Search results for: physics-informed neural network

Authors: Lori W. Gordon, Karen A. Jones

Abstract:

Today’s high-speed data connectivity depends on a vast global network of infrastructure across space, air, land, and sea, with undersea cable infrastructure (UCI) serving as the primary means for intercontinental and ‘long-haul’ communications. The UCI landscape is changing and includes an increasing variety of state actors, such as the growing economies of Brazil, Russia, India, China, and South Africa. Non-state commercial actors, such as hyper-scale content providers including Google, Facebook, Microsoft, and Amazon, are also seeking to control their data and networks through significant investments in submarine cables. Active investments by both state and non-state actors will invariably influence the growth, geopolitics, and security of this sector. Beyond these hyper-scale content providers, there are new commercial satellite communication providers. These new players include traditional geosynchronous (GEO) satellites that offer broad coverage, high throughput GEO satellites offering high capacity with spot beam technology, low earth orbit (LEO) ‘mega constellations’ – global broadband services. And potential new entrants such as High Altitude Platforms (HAPS) offer low latency connectivity, LEO constellations offer high-speed optical mesh networks, i.e., ‘fiber in the sky.’ This paper focuses on understanding the role of submarine cables within the larger context of the global data commons, spanning space, terrestrial, air, and sea networks, including an analysis of national security policy and geopolitical implications. As network operators and commercial and government stakeholders plan for emerging technologies and architectures, hedging risks for future connectivity will ensure that our data backbone will be secure for years to come.

Keywords: communications, global, infrastructure, technology

Procedia PDF Downloads 87

Authors: Emna Hlel, Salma Jamousi, Abdelmajid Ben Hamadou

Abstract:

Every model for knowledge representation to model real-world applications must be able to cope with the effects of uncertain phenomena. One of main defects of classical ontology is its inability to represent and reason with uncertainty. To remedy this defect, we try to propose a method to construct probabilistic ontology for integrating uncertain information in an ontology modeling a set of basic publications DBLP (Digital Bibliography & Library Project) using a probabilistic model.

Keywords: classical ontology, probabilistic ontology, uncertainty, Bayesian network

Procedia PDF Downloads 347

Authors: Ankan Roy, Niharika, Samir Kumar Patra

Abstract:

Anomalous nexus of complex topological assemblies and spatiotemporal epigenetic choreography at chromosomal territory may forms the most sophisticated regulatory layer of gene expression in cancer. Colon cancer is one of the leading malignant neoplasms of the lower gastrointestinal tract worldwide. There is still a paucity of information about the complex molecular mechanisms of colonic cancerogenesis. Bioinformatics prediction and analysis helps to identify essential genes and significant pathways for monitoring and conquering this deadly disease. The present study investigates and explores potential hub genes as biomarkers and effective therapeutic targets for colon cancer treatment. Colon cancer patient sample containing gene expression profile datasets, such as GSE44076, GSE20916, and GSE37364 were downloaded from Gene Expression Omnibus (GEO) database and thoroughly screened using the GEO2R tool and Funrich software to find out common 2 differentially expressed genes (DEGs). Other approaches, including Gene Ontology (GO) and KEGG pathway analysis, Protein-Protein Interaction (PPI) network construction and hub gene investigation, Overall Survival (OS) analysis, gene correlation analysis, methylation pattern analysis, and hub gene-Transcription factors regulatory network construction, were performed and validated using various bioinformatics tool. Initially, we identified 166 DEGs, including 68 up-regulated and 98 down-regulated genes. Up-regulated genes are mainly associated with the Cytokine-cytokine receptor interaction, IL17 signaling pathway, ECM-receptor interaction, Focal adhesion and PI3K-Akt pathway. Downregulated genes are enriched in metabolic pathways, retinol metabolism, Steroid hormone biosynthesis, and bile secretion. From the protein-protein interaction network, thirty hub genes with high connectivity are selected using the MCODE and cytoHubba plugin. Survival analysis, expression validation, correlation analysis, and methylation pattern analysis were further verified using TCGA data. Finally, we predicted COL1A1, COL1A2, COL4A1, SPP1, SPARC, and THBS2 as potential master regulators in colonic cancerogenesis. Moreover, our experimental data highlights that disruption of lipid raft and RAS/MAPK signaling cascade affects this gene hub at mRNA level. We identified COL1A1, COL1A2, COL4A1, SPP1, SPARC, and THBS2 as determinant hub genes in colon cancer progression. They can be considered as biomarkers for diagnosis and promising therapeutic targets in colon cancer treatment. Additionally, our experimental data advertise that signaling pathway act as connecting link between membrane hub and gene hub.

Keywords: hub genes, colon cancer, DNA methylation, epigenetic engineering, bioinformatic predictions

Procedia PDF Downloads 128

Authors: Riccardo Patriarca, Giulio Di Gravio, Francesco Costantino, Massimo Tronci

Abstract:

An accurate inventory management policy acquires a crucial role in the several high-availability sectors. In these sectors, due to the high-cost of spares and backorders, an (S-1, S) replenishment policy is necessary for high-availability items. The policy enables the shipment of a substitute efficient item anytime the inventory size decreases by one. This policy can be modelled following the Multi-Echelon Technique for Recoverable Item Control (METRIC). The METRIC is a system-based technique that allows defining the optimum stock level in a multi-echelon network, adopting measures in line with the decision-maker’s perspective. The METRIC defines an availability-cost function with inventory costs and required service levels, using as inputs data about the demand trend, the supplying and maintenance characteristics of the network and the budget/availability constraints. The traditional METRIC relies on the hypothesis that a Poisson distribution well represents the demand distribution in case of items with a low failure rate. However, in this research, we will explore the effects of using a Poisson distribution to model the demand of low failure rate items characterized by an irregular demand trend. This characteristic of a demand is not included in the traditional METRIC formulation leading to the need of revising its traditional formulation. Using the CV (Coefficient of Variation) and ADI (Average inter-Demand Interval) classification, we will define the inherent flaws of Poisson-based METRIC for irregular demand items, defining an innovative ad hoc distribution which can better fit the irregular demands. This distribution will allow defining proper stock levels to reduce stocking and backorder costs due to the high irregularities in the demand trend. A case study in the aviation domain will clarify the benefits of this innovative METRIC approach.

