumented mouthguard kinematic data as validating true impacts with video footage would not be required.

Keywords: head impacts, impact location, instrumented mouthguard, machine learning, mTBI

Procedia PDF Downloads 217

Authors: Yakup Ulusu, Faruk Ozel, Numan Eczacioglu, Abdurrahman Ozen, Sabriye Acikgoz

Abstract:

GFP discovered in the mid-1970s, has been used as a marker after replicated genetic study by scientists. In biotechnology, cell, molecular biology, the GFP gene is frequently used as a reporter of expression. In modified forms, it has been used to make biosensors. Many animals have been created that express GFP as an evidence that a gene can be expressed throughout a given organism. Proteins labeled with GFP identified locations are determined. And so, cell connections can be monitored, gene expression can be reported, protein-protein interactions can be observed and signals that create events can be detected. Additionally, monitoring GFP is noninvasive; it can be detected by under UV-light because of simply generating fluorescence. Moreover, GFP is a relatively small and inert molecule, that does not seem to treat any biological processes of interest. The synthesis of GFP has some steps like, to construct the plasmid system, transformation in E. coli, production and purification of protein. GFP carrying plasmid vector pBAD–GFPuv was digested using two different restriction endonuclease enzymes (NheI and Eco RI) and DNA fragment of GFP was gel purified before cloning. The GFP-encoding DNA fragment was ligated into pET28a plasmid using NheI and Eco RI restriction sites. The final plasmid was named pETGFP and DNA sequencing of this plasmid indicated that the hexa histidine-tagged GFP was correctly inserted. Histidine-tagged GFP was expressed in an Escherichia coli BL21 DE3 (pLysE) strain. The strain was transformed with pETGFP plasmid and grown on LuiraBertoni (LB) plates with kanamycin and chloramphenicol selection. E. coli cells were grown up to an optical density (OD 600) of 0.8 and induced by the addition of a final concentration of 1mM isopropyl-thiogalactopyranoside (IPTG) and then grown for additional 4 h. The amino-terminal hexa-histidine-tag facilitated purification of the GFP by using a His Bind affinity chromatography resin (Novagen). Purity of GFP protein was analyzed by a 12 % sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). The concentration of protein was determined by UV absorption at 280 nm (Varian Cary 50 Scan UV/VIS spectrophotometer). Synthesis of GFP-Polymer composite nanofibers was produced by using GFP solution (10mg/mL) and polymer precursor Polyvinylpyrrolidone, (PVP, Mw=1300000) as starting materials and template, respectively. For the fabrication of nanofibers with the different fiber diameter; a sol–gel solution comprising of 0.40, 0.60 and 0.80 g PVP (depending upon the desired fiber diameter) and 100 mg GFP in 10 mL water: ethanol (3:2) mixtures were prepared and then the solution was covered on collecting plate via electro spinning at 10 kV with a feed-rate of 0.25 mL h-1 using Spellman electro spinning system. Results show that GFP-based nano-fiber can be used plenty of biomedical applications such as bio-imaging, bio-mechanic, bio-material and tissue engineering.

Keywords: biomaterial, GFP, nano-fibers, protein expression

Procedia PDF Downloads 320

Authors: Mohammad Karimi Moridani, Mohammad Abdi Zadeh, Zahra Shahiazar Mazraeh

Abstract:

In recent years, the use of intelligent systems in biomedical engineering has increased dramatically, especially in the diagnosis of various diseases. Also, due to the relatively simple recording of the electrocardiogram signal (ECG), this signal is a good tool to show the function of the heart and diseases associated with it. The aim of this paper is to design an intelligent system for automatically detecting a normal electrocardiogram signal from abnormal one. Using this diagnostic system, it is possible to identify a person's heart condition in a very short time and with high accuracy. The data used in this article are from the Physionet database, available in 2016 for use by researchers to provide the best method for detecting normal signals from abnormalities. Data is of both genders and the data recording time varies between several seconds to several minutes. All data is also labeled normal or abnormal. Due to the low positional accuracy and ECG signal time limit and the similarity of the signal in some diseases with the normal signal, the heart rate variability (HRV) signal was used. Measuring and analyzing the heart rate variability with time to evaluate the activity of the heart and differentiating different types of heart failure from one another is of interest to the experts. In the preprocessing stage, after noise cancelation by the adaptive Kalman filter and extracting the R wave by the Pan and Tampkinz algorithm, R-R intervals were extracted and the HRV signal was generated. In the process of processing this paper, a new idea was presented that, in addition to using the statistical characteristics of the signal to create a return map and extraction of nonlinear characteristics of the HRV signal due to the nonlinear nature of the signal. Finally, the artificial neural networks widely used in the field of ECG signal processing as well as distinctive features were used to classify the normal signals from abnormal ones. To evaluate the efficiency of proposed classifiers in this paper, the area under curve ROC was used. The results of the simulation in the MATLAB environment showed that the AUC of the MLP and SVM neural network was 0.893 and 0.947, respectively. As well as, the results of the proposed algorithm in this paper indicated that the more use of nonlinear characteristics in normal signal classification of the patient showed better performance. Today, research is aimed at quantitatively analyzing the linear and non-linear or descriptive and random nature of the heart rate variability signal, because it has been shown that the amount of these properties can be used to indicate the health status of the individual's heart. The study of nonlinear behavior and dynamics of the heart's neural control system in the short and long-term provides new information on how the cardiovascular system functions, and has led to the development of research in this field. Given that the ECG signal contains important information and is one of the common tools used by physicians to diagnose heart disease, but due to the limited accuracy of time and the fact that some information about this signal is hidden from the viewpoint of physicians, the design of the intelligent system proposed in this paper can help physicians with greater speed and accuracy in the diagnosis of normal and patient individuals and can be used as a complementary system in the treatment centers.

Keywords: neart rate variability, signal processing, linear and non-linear features, classification methods, ROC Curve

Procedia PDF Downloads 262

Authors: Andrew N. Saylor, James R. Peters

Abstract:

Scoliosis is a complex 3D deformity of the thoracic and lumbar spines, clinically diagnosed by measurement of a Cobb angle of 10 degrees or more on a coronal X-ray. The Cobb angle is the angle made by the lines drawn along the proximal and distal endplates of the respective proximal and distal vertebrae comprising the curve. Traditionally, Cobb angles are measured manually using either a marker, straight edge, and protractor or image measurement software. The task of measuring the Cobb angle can also be represented by a function taking the spine geometry rendered using X-ray imaging as input and returning the approximate angle. Although the form of such a function may be unknown, it can be approximated using artificial neural networks (ANNs). The performance of ANNs is affected by many factors, including the choice of activation function and network architecture; however, the effects of these parameters on the accuracy of scoliotic deformity measurements are poorly understood. Therefore, the objective of this study was to systematically investigate the effect of ANN architecture and activation function on Cobb angle measurement from the coronal X-rays of scoliotic subjects. The data set for this study consisted of 609 coronal chest X-rays of scoliotic subjects divided into 481 training images and 128 test images. These data, which included labeled Cobb angle measurements, were obtained from the SpineWeb online database. In order to normalize the input data, each image was resized using bi-linear interpolation to a size of 500 × 187 pixels, and the pixel intensities were scaled to be between 0 and 1. A fully connected (dense) ANN with a fixed cost function (mean squared error), batch size (10), and learning rate (0.01) was developed using Python Version 3.7.3 and TensorFlow 1.13.1. The activation functions (sigmoid, hyperbolic tangent [tanh], or rectified linear units [ReLU]), number of hidden layers (1, 3, 5, or 10), and number of neurons per layer (10, 100, or 1000) were varied systematically to generate a total of 36 network conditions. Stochastic gradient descent with early stopping was used to train each network. Three trials were run per condition, and the final mean squared errors and mean absolute errors were averaged to quantify the network response for each condition. The network that performed the best used ReLU neurons had three hidden layers, and 100 neurons per layer. The average mean squared error of this network was 222.28 ± 30 degrees2, and the average mean absolute error was 11.96 ± 0.64 degrees. It is also notable that while most of the networks performed similarly, the networks using ReLU neurons, 10 hidden layers, and 1000 neurons per layer, and those using Tanh neurons, one hidden layer, and 10 neurons per layer performed markedly worse with average mean squared errors greater than 400 degrees2 and average mean absolute errors greater than 16 degrees. From the results of this study, it can be seen that the choice of ANN architecture and activation function has a clear impact on Cobb angle inference from coronal X-rays of scoliotic subjects.

