Search results for: reward processing

Authors: Sourabh Jain, S. S. Jain

Abstract:

Intelligent transportation system (ITS) is the application of technologies for developing a user–friendly transportation system to extend the safety and efficiency of urban transportation systems in developing countries. These systems involve vehicles, drivers, passengers, road operators, managers of transport services; all interacting with each other and the surroundings to boost the security and capacity of road systems. The goal of urban corridor management using ITS in road transport is to achieve improvements in mobility, safety, and the productivity of the transportation system within the available facilities through the integrated application of advanced monitoring, communications, computer, display, and control process technologies, both in the vehicle and on the road. Intelligent transportation system is a product of the revolution in information and communications technologies that is the hallmark of the digital age. The basic ITS technology is oriented on three main directions: communications, information, integration. Information acquisition (collection), processing, integration, and sorting are the basic activities of ITS. In the paper, attempts have been made to present the endeavor that was made to interpret and evaluate the performance of the 27.4 Km long study corridor having eight intersections and four flyovers. The corridor consisting of six lanes as well as eight lanes divided road network. Two categories of data have been collected such as traffic data (traffic volume, spot speed, delay) and road characteristics data (no. of lanes, lane width, bus stops, mid-block sections, intersections, flyovers). The instruments used for collecting the data were video camera, stop watch, radar gun, and mobile GPS (GPS tracker lite). From the analysis, the performance interpretations incorporated were the identification of peak and off-peak hours, congestion and level of service (LOS) at midblock sections and delay followed by plotting the speed contours. The paper proposed the urban corridor management strategies based on sensors integrated into both vehicles and on the roads that those have to be efficiently executable, cost-effective, and familiar to road users. It will be useful to reduce congestion, fuel consumption, and pollution so as to provide comfort, safety, and efficiency to the users.

Keywords: ITS strategies, congestion, planning, mobility, safety

Procedia PDF Downloads 179

Authors: Anoop T. R., Otman Basir, Robert F. Hess, Eileen E. Birch, Brooke A. Koritala, Reed M. Jost, Becky Luu, David Stager, Ben Thompson

Abstract:

Strabismus refers to a misalignment of the eyes. Early detection and treatment of strabismus in childhood can prevent the development of permanent vision loss due to abnormal development of visual brain areas. We developed a two-stage method for strabismus detection and classification based on photographs of the face. The first stage detects the presence or absence of strabismus, and the second stage classifies the type of strabismus. The first stage comprises face detection using Haar cascade, facial landmark estimation, face alignment, aligned face landmark detection, segmentation of the eye region, and detection of strabismus using VGG 16 convolution neural networks. Face alignment transforms the face to a canonical pose to ensure consistency in subsequent analysis. Using facial landmarks, the eye region is segmented from the aligned face and fed into a VGG 16 CNN model, which has been trained to classify strabismus. The CNN determines whether strabismus is present and classifies the type of strabismus (exotropia, esotropia, and vertical deviation). If stage 1 detects strabismus, the eye region image is fed into stage 2, which starts with the estimation of pupil center coordinates using mask R-CNN deep neural networks. Then, the distance between the pupil coordinates and eye landmarks is calculated along with the angle that the pupil coordinates make with the horizontal and vertical axis. The distance and angle information is used to characterize the degree and direction of the strabismic eye misalignment. This model was tested on 100 clinically labeled images of children with (n = 50) and without (n = 50) strabismus. The True Positive Rate (TPR) and False Positive Rate (FPR) of the first stage were 94% and 6% respectively. The classification stage has produced a TPR of 94.73%, 94.44%, and 100% for esotropia, exotropia, and vertical deviations, respectively. This method also had an FPR of 5.26%, 5.55%, and 0% for esotropia, exotropia, and vertical deviation, respectively. The addition of one more feature related to the location of corneal light reflections may reduce the FPR, which was primarily due to children with pseudo-strabismus (the appearance of strabismus due to a wide nasal bridge or skin folds on the nasal side of the eyes).

Keywords: strabismus, deep neural networks, face detection, facial landmarks, face alignment, segmentation, VGG 16, mask R-CNN, pupil coordinates, angle deviation, horizontal and vertical deviation

Procedia PDF Downloads 93

Authors: Marzouk Benali, Jawad Jeaidi, Behrang Mansoornejad, Olumoye Ajao, Banafsheh Gilani, Nima Ghavidel Mehr

Abstract:

Canadian forest industry is seeking to identify and implement transformational strategies for enhanced financial performance through the emerging bioeconomy or more specifically through the concept of the biorefinery. For example, processing forest residues or surplus of biomass available on the mill sites for the production of biofuels, biochemicals and/or biomaterials is one of the attractive strategies along with traditional wood and paper products and cogenerated energy. There are many possible process-product biorefinery pathways, each associated with specific product portfolios with different levels of risk. Thus, it is not obvious which unique strategy forest industry should select and implement. Therefore, there is a need for analytical and design tools that enable evaluating biorefinery strategies based on a set of criteria considering a perspective of sustainability over the short and long terms, while selecting the existing core products as well as selecting the new product portfolio. In addition, it is critical to assess the manufacturing flexibility to internalize the risk from market price volatility of each targeted bio-based product in the product portfolio, prior to invest heavily in any biorefinery strategy. The proposed paper will focus on introducing a systematic methodology for designing integrated biorefineries using process systems engineering tools as well as a multi-criteria decision making framework to put forward the most effective biorefinery strategies that fulfill the needs of the forest industry. Topics to be covered will include market analysis, techno-economic assessment, cost accounting, energy integration analysis, life cycle assessment and supply chain analysis. This will be followed by describing the vision as well as the key features and functionalities of the I-BIOREF software platform, developed by CanmetENERGY of Natural Resources Canada. Two industrial case studies will be presented to support the robustness and flexibility of I-BIOREF software platform: i) An integrated Canadian Kraft pulp mill with lignin recovery process (namely, LignoBoost™); ii) A standalone biorefinery based on ethanol-organosolv process.