Keywords: METRIC, inventory management, irregular demand, spare parts

Procedia PDF Downloads 347

Authors: Danjuma Bawa

Abstract:

This paper aimed at exploring the capabilities of Location-Allocation model in complementing the strides of the existing physical planning models in the location and distribution of facilities for regional consumption. The paper was designed to provide a blueprint to the Nigerian government and other donor agencies especially the Fertilizer Distribution Initiative (FDI) by the federal government for the revitalization of the terrorism ravaged regions. Theoretical underpinnings of central place theory related to spatial distribution, interrelationships, and threshold prerequisites were reviewed. The study showcased how Location-Allocation Model (L-AM) alongside Central Place Theory (CPT) was applied in Geographic Information System (GIS) environment to; map and analyze the spatial distribution of settlements; exploit their physical and economic interrelationships, and to explore their hierarchical and opportunistic influences. The study was purely spatial qualitative research which largely used secondary data such as; spatial location and distribution of settlements, population figures of settlements, network of roads linking them and other landform features. These were sourced from government ministries and open source consortium. GIS was used as a tool for processing and analyzing such spatial features within the dictum of CPT and L-AM to produce a comprehensive spatial digital plan for equitable and judicious location and distribution of fertilizer deports in the study area in an optimal way. Population threshold was used as yardstick for selecting suitable settlements that could stand as service centers to other hinterlands; this was accomplished using the query syntax in ArcMapTM. ArcGISTM’ network analyst was used in conducting location-allocation analysis for apportioning of groups of settlements around such service centers within a given threshold distance. Most of the techniques and models ever used by utility planners have been centered on straight distance to settlements using Euclidean distances. Such models neglect impedance cutoffs and the routing capabilities of networks. CPT and L-AM take into consideration both the influential characteristics of settlements and their routing connectivity. The study was undertaken in two terrorism ravaged Local Government Areas of Adamawa state. Four (4) existing depots in the study area were identified. 20 more depots in 20 villages were proposed using suitability analysis. Out of the 300 settlements mapped in the study area about 280 of such settlements where optimally grouped and allocated to the selected service centers respectfully within 2km impedance cutoff. This study complements the giant strides by the federal government of Nigeria by providing a blueprint for ensuring proper distribution of these public goods in the spirit of bringing succor to these terrorism ravaged populace. This will ardently at the same time help in boosting agricultural activities thereby lowering food shortage and raising per capita income as espoused by the government.

Keywords: central place theory, GIS, location-allocation, network analysis, urban and regional planning, welfare economics

Procedia PDF Downloads 147

Authors: Getaneh Berie Tarekegn

Abstract:

A precise localization system is crucial for many artificial intelligence Internet of Things (AI-IoT) applications in the era of smart cities. Their applications include traffic monitoring, emergency alarming, environmental monitoring, location-based advertising, intelligent transportation, and smart health care. The most common method for providing continuous positioning services in outdoor environments is by using a global navigation satellite system (GNSS). Due to nonline-of-sight, multipath, and weather conditions, GNSS systems do not perform well in dense urban, urban, and suburban areas.This paper proposes a generative AI-based positioning scheme for large-scale wireless settings using fingerprinting techniques. In this article, we presented a novel semi-supervised deep convolutional generative adversarial network (S-DCGAN)-based radio map construction method for real-time device localization. We also employed a reliable signal fingerprint feature extraction method with t-distributed stochastic neighbor embedding (t-SNE), which extracts dominant features while eliminating noise from hybrid WLAN and long-term evolution (LTE) fingerprints. The proposed scheme reduced the workload of site surveying required to build the fingerprint database by up to 78.5% and significantly improved positioning accuracy. The results show that the average positioning error of GAILoc is less than 39 cm, and more than 90% of the errors are less than 82 cm. That is, numerical results proved that, in comparison to traditional methods, the proposed SRCLoc method can significantly improve positioning performance and reduce radio map construction costs.

Keywords: location-aware services, feature extraction technique, generative adversarial network, long short-term memory, support vector machine

Procedia PDF Downloads 75

Authors: Lai Yi-Hsiu

Abstract:

The present paper aims to examine the language processing of Chinese-speaking seniors with Alzheimer’s disease (AD) from the perspective of temporal cues. Twenty healthy adults, 17 healthy seniors, and 13 seniors with AD in Taiwan participated in this study to tell stories based on two sets of pictures. Nine temporal cues were fetched and analyzed. Oral productions in Mandarin Chinese were compared and discussed to examine to what extent and in what way these three groups of participants performed with significant differences. Results indicated that the age effects were significant in filled pauses. The dementia effects were significant in mean duration of pauses, empty pauses, filled pauses, lexical pauses, normalized mean duration of filled pauses and lexical pauses. The findings reported in the current paper help characterize the nature of language processing in seniors with or without AD, and contribute to the interactions between the AD neural mechanism and their temporal parameters.