Keywords: scoliosis, artificial neural networks, cobb angle, medical imaging

Procedia PDF Downloads 129

Authors: Fernanda Bravo Cornejo, Camilo Cerda Sarabia, Belén Díaz Díaz, Diego Santibañez Oyarce, Esteban Gómez Terán, Hugo Osses Prado, Raúl Caulier-Cisterna, Jorge Vergara-Quezada, Ana Moya-Beltrán

Abstract:

Streptococcus pyogenes is a gram-positive bacteria involved in a wide range of diseases and is a major-human-specific bacterial pathogen. In Chile, this year the 'Ministerio de Salud' declared an alert due to the increase in strains throughout the year. This increase can be attributed to the multitude of factors including antimicrobial resistance (AMR) and Virulence Factors (VF). Understanding these VF and AMR is crucial for developing effective strategies and improving public health responses. Moreover, experimental identification and characterization of these pathogenic mechanisms are labor-intensive and time-consuming. Therefore, new computational methods are required to provide robust techniques for accelerating this identification. Advances in Machine Learning (ML) algorithms represent the opportunity to refine and accelerate the discovery of VF associated with Streptococcus pyogenes. In this work, we evaluate the accuracy of various machine learning models in predicting the virulence factors and antimicrobial resistance of Streptococcus pyogenes, with the objective of providing new methods for identifying the pathogenic mechanisms of this organism.Our comprehensive approach involved the download of 32,798 genbank files of S. pyogenes from NCBI dataset, coupled with the incorporation of data from Virulence Factor Database (VFDB) and Antibiotic Resistance Database (CARD) which contains sequences of AMR gene sequence and resistance profiles. These datasets provided labeled examples of both virulent and non-virulent genes, enabling a robust foundation for feature extraction and model training. We employed preprocessing, characterization and feature extraction techniques on primary nucleotide/amino acid sequences and selected the optimal more for model training. The feature set was constructed using sequence-based descriptors (e.g., k-mers and One-hot encoding), and functional annotations based on database prediction. The ML models compared are logistic regression, decision trees, support vector machines, neural networks among others. The results of this work show some differences in accuracy between the algorithms, these differences allow us to identify different aspects that represent unique opportunities for a more precise and efficient characterization and identification of VF and AMR. This comparative analysis underscores the value of integrating machine learning techniques in predicting S. pyogenes virulence and AMR, offering potential pathways for more effective diagnostic and therapeutic strategies. Future work will focus on incorporating additional omics data, such as transcriptomics, and exploring advanced deep learning models to further enhance predictive capabilities.

Keywords: antibiotic resistance, streptococcus pyogenes, virulence factors., machine learning

Procedia PDF Downloads 30

Authors: George Zhou, Yunchan Chen, Candace Chien

Abstract:

Kidney replacement therapy is the current standard of care for end-stage renal diseases. In-center or home hemodialysis remains an integral component of the therapeutic regimen. Arteriovenous fistulas (AVF) make up the vascular circuit through which blood is filtered and returned. Naturally, AVF patency determines whether adequate clearance and filtration can be achieved and directly influences clinical outcomes. Our aim was to build a deep learning model for automated AVF stenosis screening based on the sound of blood flow through the AVF. A total of 311 patients with AVF were enrolled in this study. Blood flow sounds were collected using a digital stethoscope. For each patient, blood flow sounds were collected at 6 different locations along the patient’s AVF. The 6 locations are artery, anastomosis, distal vein, middle vein, proximal vein, and venous arch. A total of 1866 sounds were collected. The blood flow sounds are labeled as “patent” (normal) or “stenotic” (abnormal). The labels are validated from concurrent ultrasound. Our dataset included 1527 “patent” and 339 “stenotic” sounds. We show that blood flow sounds vary significantly along the AVF. For example, the blood flow sound is loudest at the anastomosis site and softest at the cephalic arch. Contextualizing the sound with location metadata significantly improves classification performance. How to encode and incorporate categorical metadata is an active area of research1. Herein, we study ordinal (i.e., integer) encoding schemes. The numerical representation is concatenated to the flattened feature vector. We train a vision transformer (ViT) on spectrogram image representations of the sound and demonstrate that using scalar multiples of our integer encodings improves classification performance. Models are evaluated using a 10-fold cross-validation procedure. The baseline performance of our ViT without any location metadata achieves an AuROC and AuPRC of 0.68 ± 0.05 and 0.28 ± 0.09, respectively. Using the following encodings of Artery:0; Arch: 1; Proximal: 2; Middle: 3; Distal 4: Anastomosis: 5, the ViT achieves an AuROC and AuPRC of 0.69 ± 0.06 and 0.30 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 10; Proximal: 20; Middle: 30; Distal 40: Anastomosis: 50, the ViT achieves an AuROC and AuPRC of 0.74 ± 0.06 and 0.38 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 100; Proximal: 200; Middle: 300; Distal 400: Anastomosis: 500, the ViT achieves an AuROC and AuPRC of 0.78 ± 0.06 and 0.43 ± 0.11. respectively. Interestingly, we see that using increasing scalar multiples of our integer encoding scheme (i.e., encoding “venous arch” as 1,10,100) results in progressively improved performance. In theory, the integer values do not matter since we are optimizing the same loss function; the model can learn to increase or decrease the weights associated with location encodings and converge on the same solution. However, in the setting of limited data and computation resources, increasing the importance at initialization either leads to faster convergence or helps the model escape a local minimum.