Keywords: biorefinery strategies, bioproducts, co-production, multi-criteria decision making, tool

Procedia PDF Downloads 232

Authors: Jitendra Pratap, Jonathan Sivyer

Abstract:

Introduction: Dual energy/Spectral CT scanning permits simultaneous acquisition of two x-ray spectra datasets and can complement radiological diagnosis by allowing tissue characterisation (e.g., uric acid vs. non-uric acid renal stones), enhancing structures (e.g. boost iodine signal to improve contrast resolution), and quantifying substances (e.g. iodine density). However, the latter showed inconsistent results between the 2 main modes of dual energy scanning (i.e. dual source vs. dual layer). Therefore, the present study aimed to determine which technology is more accurate in quantifying iodine density. Methods: Twenty vials with known concentrations of iodine solutions were made using Optiray 350 contrast media diluted in sterile water. The concentration of iodine utilised ranged from 0.1 mg/ml to 1.0mg/ml in 0.1mg/ml increments, 1.5 mg/ml to 4.5 mg/ml in 0.5mg/ml increments followed by further concentrations at 5.0 mg/ml, 7mg/ml, 10 mg/ml and 15mg/ml. The vials were scanned using Dual Energy scan mode on a Siemens Somatom Force at 80kV/Sn150kV and 100kV/Sn150kV kilovoltage pairing. The same vials were scanned using Spectral scan mode on a Philips iQon at 120kVp and 140kVp. The images were reconstructed at 5mm thickness and 5mm increment using Br40 kernel on the Siemens Force and B Filter on Philips iQon. Post-processing of the Dual Energy data was performed on vendor-specific Siemens Syngo VIA (VB40) and Philips Intellispace Portal (Ver. 12) for the Spectral data. For each vial and scan mode, the iodine concentration was measured by placing an ROI in the coronal plane. Intraclass correlation analysis was performed on both datasets. Results: The iodine concentrations were reproduced with a high degree of accuracy for Dual Layer CT scanner. Although the Dual Source images showed a greater degree of deviation in measured iodine density for all vials, the dataset acquired at 80kV/Sn150kV had a higher accuracy. Conclusion: Spectral CT scanning by the dual layer technique has higher accuracy for quantitative measurements of iodine density compared to the dual source technique.

Keywords: CT, iodine density, spectral, dual-energy

Procedia PDF Downloads 120

Authors: Jitendar Kumar Tiwari, Ajay Mandal, N. Sathish, A. K. Srivastava

Abstract:

Since the last decade, graphene achieved great attention toward the progress of multifunction metal matrix composites, which are highly demanded in industries to develop energy-efficient systems. This study covers the two advanced aspects of the latest scientific endeavor, i.e., graphene as reinforcement in metallic materials and additive manufacturing (AM) as a processing technology. Herein, high-quality graphene and AlSi10Mg powder mechanically mixed by very low energy ball milling with 0.1 wt. % and 0.2 wt. % graphene. Mixed powder directly subjected to the powder bed fusion process, i.e., an AM technique to produce composite samples along with bare counterpart. The effects of graphene on porosity, microstructure, and mechanical properties were examined in this study. The volumetric distribution of pores was observed under X-ray computed tomography (CT). On the basis of relative density measurement by X-ray CT, it was observed that porosity increases after graphene addition, and pore morphology also transformed from spherical pores to enlarged flaky pores due to improper melting of composite powder. Furthermore, the microstructure suggests the grain refinement after graphene addition. The columnar grains were able to cross the melt pool boundaries in case of the bare sample, unlike composite samples. The smaller columnar grains were formed in composites due to heterogeneous nucleation by graphene platelets during solidification. The tensile properties get affected due to induced porosity irrespective of graphene reinforcement. The optimized tensile properties were achieved at 0.1 wt. % graphene. The increment in yield strength and ultimate tensile strength was 22% and 10%, respectively, for 0.1 wt. % graphene reinforced sample in comparison to bare counterpart while elongation decreases 20% for the same sample. The hardness indentations were taken mostly on the solid region in order to avoid the collapse of the pores. The hardness of the composite was increased progressively with graphene content. Around 30% of increment in hardness was achieved after the addition of 0.2 wt. % graphene. Therefore, it can be concluded that powder bed fusion can be adopted as a suitable technique to develop graphene reinforced AlSi10Mg composite. Though, some further process modification required to avoid the induced porosity after the addition of graphene, which can be addressed in future work.

Keywords: graphene, hardness, porosity, powder bed fusion, tensile properties

Procedia PDF Downloads 128

Authors: Joseph Fitoussi, Mohammadali Shirinbayan, Abbas Tcharkhtchi

Abstract:

Due to material flow during processing, short fiber reinforced composites structures obtained by injection or compression molding generally present strong spatial microstructure variation. On the other hand, quasi-static, dynamic, and fatigue behavior of these materials are highly dependent on microstructure parameters such as fiber orientation distribution. Indeed, because of complex damage mechanisms, SFRC structures design is a key challenge for safety and reliability. In this paper, we propose a micromechanical model allowing prediction of damage behavior of real structures as a function of microstructure spatial distribution. To this aim, a statistical damage criterion including strain rate and fatigue effect at the local scale is introduced into a Mori and Tanaka model. A critical local damage state is identified, allowing fatigue life prediction. Moreover, the multi-scale model is coupled with an experimental intrinsic link between damage under monotonic loading and fatigue life in order to build an abacus giving Tsai-Wu failure criterion parameters as a function of microstructure and targeted fatigue life. On the other hand, the micromechanical damage model gives access to the evolution of the anisotropic stiffness tensor of SFRC submitted to complex thermomechanical loading, including quasi-static, dynamic, and cyclic loading with temperature and amplitude variations. Then, the latter is used to fill out microstructure dependent material cards in finite element analysis for design optimization in the case of complex loading history. The proposed methodology is illustrated in the case of a real automotive component made of sheet molding compound (PSA 3008 tailgate). The obtained results emphasize how the proposed micromechanical methodology opens a new path for the automotive industry to lighten vehicle bodies and thereby save energy and reduce gas emission.