Keywords: language processing, Alzheimer’s disease, Mandarin Chinese, temporal cues

Procedia PDF Downloads 446

Authors: Leandros A. Maglaras, Jianmin Jiang

Abstract:

The protection of the national infrastructures from cyberattacks is one of the main issues for national and international security. The funded European Framework-7 (FP7) research project CockpitCI introduces intelligent intrusion detection, analysis and protection techniques for Critical Infrastructures (CI). The paradox is that CIs massively rely on the newest interconnected and vulnerable Information and Communication Technology (ICT), whilst the control equipment, legacy software/hardware, is typically old. Such a combination of factors may lead to very dangerous situations, exposing systems to a wide variety of attacks. To overcome such threats, the CockpitCI project combines machine learning techniques with ICT technologies to produce advanced intrusion detection, analysis and reaction tools to provide intelligence to field equipment. This will allow the field equipment to perform local decisions in order to self-identify and self-react to abnormal situations introduced by cyberattacks. In this paper, an intrusion detection module capable of detecting malicious network traffic in a Supervisory Control and Data Acquisition (SCADA) system is presented. Malicious data in a SCADA system disrupt its correct functioning and tamper with its normal operation. OCSVM is an intrusion detection mechanism that does not need any labeled data for training or any information about the kind of anomaly is expecting for the detection process. This feature makes it ideal for processing SCADA environment data and automates SCADA performance monitoring. The OCSVM module developed is trained by network traces off line and detects anomalies in the system real time. The module is part of an IDS (intrusion detection system) developed under CockpitCI project and communicates with the other parts of the system by the exchange of IDMEF messages that carry information about the source of the incident, the time and a classification of the alarm.

Keywords: cyber-security, SCADA systems, OCSVM, intrusion detection

Procedia PDF Downloads 552

Authors: Hamed Saremi, Suzan Taghavy, Seyed Mohammad Hanif Sanjari, Mostafa Kahali

Abstract:

One of the challenges in manufacturing and service companies to provide a product or service is recognized Brand to consumers in target markets. They provide most of their processes under the same capacity. But the constant threat of devastating internal and external resources to prevent a rise Brands and more companies are recognizing the stages are bankrupt. This paper has tried to identify and analyze effective indicators of brand equity and focuses on indicators and presents a model of intelligent create a model to prevent possible damage. In this study, the identified indicators of brand equity are based on literature study and according to expert opinions, set of indicators By techniques DEMATEL Then to used Multi-Step Adaptive Neural-Fuzzy Inference system (ANFIS) to design a multi-stage intelligent system for assessment of brand equity.

Keywords: brand, cosmetic product, ANFIS, DEMATEL

Procedia PDF Downloads 417

Authors: Jiří Barta, Tomáš Ludík, Jiří Urbánek

Abstract:

The topics of disaster and emergency management are highly debated among experts. Fast communication will help to deal with emergencies. Problem is with the network connection and data exchange. The paper suggests a solution, which allows possibilities and perspectives of new flexible communication platform to the protection of communication systems for crisis management. This platform is used for everyday communication and communication in crisis situations too.

Keywords: crisis management, information systems, interoperability, crisis communication, security environment, communication platform

Procedia PDF Downloads 475

Authors: Bandhan Dey, Muhsina Bintoon Yiasha, Gulam Sulaman Choudhury

Abstract:

Chest disease is one of the most problematic ailments in our regular life. There are many known chest diseases out there. Diagnosing them correctly plays a vital role in the process of treatment. There are many methods available explicitly developed for different chest diseases. But the most common approach for diagnosing these diseases is through X-ray. In this paper, we proposed a multi-classification deep learning model for diagnosing COVID-19, lung cancer, pneumonia, tuberculosis, and atelectasis from chest X-rays. In the present work, we used the transfer learning method for better accuracy and fast training phase. The performance of three architectures is considered: InceptionV3, VGG-16, and VGG-19. We evaluated these deep learning architectures using public digital chest x-ray datasets with six classes (i.e., COVID-19, lung cancer, pneumonia, tuberculosis, atelectasis, and normal). The experiments are conducted on six-classification, and we found that VGG16 outperforms other proposed models with an accuracy of 95%.

Keywords: deep learning, image classification, X-ray images, Tensorflow, Keras, chest diseases, convolutional neural networks, multi-classification

Procedia PDF Downloads 92

Authors: Ardalan Tofighi Soleimandarabi

Abstract:

Artificial intelligence has revolutionized the concrete construction industry and improved processes by increasing efficiency, accuracy, and sustainability. This article examines the applications of artificial intelligence in predicting the compressive strength of concrete, optimizing mixing plans, and improving structural health monitoring systems. Artificial intelligence-based models, such as artificial neural networks (ANN) and combined machine learning techniques, have shown better performance than traditional methods in predicting concrete properties. In addition, artificial intelligence systems have made it possible to improve quality control and real-time monitoring of structures, which helps in preventive maintenance and increases the life of infrastructure. Also, the use of artificial intelligence plays an effective role in sustainable construction by optimizing material consumption and reducing waste. Although the implementation of artificial intelligence is associated with challenges such as high initial costs and the need for specialized training, it will create a smarter, more sustainable, and more affordable future for concrete structures.