Keywords: arteriovenous fistula, blood flow sounds, metadata encoding, deep learning

Procedia PDF Downloads 87

Authors: Roxana D. Maiorescu-Murphy

Abstract:

In the US, corporations have recently embraced political stances in an attempt to respond to the external pressure exerted by activist groups. To date, research in this area remains in its infancy, and few studies have been conducted on the way stakeholder groups respond to corporate political advocacy in general and in the immediacy of such a corporate announcement in particular. The current study aims to fill in this research void. In addition, the study contributes to an emerging trajectory in the field of crisis management by focusing on the delineation between crises (unexpected events related to products and services) and scandals (crises that spur moral outrage). The present study looked at online reactions in the aftermath of Elon Musk’s endorsement of the Republican party on Twitter. Two data sets were collected from Twitter following two political endorsements made by Elon Musk on May 18, 2022, and June 15, 2022, respectively. The total sample of analysis stemming from the data two sets consisted of N=1,374 user comments written as a response to Musk’s initial tweets. Given the paucity of studies in the preceding research areas, the analysis employed a case study methodology, used in circumstances in which the phenomena to be studied had not been researched before. According to the case study methodology, which answers the questions of how and why a phenomenon occurs, this study responded to the research questions of how online users perceived Tesla and why they did so. The data were analyzed in NVivo by the use of the grounded theory methodology, which implied multiple exposures to the text and the undertaking of an inductive-deductive approach. Through multiple exposures to the data, the researcher ascertained the common themes and subthemes in the online discussion. Each theme and subtheme were later defined and labeled. Additional exposures to the text ensured that these were exhaustive. The results revealed that the CEO’s political endorsements triggered moral outrage, leading to Tesla’s facing a scandal as opposed to a crisis. The moral outrage revolved around the stakeholders’ predominant rejection of a perceived intrusion of an influential figure on a domain reserved for voters. As expected, Musk’s political endorsements led to polarizing opinions, and those who opposed his views engaged in online activism aimed to boycott the Tesla brand. These findings reveal that the moral outrage that characterizes a scandal requires communication practices that differ from those that practitioners currently borrow from the field of crisis management. Specifically, because scandals flourish in online settings, practitioners should regularly monitor stakeholder perceptions and address them in real-time. While promptness is essential when managing crises, it becomes crucial to respond immediately as a scandal is flourishing online. Finally, attempts should be made to distance a brand, its products, and its CEO from the latter’s political views.