Keywords: short fiber reinforced composite, structural design, damage, micromechanical modelling, fatigue, strain rate effect

Procedia PDF Downloads 107

Authors: Kichan Lee, Kwang-Ho Choi, Mi-Hye Hwang, Young Min Son, Bang-Hun Hyun, Byeong Yeal Jung

Abstract:

Recently, worldwide food safety authorities have been strengthening food hygiene in order to curb foodborne illness outbreaks. The hygiene status of Korean slaughterhouses has been monitored annually by Animal and Plant Quarantine Agency and provincial governments through foodborne pathogens investigation using slaughtered pig and cattle meats. This study presented the prevalence of food-borne pathogens from 2014 to 2016 in Korean slaughterhouses. Sampling, microbiological examinations, and analysis of results were performed in accordance with ‘Processing Standards and Ingredient Specifications for Livestock Products’. In total, swab samples from 337 pig carcasses (100 samples in 2014, 135 samples in 2015, 102 samples in 2016) and 319 cattle carcasses (100 samples in 2014, 119 samples in 2015, 100 samples in 2016) from twenty slaughterhouses were examined for Listeria monocytogenes, Campylobacter jejuni, Campylobacter coli, Salmonella spp., Staphylococcus aureus, Clostridium perfringens, Yersinia enterocolitica, Escherichia coli O157:H7 and non-O157 enterohemorrhagic E. coli (EHEC, serotypes O26, O45, O103, O104, O111, O121, O128 and O145) as foodborne pathogens. The samples were analyzed using cultural and PCR-based methods. Foodborne pathogens were isolated in 78 (23.1%) out of 337 pig samples. In 2014, S. aureus (n=17) was predominant, followed by Y. enterocolitica (n=7), C. perfringens (n=2) and L. monocytogenes (n=2). In 2015, C. coli (n=14) was the most prevalent, followed by L. monocytogenes (n=4), S. aureus (n=3), and C. perfringens (n=2). In 2016, S. aureus (n=16) was the most prevalent, followed by C. coli (n=13), L. monocytogenes (n=2) and C. perfringens (n=1). In case of cattle carcasses, foodborne bacteria were detected in 41 (12.9%) out of 319 samples. In 2014, S. aureus (n=16) was the most prevalent, followed by Y. enterocolitica (n=3), C. perfringens (n=3) and L. monocytogenes (n=2). In 2015, L. monocytogenes was isolated from 4 samples, S. aureus from three, C. perfringens, Y. enterocolitica and Salmonella spp. from one, respectively. In 2016, L. monocytogenes (n=6) was the most prevalent, followed by C. perfringens (n=3) C. jejuni (n=1), respectively. It was found that 10 carcass samples (4 cattle and 6 pigs) were contaminated with two bacterial pathogen tested. Interestingly, foodborne pathogens were more detected from pig carcasses than cattle carcasses. Although S. aureus was predominantly detected in this study, other foodborne pathogens were also isolated in slaughtered meats. Results of this study alerted the risk of foodborne pathogen infection for humans from slaughtered meats. Therefore, the authors insisted that it was important to enhance hygiene level of slaughterhouses according to Hazard Analysis and Critical Control Point.

Keywords: carcass, cattle, foodborne, Korea, pathogen, pig

Procedia PDF Downloads 344

Authors: Clarence Dissanayake, Anishka A. Hettiarachchi

Abstract:

Landscape architecture needs to encompass a placemaking process carefully composing and manipulating landscape elements to address perceptual needs of humans, especially aesthetic, psychological and spiritual. The objective of this qualitative investigation is to inquire the impact of elements of rock and water combination on landscape perception and related feelings, emotions, and behavior. The past empirical studies have assessed the impact of landscape elements in isolation on user preference, yet the combined effect of elements have been less considered. This research was conducted with reference to the verity of qualities of water and rock through a visual landscape quality assessment focusing on landscape qualities derived from five visual concepts (coherence, historicity imageability, naturalness, and ephemera). 'Kaludiya Pokuna' archeological site in Anuradhapura was investigated with a sample of University students (n=19, male 14, female 5, age 20-25) using a five-point Likert scale via a perception based questionnaire and a visitor employed photographic survey (VEP). Two hypothetical questions were taken into investigation concerning biophilic (naturalness) and topophilic (historicity) aspects of humans to prefer a landscape with rock and water. The findings revealed that this combination encourages both biophilic and topophilic aspects, but in varying degrees. The identified hierarchy of visual concepts based on visitor’s preference signify coherence (93%), historicity (89%), imageability (79%), naturalness (75%) and ephemera (70%) respectively. It was further revealed that this combination creates a scenery more coherent dominating information processing aspect of humans to perceive a landscape over the biophilic and topophilic aspects. Different characteristics and secondary landscape effects generated by rock and water combination were found to affect in transforming a space into a place, full filling the aesthetic and spiritual aspects of the visitors. These findings enhance a means of making places for people, resource management and historical landscape conservation. Equalization of gender based participation, taking diverse cases and increasing the sample size with more analytical photographic analysis are recommended to enhance the quality of further research.

Keywords: landscape perception, visitor’s preference, rock and water combination, visual concepts

Procedia PDF Downloads 226

Authors: Kirill Filianin, Satu-Pia Reinikainen, Tuomo Sainio

Abstract:

Frequent measurements of product steam quality create a data overload that becomes more and more difficult to handle. In the current study, plant history data with multiple variables was successfully treated by principal component analysis to detect abnormal process behavior, particularly, in copper solvent extraction. The multivariate model is based on the concentration levels of main process metals recorded by the industrial on-stream x-ray fluorescence analyzer. After mean-centering and normalization of concentration data set, two-dimensional multivariate model under principal component analysis algorithm was constructed. Normal operating conditions were defined through control limits that were assigned to squared score values on x-axis and to residual values on y-axis. 80 percent of the data set were taken as the training set and the multivariate model was tested with the remaining 20 percent of data. Model testing showed successful application of control limits to detect abnormal behavior of copper solvent extraction process as early warnings. Compared to the conventional techniques of analyzing one variable at a time, the proposed model allows to detect on-line a process failure using information from all process variables simultaneously. Complex industrial equipment combined with advanced mathematical tools may be used for on-line monitoring both of process streams’ composition and final product quality. Defining normal operating conditions of the process supports reliable decision making in a process control room. Thus, industrial x-ray fluorescence analyzers equipped with integrated data processing toolbox allows more flexibility in copper plant operation. The additional multivariate process control and monitoring procedures are recommended to apply separately for the major components and for the impurities. Principal component analysis may be utilized not only in control of major elements’ content in process streams, but also for continuous monitoring of plant feed. The proposed approach has a potential in on-line instrumentation providing fast, robust and cheap application with automation abilities.