Keywords: artificial intelligence, concrete construction, compressive strength prediction, structural health monitoring, stability

Procedia PDF Downloads 15

Authors: Hong Yu, Dirk Heider, Suresh Advani

Abstract:

Increasing demands for high strength and lightweight materials in aircraft industry prompted the wide use of carbon composites in recent decades. Carbon composite laminates used on aircraft structures are subject to lightning strikes. Unlike its metal/alloy counterparts, carbon fiber reinforced composites demonstrate smaller electrical conductivity, yielding more severe damages due to Joule heating. The anisotropic nature of composite laminates makes the electrical and thermal conduction within carbon composite laminates even more complicated. Good understanding of the electrical conduction behavior of carbon composites is the key to effective lightning protection design. The goal of this study is to numerically and experimentally investigate the impact of ultra-high temperature induced by simulated lightning strike on the electrical conduction of carbon composites. A lightning simulator is designed to apply standard lightning current waveform to composite laminates. Multiple carbon composite laminates made from IM7 and AS4 carbon fiber are tested and the transient resistance data is recorded. A microstructure based resistor network model is developed to describe the electrical and thermal conduction behavior, with consideration of temperature dependent material properties. Material degradations such as thermal and electrical breakdown are also modeled to include the effect of high current and high temperature induced by lightning strikes. Good match between the simulation results and experimental data indicates that the developed model captures the major conduction mechanisms. A parametric study is then conducted using the validated model to investigate the effect of system parameters such as fiber volume fraction, inter-ply interface quality, and lightning current waveforms.

Keywords: carbon composite, joule heating, lightning strike, resistor network

Procedia PDF Downloads 228

Authors: Qianwei Cheng, A. K. M. Mahbubur Rahman, Anis Sarker, Abu Bakar Siddik Nayem, Ovi Paul, Amin Ahsan Ali, M. Ashraful Amin, Ryosuke Shibasaki, Moinul Zaber

Abstract:

Thomas Friedman, in his famous book, argued that the world in this 21st century is flat and will continue to be flatter. This is attributed to rapid globalization and the interdependence of humanity that engendered tremendous in-flow of human migration towards the urban spaces. In order to keep the urban environment sustainable, policy makers need to plan based on extensive analysis of the urban environment. With the advent of high definition satellite images, high resolution data, computational methods such as deep neural network analysis, and hardware capable of high-speed analysis; urban planning is seeing a paradigm shift. Legacy data on urban environments are now being complemented with high-volume, high-frequency data. However, the first step of understanding urban space lies in useful categorization of the space that is usable for data collection, analysis, and visualization. In this paper, we propose a pragmatic categorization method that is readily usable for machine analysis and show applicability of the methodology on a developing world setting. Categorization to plan sustainable urban spaces should encompass the buildings and their surroundings. However, the state-of-the-art is mostly dominated by classification of building structures, building types, etc. and largely represents the developed world. Hence, these methods and models are not sufficient for developing countries such as Bangladesh, where the surrounding environment is crucial for the categorization. Moreover, these categorizations propose small-scale classifications, which give limited information, have poor scalability and are slow to compute in real time. Our proposed method is divided into two steps-categorization and automation. We categorize the urban area in terms of informal and formal spaces and take the surrounding environment into account. 50 km × 50 km Google Earth image of Dhaka, Bangladesh was visually annotated and categorized by an expert and consequently a map was drawn. The categorization is based broadly on two dimensions-the state of urbanization and the architectural form of urban environment. Consequently, the urban space is divided into four categories: 1) highly informal area; 2) moderately informal area; 3) moderately formal area; and 4) highly formal area. In total, sixteen sub-categories were identified. For semantic segmentation and automatic categorization, Google’s DeeplabV3+ model was used. The model uses Atrous convolution operation to analyze different layers of texture and shape. This allows us to enlarge the field of view of the filters to incorporate larger context. Image encompassing 70% of the urban space was used to train the model, and the remaining 30% was used for testing and validation. The model is able to segment with 75% accuracy and 60% Mean Intersection over Union (mIoU). In this paper, we propose a pragmatic categorization method that is readily applicable for automatic use in both developing and developed world context. The method can be augmented for real-time socio-economic comparative analysis among cities. It can be an essential tool for the policy makers to plan future sustainable urban spaces.

Keywords: semantic segmentation, urban environment, deep learning, urban building, classification

Procedia PDF Downloads 191

Authors: Viktor R. Stoynov, Zlatka V. Valkova-Jarvis

Abstract:

The intelligent management and optimisation of radio resource technologies will lead to a considerable improvement in the overall performance in Next Generation Networks (NGNs). Carrier Aggregation (CA) technology, also known as Spectrum Aggregation, enables more efficient use of the available spectrum by combining multiple Component Carriers (CCs) in a virtual wideband channel. LTE-A (Long Term Evolution–Advanced) CA technology can combine multiple adjacent or separate CCs in the same band or in different bands. In this way, increased data rates and dynamic load balancing can be achieved, resulting in a more reliable and efficient operation of mobile networks and the enabling of high bandwidth mobile services. In this paper, several distinct CA deployment strategies for the utilisation of spectrum bands are compared in indoor-outdoor scenarios, simulated via the recently-developed Realistic Indoor Environment Generator (RIEG). We analyse the performance of the User Equipment (UE) by integrating the average throughput, the level of fairness of radio resource allocation, and other parameters, into one summative assessment termed a Comparative Factor (CF). In addition, comparison of non-CA and CA indoor mobile networks is carried out under different load conditions: varying numbers and positions of UEs. The experimental results demonstrate that the CA technology can improve network performance, especially in the case of indoor scenarios. Additionally, we show that an increase of carrier frequency does not necessarily lead to improved CF values, due to high wall-penetration losses. The performance of users under bad-channel conditions, often located in the periphery of the cells, can be improved by intelligent CA location. Furthermore, a combination of such a deployment and effective radio resource allocation management with respect to user-fairness plays a crucial role in improving the performance of LTE-A networks.