Keywords: crisis management, communication management, Tesla, corporate political activism, Elon Musk

Procedia PDF Downloads 91

Authors: Z. Giagkazoglou, D. Loukovitis, C. Gubili, A. Imsiridou

Abstract:

Along with the increase in human population, demand for seafood has increased. Despite the strict labeling regulations that exist for most marketed species in the European Union, seafood substitution remains a persistent global issue. Food fraud occurs when food products are traded in a false or misleading way. Mislabeling occurs when one species is substituted and traded under the name of another, and it can be intentional or unintentional. Crustaceans are one of the most regularly consumed seafood in Greece. Shrimps, prawns, lobsters, crayfish, and crabs are considered a delicacy and can be encountered in a variety of market presentations (fresh, frozen, pre-cooked, peeled, etc.). With most of the external traits removed, products as such are susceptible to species substitution. DNA barcoding has proven to be the most accurate method for the detection of fraudulent seafood products. To our best knowledge, the DNA barcoding methodology is used for the first time in Greece, in order to investigate the labeling practices for crustacean products available in the market. A total of 100 tissue samples were collected from various retailers and markets across four Greek cities. In an effort to cover the highest range of products possible, different market presentations were targeted (fresh, frozen and cooked). Genomic DNA was extracted using the DNeasy Blood & Tissue Kit, according to the manufacturer's instructions. The mitochondrial gene selected as the target region of the analysis was the cytochrome c oxidase subunit I (COI). PCR products were purified and sequenced using an ABI 3500 Genetic Analyzer. Sequences were manually checked and edited using BioEdit software and compared against the ones available in GenBank and BOLD databases. Statistical analyses were conducted in R and PAST software. For most samples, COI amplification was successful, and species-level identification was possible. The preliminary results estimate moderate mislabeling rates (25%) in the identified samples. Mislabeling was most commonly detected in fresh products, with 50% of the samples in this category labeled incorrectly. Overall, the mislabeling rates detected by our study probably relate to some degree of unintentional misidentification, and lack of knowledge surrounding the legal designations by both retailers and consumers. For some species of crustaceans (i.e. Squila mantis) the mislabeling appears to be also affected by the local labeling practices. Across Greece, S. mantis is sold in the market under two common names, but only one is recognized by the country's legislation, and therefore any mislabeling is probably not profit-motivated. However, the substitution of the speckled shrimp (Metapenaus monoceros) for the distinct, giant river prawn (Macrobranchium rosenbergii), is a clear example of deliberate fraudulent substitution, aiming for profit. To our best knowledge, no scientific study investigating substitution and mislabeling rates in crustaceans has been conducted in Greece. For a better understanding of Greece's seafood market, similar DNA barcoding studies in other regions with increased touristic importance (e.g., the Greek islands) should be conducted. Regardless, the expansion of the list of species-specific designations for crustaceans in the country is advised.

Keywords: COI gene, food fraud, labelling control, molecular identification

Procedia PDF Downloads 67

Authors: Carlos E. Rodríguez-Rodríguez, Karla Ruiz-Hidalgo, Kattia Madrigal-Zúñiga, Juan Salvador Chin-Pampillo, Mario Masís-Mora, Elizabeth Carazo-Rojas

Abstract:

One of the main causes of contamination linked to agricultural activities is the spillage and disposal of pesticides, especially during the loading, mixing or cleaning of agricultural spraying equipment. One improvement in the handling of pesticides is the use of biopurification systems (BPS), simple and cheap degradation devices where the pesticides are biologically degraded at accelerated rates. The biologically active core of BPS is the biomixture, which is constituted by soil pre-exposed to the target pesticide, a lignocellulosic substrate to promote the activity of ligninolitic fungi and a humic component (peat or compost), mixed at a volumetric proportion of 50:25:25. Considering the known ability of lignocellulosic fungi to degrade a wide range of organic pollutants, and the high amount of lignocellulosic waste used in biomixture preparation, the bioaugmentation of biomixtures with these fungi represents an interesting approach for improving biomixtures. The present work aimed at evaluating the effect of the bioaugmentation of rice husk based biomixtures with the fungus Trametes versicolor in the removal of the insectice/nematicide carbofuran (CFN) and to optimize the composition of the biomixture to obtain the best performance in terms of CFN removal and mineralization, reduction in formation of transformation products and decrease in residual toxicity of the matrix. The evaluation of several lignocellulosic residues (rice husk, wood chips, coconut fiber, sugarcane bagasse or newspaper print) revealed the best colonization by T. versicolor in rice husk. Pre-colonized rice husk was then used in the bioaugmentation of biomixtures also containing soil pre-exposed to CFN and either peat (GTS biomixture) or compost (GCS biomixture). After spiking with 10 mg/kg CBF, the efficiency of the biomixture was evaluated through a multi-component approach that included: monitoring of CBF removal and production of CBF transformation products, mineralization of radioisotopically labeled carbofuran (14C-CBF) and changes in the toxicity of the matrix after the treatment (Daphnia magna acute immobilization test). Estimated half-lives of CBF in the biomixtures were 3.4 d and 8.1 d in GTS and GCS, respectively. The transformation products 3-hydroxycarbofuran and 3-ketocarbofuran were detected at the moment of CFN application, however their concentration continuously disappeared. Mineralization of 14C-CFN was also faster in GTS than GCS. The toxicological evaluation showed a complete toxicity removal in the biomixtures after 48 d of treatment. The composition of the GCS biomixture was optimized using a central composite design and response surface methodology. The design variables were the volumetric content of fungally pre-colonized rice husk and the volumetric ratio compost/soil. According to the response models, maximization of CFN removal and mineralization rate, and minimization in the accumulation of transformation products were obtained with an optimized biomixture of composition 30:43:27 (pre-colonized rice husk:compost:soil), which differs from the 50:25:25 composition commonly employed in BPS. Results suggest that fungal bioaugmentation may enhance the performance of biomixtures in CFN removal. Optimization reveals the importance of assessing new biomixture formulations in order to maximize their performance.