Keywords: abnormal process behavior, failure detection, principal component analysis, solvent extraction

Procedia PDF Downloads 310

Authors: O. Ekrami, P. Claes, S. Van Dongen

Abstract:

Fluctuating asymmetry (FA) has been studied for many years as an indicator of developmental stability or ‘genetic quality’ based on the assumption that perfect symmetry is ideally the expected outcome for a bilateral organism. Further studies have also investigated the possible link between FA and attractiveness or levels of masculinity or femininity. These hypotheses have been mostly examined using 2D images, and the structure of interest is usually presented using a limited number of landmarks. Such methods have the downside of simplifying and reducing the dimensionality of the structure, which will in return increase the error of the analysis. In an attempt to reach more conclusive and accurate results, in this study we have used high-resolution 3D scans of human faces and have developed an algorithm to measure and localize FA, taking a spatially-dense approach. A symmetric spatially dense anthropometric mask with paired vertices is non-rigidly mapped on target faces using an Iterative Closest Point (ICP) registration algorithm. A set of 19 manually indicated landmarks were used to examine the precision of our mapping step. The protocol’s accuracy in measurement and localizing FA is assessed using simulated faces with known amounts of asymmetry added to them. The results of validation of our approach show that the algorithm is perfectly capable of locating and measuring FA in 3D simulated faces. With the use of such algorithm, the additional captured information on asymmetry can be used to improve the studies of FA as an indicator of fitness or attractiveness. This algorithm can especially be of great benefit in studies of high number of subjects due to its automated and time-efficient nature. Additionally, taking a spatially dense approach provides us with information about the locality of FA, which is impossible to obtain using conventional methods. It also enables us to analyze the asymmetry of a morphological structures in a multivariate manner; This can be achieved by using methods such as Principal Components Analysis (PCA) or Factor Analysis, which can be a step towards understanding the underlying processes of asymmetry. This method can also be used in combination with genome wide association studies to help unravel the genetic bases of FA. To conclude, we introduced an algorithm to study and analyze asymmetry in human faces, with the possibility of extending the application to other morphological structures, in an automated, accurate and multi-variate framework.

Keywords: developmental stability, fluctuating asymmetry, morphometrics, 3D image processing

Procedia PDF Downloads 141

Authors: Disha Bhanot, Vinish Kathuria

Abstract:

This study focusses on the issue of distress sale in horticulture sector in India, which faces unique challenges, given the perishable nature of horticulture crops, seasonal production and paucity of post-harvest produce management links. Distress sale, from a farmer’s perspective may be defined as urgent sale of normal or distressed goods, at deeply discounted prices (way below the cost of production) and it is usually characterized by unfavorable conditions for the seller (farmer). The small and marginal farmers, often involved in subsistence farming, stand to lose substantially if they receive lower prices than expected prices (typically framed in relation to cost of production). Distress sale maximizes price uncertainty of produce leading to substantial income loss; and with increase in input costs of farming, the high variability in harvest price severely affects profit margin of farmers, thereby affecting their survival. The objective of this study is to model the occurrence of distress sale by tomato cultivators in the Indian state of Maharashtra, against the background of differential access to set of factors such as - capital, irrigation facilities, warehousing, storage and processing facilities, and institutional arrangements for procurement etc. Data is being collected using primary survey of over 200 farmers in key tomato growing areas of Maharashtra, asking information on the above factors in addition to seeking information on cost of cultivation, selling price, time gap between harvesting and selling, role of middleman in selling, besides other socio-economic variables. Farmers selling their produce far below the cost of production would indicate an occurrence of distress sale. Occurrence of distress sale would then be modelled as a function of farm, household and institutional characteristics. Heckman-two-stage model would be applied to find the probability/likelihood of a famer falling into distress sale as well as to ascertain how the extent of distress sale varies in presence/absence of various factors. Findings of the study would recommend suitable interventions and promotion of strategies that would help farmers better manage price uncertainties, avoid distress sale and increase profit margins, having direct implications on poverty.

Keywords: distress sale, horticulture, income loss, India, price uncertainity

Procedia PDF Downloads 243

Authors: Svetlana Kostrubina, Anastasia Prokopeva

Abstract:

This research presents a comprehensive study of the word formation processes in computer terminology within English and Russian languages and provides listeners with a system of exercises for training these skills. The originality is that this study focuses on a comparative approach, which shows both general patterns and specific features of English and Russian computer terms word formation. The key point is the system of exercises development for training computer terminology based on Bloom’s taxonomy. Data contain 486 units (228 English terms from the Glossary of Computer Terms and 258 Russian terms from the Terminological Dictionary-Reference Book). The objective is to identify the main affixation models in the English and Russian computer terms formation and to develop exercises. To achieve this goal, the authors employed Bloom’s Taxonomy as a methodological framework to create a systematic exercise program aimed at enhancing students’ cognitive skills in analyzing, applying, and evaluating computer terms. The exercises are appropriate for various levels of learning, from basic recall of definitions to higher-order thinking skills, such as synthesizing new terms and critically assessing their usage in different contexts. Methodology also includes: a method of scientific and theoretical analysis for systematization of linguistic concepts and clarification of the conceptual and terminological apparatus; a method of nominative and derivative analysis for identifying word-formation types; a method of word-formation analysis for organizing linguistic units; a classification method for determining structural types of abbreviations applicable to the field of computer communication; a quantitative analysis technique for determining the productivity of methods for forming abbreviations of computer vocabulary based on the English and Russian computer terms, as well as a technique of tabular data processing for a visual presentation of the results obtained. a technique of interlingua comparison for identifying common and different features of abbreviations of computer terms in the Russian and English languages. The research shows that affixation retains its productivity in the English and Russian computer terms formation. Bloom’s taxonomy allows us to plan a training program and predict the effectiveness of the compiled program based on the assessment of the teaching methods used.