Keywords: comparative factor, carrier aggregation, indoor mobile network, resource allocation

Procedia PDF Downloads 178

Authors: Swee Yang Low, Dong Eon Kim, Canh Tien Trinh Nguyen, Yixiong Cai, Shie-Yui Liong

Abstract:

Nee Soon swamp forest is one of the last vestiges of tropical wetland in Singapore. Understanding the hydrological regime of the swamp forest and implications for water quality is critical to guide stakeholders in implementing effective measures to preserve the wetland against anthropogenic impacts. In particular, although current field measurement data do not indicate a concern with organic pollution, reviewing the ways in which the wetland responds to elevated organic waste influx (and the corresponding impact on dissolved oxygen, DO) can help identify potential hotspots, and the impact on the outflow from the catchment which drains into downstream controlled watercourses. An integrated water quality model is therefore developed in this study to investigate spatial and temporal concentrations of DO levels and organic pollution (as quantified by biochemical oxygen demand, BOD) within the catchment’s river network under hypothetical, projected scenarios of spiked upstream inflow. The model was developed using MIKE HYDRO for modelling the study domain, as well as the MIKE ECO Lab numerical laboratory for characterising water quality processes. Model parameters are calibrated against time series of observed discharges at three measurement stations along the river network. Over a simulation period of April 2014 to December 2015, the calibrated model predicted that a continuous spiked inflow of 400 mg/l BOD will elevate downstream concentrations at the catchment outlet to an average of 12 mg/l, from an assumed nominal baseline BOD of 1 mg/l. Levels of DO were decreased from an initial 5 mg/l to 0.4 mg/l. Though a scenario of spiked organic influx at the swamp forest’s undeveloped upstream sub-catchments is currently unlikely to occur, the outcomes nevertheless will be beneficial for future planning studies in understanding how the water quality of the catchment will be impacted should urban redevelopment works be considered around the swamp forest.

Keywords: hydrology, modeling, water quality, wetland

Procedia PDF Downloads 140

Authors: Armineh Nourbakhsh, Sameena Shah, Carolyn Rose

Abstract:

Layout-aware language models have been used to create multimodal representations for documents that are in image form, achieving relatively high accuracy in document understanding tasks. However, the large number of parameters in the resulting models makes building and using them prohibitive without access to high-performing processing units with large memory capacity. We propose an alternative approach that can create efficient representations without the need for a neural visual backbone. This leads to an 80% reduction in the number of parameters compared to the smallest SOTA model, widely expanding applicability. In addition, our layout embeddings are pre-trained on spatial and visual cues alone and only fused with text embeddings in downstream tasks, which can facilitate applicability to low-resource of multi-lingual domains. Despite using 2.5% of training data, we show competitive performance on two form understanding tasks: semantic labeling and link prediction.

Keywords: layout understanding, form understanding, multimodal document understanding, bias-augmented attention

Procedia PDF Downloads 148

Authors: Anshika Kankane, Dongshik Kang

Abstract:

Marine debris has a significant influence on coastal environments, damaging biodiversity, and causing loss and damage to marine and ocean sector. A functional cost-effective and automatic approach has been used to look up at this problem. Computer vision combined with a deep learning-based model is being proposed to identify and categorize marine debris of seven kinds on different beach locations of Japan. This research compares state-of-the-art deep learning models with a suggested model architecture that is utilized as a feature extractor for debris categorization. The model is being proposed to detect seven categories of litter using a manually constructed debris dataset, with the help of Mask R-CNN for instance segmentation and a shape matching network called HOGShape, which can then be cleaned on time by clean-up organizations using warning notifications of the system. The manually constructed dataset for this system is created by annotating the images taken by fixed KaKaXi camera using CVAT annotation tool with seven kinds of category labels. A pre-trained HOG feature extractor on LIBSVM is being used along with multiple templates matching on HOG maps of images and HOG maps of templates to improve the predicted masked images obtained via Mask R-CNN training. This system intends to timely alert the cleanup organizations with the warning notifications using live recorded beach debris data. The suggested network results in the improvement of misclassified debris masks of debris objects with different illuminations, shapes, viewpoints and litter with occlusions which have vague visibility.

Keywords: computer vision, debris, deep learning, fixed live camera images, histogram of gradients feature extractor, instance segmentation, manually annotated dataset, multiple template matching

Procedia PDF Downloads 106

Authors: Elmahdi Mohammadine, Ahouzi Esmail, Najid Abdellah

Abstract:

This contribution aims to present a new protected hybrid WDM/TDM PON architecture using Wavelength Selective Switches and Optical Line Protection devices. The objective from using these technologies is to improve flexibility and enhance the protection of GPON networks.

Keywords: Wavlenght Division Multiplexed Passive Optical Network (WDM-PON), Time Division Multiplexed PON (TDM-PON), architecture, Protection, Wavelength Selective Switches (WSS), Optical Line Protection (OLP)

Procedia PDF Downloads 542

Authors: M. Ahmadi, Sh. Damanabi, F. Sadoughi

Abstract:

National Health Information System plays an important role in ensuring timely and reliable access to Health information which is essential for strategic and operational decisions that improve health, quality and effectiveness of health care. In other words, by using the National Health information system you can improve the quality of health data, information and knowledge used to support decision making at all levels and areas of the health sector. Since full identification of the components of this system for better planning and management influential factors of performance seems necessary, therefore, in this study, different attitudes towards components of this system are explored comparatively. Methods: This is a descriptive and comparative kind of study. The society includes printed and electronic documents containing components of the national health information system in three parts: input, process, and output. In this context, search for information using library resources and internet search were conducted and data analysis was expressed using comparative tables and qualitative data. Results: The findings showed that there are three different perspectives presenting the components of national health information system, Lippeveld, Sauerborn, and Bodart Model in 2000, Health Metrics Network (HMN) model from World Health Organization in 2008 and Gattini’s 2009 model. All three models outlined above in the input (resources and structure) require components of management and leadership, planning and design programs, supply of staff, software and hardware facilities, and equipment. In addition, in the ‘process’ section from three models, we pointed up the actions ensuring the quality of health information system and in output section, except Lippeveld Model, two other models consider information products, usage and distribution of information as components of the national health information system. Conclusion: The results showed that all the three models have had a brief discussion about the components of health information in input section. However, Lippeveld model has overlooked the components of national health information in process and output sections. Therefore, it seems that the health measurement model of network has a comprehensive presentation for the components of health system in all three sections-input, process, and output.