Keywords: bioaugmentation, biopurification systems, degradation, fungi, pesticides, toxicity

Procedia PDF Downloads 311

Authors: Clea Southall

Abstract:

Many children leave primary school having formed preconceptions about their relationship with science. Thus, primary school represents a critical window for stimulating scientific interest in younger children. Engagement relies on the provision of hands-on activities coupled with an ability to capture a child’s innate curiosity. This requires children to perceive science topics as interesting and relevant to their everyday life. Teachers and pupils alike have suggested the school curriculum be tailored to help stimulate scientific interest. Young children are naturally inquisitive about the human body; the brain is one topic which frequently engages pupils, although it is not currently included in the UK primary curriculum. Teaching children about the brain could have wider societal impacts such as increasing knowledge of neurological disorders. However, many primary school teachers do not receive formal neuroscience training and may feel apprehensive about delivering lessons on the nervous system. This is exacerbated by a lack of educational neuroscience resources. One solution is for undergraduates to form partnerships with schools - delivering engaging lessons and supplementing teacher knowledge. The aim of this project was to evaluate the success of a short lesson on the brain delivered by an undergraduate neuroscientist to primary school pupils. Prior to entering schools, semi-structured online interviews were conducted with teachers to gain pedagogical advice and relevant websites were searched for neuroscience resources. Subsequently, a single lesson plan was created comprising of four hands-on activities. The activities were devised in a top-down manner, beginning with learning about the brain as an entity, before focusing on individual neurons. Students were asked to label a ‘brain map’ to assess prior knowledge of brain structure and function. They viewed animal brains and created ‘pipe-cleaner neurons’ which were later used to depict electrical transmission. The same session was delivered by an undergraduate student to 570 key stage 2 (KS2) pupils across five schools in Leeds, UK. Post-session surveys, designed for teachers and pupils respectively, were used to evaluate the session. Children in all year groups had relatively poor knowledge of brain structure and function at the beginning of the session. When asked to label four brain regions with their respective functions, older pupils labeled a mean of 1.5 (± 1.0) brain regions compared to 0.8 (± 0.96) for younger pupils (p=0.002). However, by the end of the session, 95% of pupils felt their knowledge of the brain had increased. Hands-on activities were rated most popular by pupils and were considered the most successful aspect of the session by teachers. Although only half the teachers were aware of neuroscience educational resources, nearly all (95%) felt they would have more confidence in teaching a similar session in the future. All teachers felt the session was engaging and that the content could be linked to the current curriculum. Thus, a short fifty-minute session can successfully enhance pupils’ knowledge of a new topic: the brain. Partnerships with an undergraduate student can provide an alternative method for supplementing teacher knowledge, increasing their confidence in delivering future lessons on the nervous system.