Keywords: word formation, affixation, computer terms, Bloom's taxonomy

Procedia PDF Downloads 14

Authors: Moayyad Shawaqfah, Ibtehal Alqdah, Amjad Adaileh

Abstract:

Coridoor area (400 km2) lies to the north – east of Amman (60 km). It lies between 285-305 E longitude and 165-185 N latitude (according to Palestine Grid). It been subjected to exploitation of groundwater from new eleven wells since the 1999 with a total discharge of 11 MCM in addition to the previous discharge rate from the well field 14.7 MCM. Consequently, the aquifer balance is disturbed and a major decline in water level. Therefore, suitable groundwater resources management is required to overcome the problems of over pumping and its effect on groundwater quality. Three–dimensional groundwater flow model Processing Modeflow for Windows Pro (PMWIN PRO, 2003) has been used in order to calculate the groundwater budget, aquifer characteristics, and to predict the aquifer response under different stresses for the next 20 years (2035). The model was calibrated for steady state conditions by trial and error calibration. The calibration was performed by matching observed and calculated initial heads for year 2001. Drawdown data for period 2001-2010 were used to calibrate transient model by matching calculated with observed one, after that, the transient model was validated by using the drawdown data for the period 2011-2014. The hydraulic conductivities of the Basalt- A7/B2 aquifer System are ranging between 1.0 and 8.0 m/day. The low conductivity value was found at the north-west and south-western parts of the study area, the high conductivity value was found at north-western corner of the study area and the average storage coefficient is about 0.025. The water balance for the Basalt and B2/A7 formation at steady state condition with a discrepancy of 0.003%. The major inflows come from Jebal Al Arab through the basalt and through the limestone aquifer (B2/A7 12.28 MCMY aquifer and from excess rainfall is about 0.68 MCM/a. While the major outflows from the Basalt-B2/A7 aquifer system are toward Azraq basin with about 5.03 MCMY and leakage to A1/6 aquitard with 7.89 MCMY. Four scenarios have been performed to predict aquifer system responses under different conditions. Scenario no.2 was found to be the best one which indicates that the reduction the abstraction rates by 50% of current withdrawal rate (25.08 MCMY) to 12.54 MCMY. The maximum drawdowns were decreased to reach about, 7.67 and 8.38m in the years 2025 and 2035 respectively.

Keywords: Amman/Zarqa Basin, Jordan, groundwater management, groundwater modeling, modflow

Procedia PDF Downloads 216

Authors: Pedro Augusto Da Silva E Souza Miranda, Niloofar Jalali, Shweta Mistry

Abstract:

BACKGROUND: Software developers utilize automated solutions to scrape users’ reviews to extract meaningful knowledge to identify problems (e.g., bugs, compatibility issues) and possible enhancements (e.g., users’ requests) to their solutions. However, most of these solutions do not consider the health risk aspects to users. Recent works have shed light on the importance of including health risk considerations in the development cycle of mHealth apps to prevent harm to its users. PROBLEM: The COVID-19 Pandemic in Canada (and World) is currently forcing physical distancing upon the general population. This new lifestyle made the usage of mHealth applications more essential than ever, with a projected market forecast of 332 billion dollars by 2025. However, this new insurgency in mHealth usage comes with possible risks to users’ health due to mHealth apps problems (e.g., wrong insulin dosage indication due to a UI error). OBJECTIVE: These works aim to raise awareness amongst mHealth developers of the importance of considering risks to users’ health within their development lifecycle. Moreover, this work also aims to help mHealth developers with a Proof-of-Concept (POC) solution to understand, process, and identify possible health risks to users of mHealth apps based on users’ reviews. METHODS: We conducted a mixed-method study design. We developed a crawler to mine the negative reviews from two samples of mHealth apps (my fitness, medisafe) from the Google Play store users. For each mHealth app, we performed the following steps: • The reviews are divided into two groups, before starting the COVID-19 (reviews’ submission date before 15 Feb 2019) and during the COVID-19 (reviews’ submission date starts from 16 Feb 2019 till Dec 2020). For each period, the Latent Dirichlet Allocation (LDA) topic model was used to identify the different clusters of reviews based on similar topics of review The topics before and during COVID-19 are compared, and the significant difference in frequency and severity of similar topics are identified. RESULTS: We successfully scraped, filtered, processed, and identified health-related topics in both qualitative and quantitative approaches. The results demonstrated the similarity between topics before and during the COVID-19.

Keywords: natural language processing (NLP), topic modeling, mHealth, COVID-19, software engineering, telemedicine, health risks

Procedia PDF Downloads 130

Authors: Namasivayam Navaranjan, Eric Dimla

Abstract:

The world needs more quality food, shelter and transportation to meet the demands of growing population and improving living standard of those who currently live below the poverty line. Materials are essential commodities for various applications including food and pharmaceutical packaging, building and automobile. Petroleum based plastics are widely used materials amongst others for these applications and their demand is expected to increase. Use of plastics has environment related issues because considerable amount of plastic used worldwide is disposed in landfills, where its resources are wasted, the material takes up valuable space and blights communities. Some countries have been implementing regulations and/or legislations to increase reuse, recycle, renew and remanufacture materials as well as to minimise the use of non-environmentally friendly materials such as petroleum plastics. However, issue of material waste is still a concern in the countries who have low environmental regulations. Development of materials, mostly bioplastics from renewable biomass resources has become popular in the last decade. It is widely believed that the potential for up to 90% substitution of total plastics consumption by bioplastics is technically possible. The global demand for bioplastics is estimated to be approximately six times larger than in 2010. Recently, standard polymers like polyethylene (PE), polypropylene (PP), Polyvinyl Chloride (PVC) or Polyethylene terephthalate (PET), but also high-performance polymers such as polyamides or polyesters have been totally or partially substituted by their renewable equivalents. An example is Polylactide (PLA) being used as a substitute in films and injection moulded products made of petroleum plastics, e.g. PET. The starting raw materials for bio-based materials are usually sugars or starches that are mostly derived from food resources, partially also recycled materials from food or wood processing. The risk in lower food availability by increasing price of basic grains as a result of competition with biomass-based product sectors for feedstock also needs to be considered for the future bioplastic production. Manufacturing of bioplastic materials is often still reliant upon petroleum as an energy and materials source. Life Cycle Assessment (LCA) of bioplastic products has being conducted to determine the sustainability of a production route. However, the accuracy of LCA depends on several factors and needs improvement. Low oil price and high production cost may also limit the technically possible growth of these plastics in the coming years.