Keywords: National Health Information System, components of the NHIS, Lippeveld Model

Procedia PDF Downloads 421

Authors: S. Kawther, C. Marshall

Abstract:

Educational neuroscience has the intention of transforming research findings of the underpinning neural processes of learning to educational practices. A main criticism of the field, hitherto, is that less focus has been put on studying the in-progress practical application of these findings. Therefore, this study aims to gain a better understanding of teachers’ perceptions of the practical application and utilization of brain knowledge. This was approached by investigating the answer to 'What are the facilitators and barriers for bringing research from neuroscience to bear on education?'. Following a qualitative design, semi-structured interviews were conducted with 12 teachers who had a proficient course in educational neuroscience. Thematic analysis was performed on the transcribed data applying Braun & Clark’s steps. Findings emerged with four main themes: time, knowledge, teacher’s involvement, and system. These themes revealed that some effective brain-based practices are being engaged in by the teachers. However, the lack of guidance and challenges regarding this implementation were also found. This study discusses findings in light of the development of educational neuroscience implementation.

Keywords: brain-based, educational neuroscience, neuroeducation, neuroscience-informed

Procedia PDF Downloads 167

Authors: Hadeer M. Sayed, Hesham E. El Deeb, Shereen A. Taie

Abstract:

In the big data era, we are facing a diversity of datasets from different sources in different domains that describe a single life event. These datasets consist of multiple modalities, each of which has a different representation, distribution, scale, and density. Multimodal fusion is the concept of integrating information from multiple modalities in a joint representation with the goal of predicting an outcome through a classification task or regression task. In this paper, multimodal fusion techniques are classified into two main classes: model-agnostic techniques and model-based approaches. It provides a comprehensive study of recent research in each class and outlines the benefits and limitations of each of them. Furthermore, the audiovisual speech recognition task is expressed as a case study of multimodal data fusion approaches, and the open issues through the limitations of the current studies are presented. This paper can be considered a powerful guide for interested researchers in the field of multimodal data fusion and audiovisual speech recognition particularly.

Keywords: multimodal data, data fusion, audio-visual speech recognition, neural networks

Procedia PDF Downloads 112

Authors: Eiji Nakamachi, Ryota Sakiyama, Koji Yamamoto, Yusuke Morita, Hidetoshi Sakamoto

Abstract:

The regeneration of injured central nerve network caused by the cerebrovascular accidents is difficult, because of poor regeneration capability of central nerve system composed of the brain and the spinal cord. Recently, new regeneration methods such as transplant of nerve cells and supply of nerve nutritional factor were proposed and examined. However, there still remain many problems with the canceration of engrafted cells and so on and it is strongly required to establish an efficacious treating method of a central nerve system. Blackman proposed the electromagnetic stimulation method to enhance the axonal nerve extension. In this study, we try to design and fabricate a new three-dimensional (3D) bio-reactor, which can load a uniform AC magnetic field stimulation on PC12 cells in the extracellular environment for enhancement of an axonal nerve extension and 3D nerve network generation. Simultaneously, we measure the morphology of PC12 cell bodies, axons, and dendrites by the multiphoton excitation fluorescence microscope (MPM) and evaluate the effectiveness of the uniform AC magnetic stimulation to enhance the axonal nerve extension. Firstly, we designed and fabricated the uniform AC magnetic field stimulation bio-reactor. For the AC magnetic stimulation system, we used the laminated silicon steel sheets for a yoke structure of 3D chamber, which had a high magnetic permeability. Next, we adopted the pole piece structure and installed similar specification coils on both sides of the yoke. We searched an optimum pole piece structure using the magnetic field finite element (FE) analyses and the response surface methodology. We confirmed that the optimum 3D chamber structure showed a uniform magnetic flux density in the PC12 cell culture area by using FE analysis. Then, we fabricated the uniform AC magnetic field stimulation bio-reactor by adopting analytically determined specifications, such as the size of chamber and electromagnetic conditions. We confirmed that measurement results of magnetic field in the chamber showed a good agreement with FE results. Secondly, we fabricated a dish, which set inside the uniform AC magnetic field stimulation of bio-reactor. PC12 cells were disseminated with collagen gel and could be 3D cultured in the dish. The collagen gel were poured in the dish. The collagen gel, which had a disk shape of 6 mm diameter and 3mm height, was set on the membrane filter, which was located at 4 mm height from the bottom of dish. The disk was full filled with the culture medium inside the dish. Finally, we evaluated the effectiveness of the uniform AC magnetic field stimulation to enhance the nurve axonal extension. We confirmed that a 6.8 increase in the average axonal extension length of PC12 under the uniform AC magnetic field stimulation at 7 days culture in our bio-reactor, and a 24.7 increase in the maximum axonal extension length. Further, we confirmed that a 60 increase in the number of dendrites of PC12 under the uniform AC magnetic field stimulation. Finally, we confirm the availability of our uniform AC magnetic stimulation bio-reactor for the nerve axonal extension and the nerve network generation.