Keywords: education, neuroscience, primary school, undergraduate

Procedia PDF Downloads 211

Authors: Neha Adsul, Rohit Shah

Abstract:

Background: The global history of nursing is exclusively a history of deep contradictions as it seeks to negotiate inclusion in an already gendered world. Although a powerful 'clinical gaze exists, nurses have toiled to re-negotiate and subvert the 'medical gaze' by practicing the 'therapeutic gaze' to tether back 'care into nursing practice.' This helps address the duality of the 'body' and 'mind' wherein the patient is not just limited to being an object of medical inquiry. Nevertheless, there has been a consistent effort to fit 'nursing' into being an art or an emerging science over the years. Especially with advances in hospital-based techno-centric medical practices, the boundaries between technology and nursing practices are becoming more blurred as the technical process becomes synonymous with nursing, eroding the essence of nursing care. Aim: This paper examines the history of nursing and offers insights into how gendered relations and the ideological belief of 'nursing as gendered work' have propagated to the subjugation of the nursing profession. It further aims to provide insights into the patriarchally imbibed techno-centrism that negates the gendered caregiving which lies at the crux of a nurse's work. Method: A literature search was carried out using Google Scholar, Web of Science and PubMed databases. Search words included: technology and nursing, medical technology and nursing, history of nursing, sociology and nursing and nursing care. The history of nursing is presented in a discussion that weaves together the historical events of the 'Birth of the Clinic' and the shift from 'bed-side medicine' to 'hospital-based medicine' that legitimizes exploitation of the bodies of patients to the 'medical gaze while the emergence of nursing as acquiescent to instrumental, technical, positivist and dominant views of medicine. The resultant power asymmetries, wherein in contemporary nursing, the constant struggle of nurses to juggle between being the physicians "operational right arm" to harboring that subjective understanding of the patients to refrain from de-humanizing nursing-care. Findings: The nursing profession suffers from being rendered invisible due to gendered relations having patrifocal societal roots. This perpetuates a notion rooted in the idea that emphasizes empiricism and has resulted in theoretical and epistemological fragmentation of the understanding of body and mind as separate entities. Nurses operate within this structure while constantly being at the brink of being pushed beyond the legitimate professional boundaries while being labeled as being 'unscientific' as the work does not always corroborate and align with the existing dominant positivist lines of inquiries. Conclusion: When understood in this broader context of how nursing as a practice has evolved over the years, it provides a particularly crucial testbed for understanding contemporary gender relations. Not because nurses like to live in a gendered work trap but because the gendered relations at work are written in a covert narcissistic patriarchal milieu that fails to recognize the value of intangible yet utmost necessary 'caring work in nursing. This research urges and calls for preserving and revering the humane aspect of nursing care alongside the emerging tech-savvy expectations from nursing work.

Keywords: nursing history, technocentric, power relations, scientific duality

Procedia PDF Downloads 145

Authors: Jenson George, Thoa Nguyen, Garth Sanewski, Craig Hardner, Heather Eunice Smyth

Abstract:

Pineapple (Ananas comosus), with its unique sweet flavour, is one of the most popular tropical, non-climacteric fruits consumed worldwide. It is also the third most important tropical fruit in world production. In Australia, 99% of the pineapple production is from the Queensland state due to the favourable subtropical climatic conditions. The flavourful fruit is known to contain around 500 volatile organic compounds (VOC) at varying concentrations and greatly contribute to the flavour quality of pineapple fruit by providing distinct aroma sensory properties that are sweet, fruity, tropical, pineapple-like, caramel-like, coconut-like, etc. The aroma of pineapple is one of the important factors attracting consumers and strengthening the marketplace. To better understand the aroma of Australian-grown pineapples, the matrix-matched Gas chromatography–mass spectrometry (GC-MS), Head Space - Solid-phase microextraction (HS-SPME), Stable-isotope dilution analysis (SIDA) method was developed and validated. The developed method represents a significant improvement over current methods with the incorporation of multiple external reference standards, multiple isotopes labeled internal standards, and a matching model system of pineapple fruit matrix. This method was employed to quantify 28 key aroma compounds in more than 200 genetically diverse pineapple varieties from a breeding program. The Australian pineapple cultivars varied in content and composition of free volatile compounds, which were predominantly comprised of esters, followed by terpenes, alcohols, aldehydes, and ketones. Using selected commercial cultivars grown in Australia, and by employing the sensorial analysis, the appearance (colour), aroma (intensity, sweet, vinegar/tang, tropical fruits, floral, coconut, green, metallic, vegetal, fresh, peppery, fermented, eggy/sulphurous) and texture (crunchiness, fibrousness, and juiciness) were obtained. Relationships between sensory descriptors and volatiles were explored by applying multivariate analysis (PCA) to the sensorial and chemical data. The key aroma compounds of pineapple exhibited a positive correlation with corresponding sensory properties. The sensory and volatile data were also used to explore genetic diversity in the breeding population. GWAS was employed to unravel the genetic control of the pineapple volatilome and its interplay with fruit sensory characteristics. This study enhances our understanding of pineapple aroma (flavour) compounds, their biosynthetic pathways and expands breeding option for pineapple cultivars. This research provides foundational knowledge to support breeding programs, post-harvest and target market studies, and efforts to optimise the flavour of commercial pineapple varieties and their parent lines to produce better tasting fruits for consumers.