Keywords: bioplastics, plastics, renewable resources, biomass

Procedia PDF Downloads 308

Authors: Jasmal A. Syamsu, Muhammad Yusuf, Hikmah M. Ali, Mawardi A. Asja, Zulkharnaim

Abstract:

This study was conducted and aimed in identifying and mapping the pasture as an ecological base of beef cattle. A survey was carried out during a period of April to June 2016, in Suppa, Mattirobulu, the district of Pinrang, South Sulawesi province. The mapping process of grazing area was conducted in several stages; inputting and tracking of data points into Google Earth Pro (version 7.1.4.1529), affirmation and confirmation of tracking line visualized by satellite with a variety of records at the point, a certain point and tracking input data into ArcMap Application (ArcGIS version 10.1), data processing DEM/SRTM (S04E119) with respect to the location of the grazing areas, creation of a contour map (a distance of 5 m) and mapping tilt (slope) of land and land cover map-making. Analysis of land cover, particularly the state of the vegetation was done through the identification procedure NDVI (Normalized Differences Vegetation Index). This procedure was performed by making use of the Landsat-8. The results showed that the topography of the grazing areas of hills and some sloping surfaces and flat with elevation vary from 74 to 145 above sea level (asl), while the requirements for growing superior grass and legume is an altitude of up to 143-159 asl. Slope varied between 0 - > 40% and was dominated by a slope of 0-15%, according to the slope/topography pasture maximum of 15%. The range of NDVI values for pasture image analysis results was between 0.1 and 0.27. Characteristics of vegetation cover of pasture land in the category of vegetation density were low, 70% of the land was the land for cattle grazing, while the remaining approximately 30% was a grove and forest included plant water where the place for shelter of the cattle during the heat and drinking water supply. There are seven types of graminae and 5 types of legume that was dominant in the region. Proportionally, graminae class dominated up 75.6% and legume crops up to 22.1% and the remaining 2.3% was another plant trees that grow in the region. The dominant weed species in the region were Cromolaenaodorata and Lantana camara, besides that there were 6 types of floor plant that did not include as forage fodder.

Keywords: pastoral, ecology, mapping, beef cattle

Procedia PDF Downloads 354

Authors: Sanjay Rathee, Arti Kashyap

Abstract:

Extraction of useful information from large datasets is one of the most important research problems. Association rule mining is one of the best methods for this purpose. Finding possible associations between items in large transaction based datasets (finding frequent patterns) is most important part of the association rule mining. There exist many algorithms to find frequent patterns but Apriori algorithm always remains a preferred choice due to its ease of implementation and natural tendency to be parallelized. Many single-machine based Apriori variants exist but massive amount of data available these days is above capacity of a single machine. Therefore, to meet the demands of this ever-growing huge data, there is a need of multiple machines based Apriori algorithm. For these types of distributed applications, MapReduce is a popular fault-tolerant framework. Hadoop is one of the best open-source software frameworks with MapReduce approach for distributed storage and distributed processing of huge datasets using clusters built from commodity hardware. However, heavy disk I/O operation at each iteration of a highly iterative algorithm like Apriori makes Hadoop inefficient. A number of MapReduce-based platforms are being developed for parallel computing in recent years. Among them, two platforms, namely, Spark and Flink have attracted a lot of attention because of their inbuilt support to distributed computations. Earlier we proposed a reduced- Apriori algorithm on Spark platform which outperforms parallel Apriori, one because of use of Spark and secondly because of the improvement we proposed in standard Apriori. Therefore, this work is a natural sequel of our work and targets on implementing, testing and benchmarking Apriori and Reduced-Apriori and our new algorithm ReducedAll-Apriori on Apache Flink and compares it with Spark implementation. Flink, a streaming dataflow engine, overcomes disk I/O bottlenecks in MapReduce, providing an ideal platform for distributed Apriori. Flink's pipelining based structure allows starting a next iteration as soon as partial results of earlier iteration are available. Therefore, there is no need to wait for all reducers result to start a next iteration. We conduct in-depth experiments to gain insight into the effectiveness, efficiency and scalability of the Apriori and RA-Apriori algorithm on Flink.

Keywords: apriori, apache flink, Mapreduce, spark, Hadoop, R-Apriori, frequent itemset mining

Procedia PDF Downloads 294

Authors: Desmond F. Bellamy

Abstract:

Discourses about animal rights usually assume an ontological abyss between human and animal. This supposition of non-animality allows us to utilise and exploit non-humans, particularly those with commercial value, with little regard for their rights or interests. We can and do confine them, inflict painful treatments such as castration and branding, and slaughter them at an age determined only by financial considerations. This paper explores the way images and texts depicting human cannibalism reflect this deprivation of rights back onto our species and examines how this offers new perspectives on our granting or withholding of rights to farmed animals. The animals we eat – sheep, pigs, cows, chickens and a small handful of other species – are during processing de-animalised, turned into commodities, and made unrecognisable as formerly living beings. To do the same to a human requires the cannibal to enact another step – humans must first be considered as animals before they can be commodified or de-animalised. Different iterations of cannibalism in a selection of fiction and non-fiction texts will be considered: survivalism (necessitated by catastrophe or dystopian social collapse), the primitive savage of colonial discourses, and the inhuman psychopath. Each type of cannibalism shows alternative ways humans can be animalised and thereby dispossessed of both their human and animal rights. Human rights, summarised in the UN Universal Declaration of Human Rights as ‘life, liberty, and security of person’ are stubbornly denied to many humans, and are refused to virtually all farmed non-humans. How might this paradigm be transformed by seeing the animal victim replaced by an animalised human? People are fascinated as well as repulsed by cannibalism, as demonstrated by the upsurge of films on the subject in the last few decades. Cannibalism is, at its most basic, about envisaging and treating humans as objects: meat. It is on the dinner plate that the abyss between human and ‘animal’ is most challenged. We grasp at a conscious level that we are a species of animal and may become, if in the wrong place (e.g., shark-infested water), ‘just food’. Culturally, however, strong traditions insist that humans are much more than ‘just meat’ and deserve a better fate than torment and death. The billions of animals on death row awaiting human consumption would ask the same if they could. Depictions of cannibalism demonstrate in graphic ways that humans are animals, made of meat and that we can also be butchered and eaten. These depictions of us as having the same fleshiness as non-human animals reminds us that they have the same capacities for pain and pleasure as we do. Depictions of cannibalism, therefore, unconsciously aid in deconstructing the human/animal binary and give a unique glimpse into the often unnoticed repudiation of animal rights.