Keywords: nerve regeneration, axonal extension , PC12 cell, magnetic field, three-dimensional bio-reactor

Procedia PDF Downloads 168

Authors: Joanna Skopinska-Wisniewska, Anna Bajek, Marta Ziegler-Borowska, Alina Sionkowska

Abstract:

Hydrogels are a class of materials widely used in medicine for many years. Proteins, such as collagen, due to the presence of a large number of functional groups are easily wettable by polar solvents and can create hydrogels. The supramolecular network capable to swelling is created by cross-linking of the biopolymers using various reagents. Many cross-linking agents has been tested for last years, however, researchers still are looking for a new, more secure reactants. Squaric acid, 3,4-dihydroxy 3-cyclobutene 1,2- dione, is a very strong acid, which possess flat and rigid structure. Due to the presence of two carboxyl groups the squaric acid willingly reacts with amino groups of collagen. The main purpose of this study was to investigate the influence of addition of squaric acid on the chemical, physical and biological properties of collagen materials. The collagen type I was extracted from rat tail tendons and 1% solution in 0.1M acetic acid was prepared. The samples were cross-linked by the addition of 5%, 10% and 20% of squaric acid. The mixtures of all reagents were incubated 30 min on magnetic stirrer and then dialyzed against deionized water. The FTIR spectra show that the collagen structure is not changed by cross-linking by squaric acid. Although the mechanical properties of the collagen material deteriorate, the temperature of thermal denaturation of collagen increases after cross-linking, what indicates that the protein network was created. The lyophilized collagen gels exhibit porous structure and the pore size decreases with the higher addition of squaric acid. Also the swelling ability is lower after the cross-linking. The in vitro study demonstrates that the materials are attractive for 3T3 cells. The addition of squaric acid causes formation of cross-ling bonds in the collagen materials and the transparent, stiff hydrogels are obtained. The changes of physicochemical properties of the material are typical for cross-linking process, except mechanical properties – it requires further experiments. However, the results let us to conclude that squaric acid is a suitable cross-linker for protein materials for medicine and tissue engineering.

Keywords: collagen, squaric acid, cross-linking, hydrogel

Procedia PDF Downloads 388

Authors: Mohammad Hossein Mohammadi Sanjani, Ashknaz Oraee, Oriol Gomis Bellmunt, Vinicius Albernaz Lacerda Freitas

Abstract:

The use of distributed and renewable sources in power systems has grown significantly, recently. One the most popular sources are wind farms which have grown massively. However, ¬wind farms are connected to the grid, this can cause problems such as reduced voltage stability, frequency fluctuations and reduced dynamic stability. Variable speed generators (asynchronous) are used due to the uncontrollability of wind speed specially Doubley Fed Induction Generators (DFIG). The most important disadvantage of DFIGs is its sensitivity to voltage drop. In the case of faults, a large volume of reactive power is induced therefore, use of FACTS devices such as SVC and STATCOM are suitable for improving system output performance. They increase the capacity of lines and also passes network fault conditions. In this paper, in addition to modeling the reactive power control system in a DFIG with converter, FACTS devices have been used in a DFIG wind turbine to improve the stability of the power system containing two synchronous sources. In the following paper, recent optimal control systems have been designed to minimize fluctuations caused by system disturbances, for FACTS devices employed. For this purpose, a suitable method for the selection of nine parameters for MPSH-phase-post-phase compensators of reactive power compensators is proposed. The design algorithm is formulated ¬¬as an optimization problem searching for optimal parameters in the controller. Simulation results show that the proposed controller Improves the stability of the network and the fluctuations are at desired speed.

Keywords: renewable energy sources, optimization wind power plant, stability, reactive power compensator, double-feed induction generator, optimal control, genetic algorithm

Procedia PDF Downloads 95

Authors: Azene Zenebe

Abstract:

Deep learning is a subset of machine learning which incorporates techniques for the construction of artificial neural networks and found to be useful for modeling complex problems with large dataset. Deep learning requires a very high power computational and longer time for training. By aggregating computing power, high performance computer (HPC) has emerged as an approach to resolving advanced problems and performing data-driven research activities. Cyber threat intelligence (CIT) is actionable information or insight an organization or individual uses to understand the threats that have, will, or are currently targeting the organization. Results of review of literature will be presented along with results of experimental study that compares the performance of tree-based and function-base machine learning including deep learning algorithms using secondary dataset collected from darknet.

Keywords: deep-learning, cyber security, cyber threat modeling, tree-based machine learning, function-based machine learning, data science

Procedia PDF Downloads 154

Authors: Chad Goldsworthy, B. Rajeswari Matam

Abstract:

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

Keywords: convolutional neural networks, data imbalance, deep learning, focal loss, species classification, wildlife conservation

Procedia PDF Downloads 191

Authors: Gabrielle Robertson, Matthew Downs, Joseph Dagher

Abstract:

Iron deposition in the brain has been linked with a host of neurological disorders such as Alzheimer’s, Parkinson’s, and Multiple Sclerosis. While some treatment options exist, there are no objective measurement tools that allow for the monitoring of iron levels in the brain in vivo. An emerging Magnetic Resonance Imaging (MRI) method has been recently proposed to deduce iron concentration through quantitative measurement of magnetic susceptibility. This is a multi-step process that involves repeated modeling of physical processes via approximate numerical solutions. For example, the last two steps of this Quantitative Susceptibility Mapping (QSM) method involve I) mapping magnetic field into magnetic susceptibility and II) mapping magnetic susceptibility into iron concentration. Process I involves solving an ill-posed inverse problem by using regularization via injection of prior belief. The end result from Process II highly depends on the model used to describe the molecular content of each voxel (type of iron, water fraction, etc.) Due to these factors, the accuracy and repeatability of QSM have been an active area of research in the MRI and medical imaging community. This work aims to estimate iron concentration in the brain via a single step. A synthetic numerical model of the human head was created by automatically and manually segmenting the human head on a high-resolution grid (640x640x640, 0.4mm³) yielding detailed structures such as microvasculature and subcortical regions as well as bone, soft tissue, Cerebral Spinal Fluid, sinuses, arteries, and eyes. Each segmented region was then assigned tissue properties such as relaxation rates, proton density, electromagnetic tissue properties and iron concentration. These tissue property values were randomly selected from a Probability Distribution Function derived from a thorough literature review. In addition to having unique tissue property values, different synthetic head realizations also possess unique structural geometry created by morphing the boundary regions of different areas within normal physical constraints. This model of the human brain is then used to create synthetic MRI measurements. This is repeated thousands of times, for different head shapes, volume, tissue properties and noise realizations. Collectively, this constitutes a training-set that is similar to in vivo data, but larger than datasets available from clinical measurements. This 3D convolutional U-Net neural network architecture was used to train data-driven Deep Learning models to solve for iron concentrations from raw MRI measurements. The performance was then tested on both synthetic data not used in training as well as real in vivo data. Results showed that the model trained on synthetic MRI measurements is able to directly learn iron concentrations in areas of interest more effectively than other existing QSM reconstruction methods. For comparison, models trained on random geometric shapes (as proposed in the Deep QSM method) are less effective than models trained on realistic synthetic head models. Such an accurate method for the quantitative measurement of iron deposits in the brain would be of important value in clinical studies aiming to understand the role of iron in neurological disease.

Keywords: magnetic resonance imaging, MRI, iron deposition, machine learning, quantitative susceptibility mapping

Procedia PDF Downloads 136

Authors: O. Gsib, U. Peirera, C. Egles, S. A. Bencherif

Abstract:

In skin regenerative medicine, one of the critical issues is to produce a three-dimensional scaffold with optimized porosity for dermal fibroblast infiltration and neovascularization, which exhibits high mechanical properties and displays sufficient wound healing characteristics. In this study, we report on the synthesis and characterization of macroporous sequential interpenetrating polymer networks (IPNs) combining skin wound healing properties of fibrin with the excellent physical properties of polyethylene glycol (PEG). Fibrin fibers serve as a provisional biologically active network to promote cell adhesion and proliferation while PEG provides the mechanical stability to maintain the entire 3D construct. After having modified both PEG and Serum Albumin (used for promoting enzymatic degradability) by adding methacrylate residues (PEGDM and SAM, respectively), Fibrin/PEGDM-SAM sequential IPNs were synthesized as follows: Macroporous sponges were first produced from PEGDM-SAM hydrogels by a freeze-drying technique and then rehydrated by adding the fibrin precursors. Environmental Scanning Electron Microscopy (ESEM) and Confocal Laser Scanning Microscopy (CLSM) were used to characterize their microstructure. Human dermal fibroblasts were cultivated during one week in the constructs and different cell culture parameters (viability, morphology, proliferation) were evaluated. Subcutaneous implantations of the scaffolds were conducted on five-week old male nude mice to investigate their biocompatibility in vivo. We successfully synthesized interconnected and macroporous Fibrin/PEGDM-SAM sequential IPNs. The viability of primary dermal fibroblasts was well maintained (above 90%) after 2 days of culture. Cells were able to adhere, spread and proliferate in the scaffolds suggesting the suitable porosity and intrinsic biologic properties of the constructs. The fibrin network adopted a spider web shape that covered partially the pores allowing easier cell infiltration into the macroporous structure. To further characterize the in vitro cell behavior, cell proliferation (EdU incorporation, MTS assay) is being studied. Preliminary histological analysis of animal studies indicated the persistence of hydrogels even after one-month post implantation and confirmed the absence of inflammation response, good biocompatibility and biointegration of our scaffolds within the surrounding tissues. These results suggest that our Fibrin/PEGDM-SAM IPNs could be considered as potential candidates for dermis regenerative medicine. Histological analysis will be completed to further assess scaffold remodeling including de novo extracellular matrix protein synthesis and early stage angiogenesis analysis. Compression measurements will be conducted to investigate the mechanical properties.

Keywords: fibrin, hydrogels for dermal reconstruction, polyethylene glycol, semi-interpenetrating polymer network

Procedia PDF Downloads 236

Authors: Sham Kidane, Fitsum Gaim, Ibrahim Abdella, Sirak Asmerom, Yoel Ghebrihiwot, Simon Mulugeta, Natnael Ambassager

Abstract:

The lack of annotated datasets is a bottleneck to the progress of NLP in low-resourced languages. The work presented here consists of large-scale annotated datasets and models for the named entity recognition (NER) system for the Tigrinya language. Our manually constructed corpus comprises over 340K words tagged for NER, with over 118K of the tokens also having parts-of-speech (POS) tags, annotated with 12 distinct classes of entities, represented using several types of tagging schemes. We conducted extensive experiments covering convolutional neural networks and transformer models; the highest performance achieved is 88.8% weighted F1-score. These results are especially noteworthy given the unique challenges posed by Tigrinya’s distinct grammatical structure and complex word morphologies. The system can be an essential building block for the advancement of NLP systems in Tigrinya and other related low-resourced languages and serve as a bridge for cross-referencing against higher-resourced languages.

Keywords: Tigrinya NER corpus, TiBERT, TiRoBERTa, BiLSTM-CRF

Procedia PDF Downloads 130