Keywords: Ananas comosus, pineapple, flavour, volatile organic compounds, aroma, Gas chromatography–mass spectrometry (GC-MS), Head Space - Solid-phase microextraction (HS-SPME), Stable-isotope dilution analysis (SIDA).

Procedia PDF Downloads 57

Authors: Huan Peng, Chenye Zeng, Zhao Wang

Abstract:

Alzheimer’s disease (AD) is an age-related neurodegenerative disease that is a leading cause of dementia syndromes and has become a huge burden on society and families. The main pathological features of AD involve excessive deposition of β-amyloid (Aβ) and Tau proteins in the brain, resulting in loss of neurons, expansion of neuroinflammation, and cognitive dysfunction in patients. Researchers have found effective drugs to clear the brain of error-accumulating proteins or to slow the loss of neurons, but their direct administration has key bottlenecks such as single-drug limitation, rapid blood clearance rate, impenetrable blood-brain barrier (BBB), and poor ability to target tissues and cells. Therefore, we are committed to seeking a suitable and efficient delivery system. Inspired by the possibility that exosomes may be involved in the secretion and transport mechanism of many signaling molecules or proteins in the brain, exosomes have attracted extensive attention as natural nanoscale drug carriers. We selected exosomes derived from bone marrow mesenchymal stem cells (MSC-EXO) with low immunogenicity and exosomes derived from hippocampal neurons (HT22-EXO) that may have excellent homing ability to overcome the deficiencies of oral or injectable pathways and bypass the BBB through nasal administration and evaluated their delivery ability and effect on AD. First, MSC-EXO and HT22 cells were isolated and cultured, and MSCs were identified by microimaging and flow cytometry. Then MSC-EXO and HT22-EXO were obtained by gradient centrifugation and qEV SEC separation column, and a series of physicochemical characterization were performed by transmission electron microscope, western blot, nanoparticle tracking analysis and dynamic light scattering. Next, exosomes labeled with lipophilic fluorescent dye were administered to WT mice and APP/PS1 mice to obtain fluorescence images of various organs at different times. Finally, APP/PS1 mice were administered intranasally with two exosomes 20 times over 40 days and 20 μL each time. Behavioral analysis and pathological section analysis of the hippocampus were performed after the experiment. The results showed that MSC-EXO and HT22-EXO were successfully isolated and characterized, and they had good biocompatibility. MSC-EXO showed excellent brain enrichment in APP/PS1 mice after intranasal administration, could improve the neuronal damage and reduce inflammation levels in the hippocampus of APP/PS1 mice, and the improvement effect was significantly better than HT22-EXO. However, intranasal administration of the two exosomes did not cause depression and anxious-like phenotypes in APP/PS1 mice, nor significantly improved the short-term or spatial learning and memory ability of APP/PS1 mice, and had no significant effect on the content of Aβ plaques in the hippocampus, which also meant that MSC-EXO could use their own advantages in combination with other drugs to clear Aβ plaques. The possibility of realizing highly effective non-invasive synergistic treatment for AD provides new strategies and ideas for clinical research.

Keywords: Alzheimer’s disease, exosomes derived from mesenchymal stem cell, intranasal administration, therapy-carrier dual roles

Procedia PDF Downloads 62

Authors: B. R. Campomanes-Alvarez, P. Quiros, B. Fernandez

Abstract:

Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.

Keywords: automatic speech recognition, deep neural networks, machine learning, semi-supervised learning

Procedia PDF Downloads 339