Keywords: animal rights, cannibalism, human/animal binary, objectification

Procedia PDF Downloads 138

Authors: Jayati Bhattacharjee

Abstract:

Teaching is a very dynamic profession. Teaching Science is as much challenging as Learning the subject if not more. For instance teaching of Chemistry. From the introductory concepts of subatomic particles to atoms of elements and their symbols and further presenting the chemical equation and so forth is a challenge on both side of the equation Teaching Learning. This paper combines the Neuroscience of Learning and memory with the knowledge of Learning style (VAK) and presents an effective tool for the teacher to authenticate Learning. The model of ‘Working Memory’, the Visio-spatial sketchpad, the central executive and the phonological loop that transforms short-term memory to long term memory actually supports the psychological theory of Learning style i.e. Visual –Auditory-Kinesthetic. A closer examination of David Kolbe’s learning model suggests that learning requires abilities that are polar opposites, and that the learner must continually choose which set of learning abilities he or she will use in a specific learning situation. In grasping experience some of us perceive new information through experiencing the concrete, tangible, felt qualities of the world, relying on our senses and immersing ourselves in concrete reality. Others tend to perceive, grasp, or take hold of new information through symbolic representation or abstract conceptualization – thinking about, analyzing, or systematically planning, rather than using sensation as a guide. Similarly, in transforming or processing experience some of us tend to carefully watch others who are involved in the experience and reflect on what happens, while others choose to jump right in and start doing things. The watchers favor reflective observation, while the doers favor active experimentation. Any lesson plan based on the model of Prescriptive design: C+O=M (C: Instructional condition; O: Instructional Outcome; M: Instructional method). The desired outcome and conditions are independent variables whereas the instructional method is dependent hence can be planned and suited to maximize the learning outcome. The assessment for learning rather than of learning can encourage, build confidence and hope amongst the learners and go a long way to replace the anxiety and hopelessness that a student experiences while learning Science with a human touch in it. Application of this model has been tried in teaching chemistry to high school students as well as in workshops with teachers. The response received has proven the desirable results.

Keywords: working memory model, learning style, prescriptive design, assessment for learning

Procedia PDF Downloads 351

Authors: Birgul Hizlar, Sibel Karakaya

Abstract:

Recently, increasing attention has been given to carotenoid bioaccessibility and bioavailability in the field of nutrition research. As a consequence of their lipophilic nature and their specific localization in plant-based tissues, carotenoid bioaccessibility and bioavailability is generally quite low in raw fruits and vegetables, since carotenoids need to be released from the cellular matrix and incorporated in the lipid fraction during digestion before being absorbed. Today’s approach related to improving the bioaccessibility is to design food matrix. Recently, the newest approach, excipient food, has been introduced to improve the bioavailability of orally administered bioactive compounds. The main idea is combining food and another food (the excipient food) whose composition and/or structure is specifically designed for improving health benefits. In this study, effects of food processing, food matrix and the addition of excipient foods on the carotenoid bioaccessibility of carrots were determined. Different excipient foods (olive oil, lemon juice and whey curd) and different food matrices (grating, boiling and mashing) were used. Total carotenoid contents of the grated, boiled and mashed carrots were 57.23, 51.11 and 62.10 μg/g respectively. No significant differences among these values indicated that these treatments had no effect on the release of carotenoids from the food matrix. Contrary to, changes in the food matrix, especially mashing caused significant increase in the carotenoid bioaccessibility. Although the carotenoid bioaccessibility was 10.76% in grated carrots, this value was 18.19% in mashed carrots (p<0.05). Addition of olive oil and lemon juice as excipients into the grated carrots caused 1.23 times and 1.67 times increase in the carotenoid content and the carotenoid bioaccessibility respectively. However, addition of the excipient foods in the boiled carrot samples did not influence the release of carotenoid from the food matrix. Whereas, up to 1.9 fold increase in the carotenoid bioaccessibility was determined by the addition of the excipient foods into the boiled carrots. The bioaccessibility increased from 14.20% to 27.12% by the addition of olive oil, lemon juice and whey curd. The highest carotenoid content among mashed carrots was found in the mashed carrots incorporated with olive oil and lemon juice. This combination also caused a significant increase in the carotenoid bioaccessibility from 18.19% to 29.94% (p<0.05). When compared the results related with the effect of the treatments on the carotenoid bioaccessibility, mashed carrots containing olive oil, lemon juice and whey curd had the highest carotenoid bioaccessibility. The increase in the bioaccessibility was approximately 81% when compared to grated and mashed samples containing olive oil, lemon juice and whey curd. In conclusion, these results demonstrated that the food matrix and addition of the excipient foods had a significant effect on the carotenoid content and the carotenoid bioaccessibility.

Keywords: carrot, carotenoids, excipient foods, food matrix

Procedia PDF Downloads 457

Authors: Buruk Kitachew Wossenyeleh

Abstract:

Groundwater is the largest freshwater reservoir in the world. Like the other reservoirs of the hydrologic cycle, it is a finite resource. This study focused on the groundwater modeling of the Ter Kamerenbos well field to understand the groundwater flow system and the impact of different scenarios. The study area covers 68.9Km2 in the Brussels Capital Region and is situated in two river catchments, i.e., Zenne River and Woluwe Stream. The aquifer system has three layers, but in the modeling, they are considered as one layer due to their hydrogeological properties. The catchment aquifer system is replenished by direct recharge from rainfall. The groundwater recharge of the catchment is determined using the spatially distributed water balance model called WetSpass, and it varies annually from zero to 340mm. This groundwater recharge is used as the top boundary condition for the groundwater modeling of the study area. During the groundwater modeling using Processing MODFLOW, constant head boundary conditions are used in the north and south boundaries of the study area. For the east and west boundaries of the study area, head-dependent flow boundary conditions are used. The groundwater model is calibrated manually and automatically using observed hydraulic heads in 12 observation wells. The model performance evaluation showed that the root means the square error is 1.89m and that the NSE is 0.98. The head contour map of the simulated hydraulic heads indicates the flow direction in the catchment, mainly from the Woluwe to Zenne catchment. The simulated head in the study area varies from 13m to 78m. The higher hydraulic heads are found in the southwest of the study area, which has the forest as a land-use type. This calibrated model was run for the climate change scenario and well operation scenario. Climate change may cause the groundwater recharge to increase by 43% and decrease by 30% in 2100 from current conditions for the high and low climate change scenario, respectively. The groundwater head varies for a high climate change scenario from 13m to 82m, whereas for a low climate change scenario, it varies from 13m to 76m. If doubling of the pumping discharge assumed, the groundwater head varies from 13m to 76.5m. However, if the shutdown of the pumps is assumed, the head varies in the range of 13m to 79m. It is concluded that the groundwater model is done in a satisfactory way with some limitations, and the model output can be used to understand the aquifer system under steady-state conditions. Finally, some recommendations are made for the future use and improvement of the model.

Keywords: Ter Kamerenbos, groundwater modelling, WetSpass, climate change, well operation

Procedia PDF Downloads 152

Authors: Fang Liu, JiaJia Yao, GuanLin Wu, ZuMaoLi, XueYan Yang, HePeng Zhang, ZhiPeng Sun, JunShuai Xue

Abstract:

The explosive increasing needs of data processing and information storage strongly drive the advancement of the binary logic system to multiple-valued logic system. Inherent negative differential resistance characteristic, ultra-high-speed switching time, and robust anti-irradiation capability make III-nitride resonant tunneling diode one of the most promising candidates for multi-valued logic devices. Here we report the monolithic integration of GaN resonant tunneling diodes in series to realize multiple negative differential resistance regions, obtaining at least three stable operating states. A multiply-by-three circuit is achieved by this combination, increasing the frequency of the input triangular wave from f0 to 3f0. The resonant tunneling diodes are grown by plasma-assistedmolecular beam epitaxy on free-standing c-plane GaN substrates, comprising double barriers and a single quantum well both at the atomic level. Device with a peak current density of 183kA/cm² in conjunction with a peak-to-valley current ratio (PVCR) of 2.07 is observed, which is the best result reported in nitride-based resonant tunneling diodes. Microwave oscillation event at room temperature was discovered with a fundamental frequency of 0.31GHz and an output power of 5.37μW, verifying the high repeatability and robustness of our device. The switching behavior measurement was successfully carried out, featuring rise and fall times in the order of picoseconds, which can be used in high-speed digital circuits. Limited by the measuring equipment and the layer structure, the switching time can be further improved. In general, this article presents a novel nitride device with multiple negative differential regions driven by the resonant tunneling mechanism, which can be used in high-speed multiple value logic field with reduced circuit complexity, demonstrating a new solution of nitride devices to break through the limitations of binary logic.

Keywords: GaN resonant tunneling diode, negative differential resistance, multiple-valued logic system, switching time, peak-to-valley current ratio

Procedia PDF Downloads 100

Authors: Yang Cao

Abstract:

Analyzing the oral production of Chinese-speaking learners of English as a second language (L2), we can find a large variety of verb inflections – Why does it seem so hard for them to use consistent correct past morphologies in obligatory past contexts? Failed Functional Features Hypothesis (FFFH) attributes the rather non-target-like performance to the absence of [±past] feature in their L1 Chinese, arguing that for post puberty learners, new features in L2 are no more accessible. By contrast, Missing Surface Inflection Hypothesis (MSIH) tends to believe that all features are actually acquirable for late L2 learners, while due to the mapping difficulties from features to forms, it is hard for them to realize the consistent past morphologies on the surface. However, most of the studies are limited to the verb morphologies in finite clauses and few studies have ever attempted to figure out these learners’ performance in non-finite clauses. Additionally, it has been discussed that Chinese learners may be able to tell the finite/infinite distinction (i.e. the [±finite] feature might be selected in Chinese, even though the existence of [±past] is denied). Therefore, adopting a self-paced reading task (SPR), the current study aims to analyze the processing patterns of Chinese-speaking learners of L2 Russian, in order to find out if they are sensitive to misuse of tense morphologies in both finite and non-finite clauses and whether they are sensitive to the finite/infinite distinction presented in Russian. The study targets L2 Russian due to its systematic morphologies in both present and past tenses. A native Russian group, as well as a group of English-speaking learners of Russian, whose L1 has definitely selected both [±finite] and [±past] features, will also be involved. By comparing and contrasting performance of the three language groups, the study is going to further examine and discuss the two theories, FFFH and MSIH. Preliminary hypotheses are: a) Russian native speakers are expected to spend longer time reading the verb forms which violate the grammar; b) it is expected that Chinese participants are, at least, sensitive to the misuse of inflected verbs in non-finite clauses, although no sensitivity to the misuse of infinitives in finite clauses might be found. Therefore, an interaction of finite and grammaticality is expected to be found, which indicate that these learners are able to tell the finite/infinite distinction; and c) having selected [±finite] and [±past], English-speaking learners of Russian are expected to behave target-likely, supporting L1 transfer.

Keywords: features, finite clauses, morphosyntax, non-finite clauses, past morphologies, present morphologies, Second Language Acquisition, self-paced reading task, verb inflections

Procedia PDF Downloads 108

Authors: Karan Sharma, Ajay Kumar

Abstract:

Epileptic seizure is a type of disease due to which electrical charge in the brain flows abruptly resulting in abnormal activity by the subject. One percent of total world population gets epileptic seizure attacks.Due to abrupt flow of charge, EEG (Electroencephalogram) waveforms change. On the display appear a lot of spikes and sharp waves in the EEG signals. Detection of epileptic seizure by using conventional methods is time-consuming. Many methods have been evolved that detect it automatically. The initial part of this paper provides the review of techniques used to detect epileptic seizure automatically. The automatic detection is based on the feature extraction and classification patterns. For better accuracy decomposition of the signal is required before feature extraction. A number of parameters are calculated by the researchers using different techniques e.g. approximate entropy, sample entropy, Fuzzy approximate entropy, intrinsic mode function, cross-correlation etc. to discriminate between a normal signal & an epileptic seizure signal.The main objective of this review paper is to present the variations in the EEG signals at both stages (i) Interictal (recording between the epileptic seizure attacks). (ii) Ictal (recording during the epileptic seizure), using most appropriate methods of analysis to provide better healthcare diagnosis. This research paper then investigates the effects of a noninvasive healing therapy on the subjects by studying the EEG signals using latest signal processing techniques. The study has been conducted with Reiki as a healing technique, beneficial for restoring balance in cases of body mind alterations associated with an epileptic seizure. Reiki is practiced around the world and is recommended for different health services as a treatment approach. Reiki is an energy medicine, specifically a biofield therapy developed in Japan in the early 20th century. It is a system involving the laying on of hands, to stimulate the body’s natural energetic system. Earlier studies have shown an apparent connection between Reiki and the autonomous nervous system. The Reiki sessions are applied by an experienced therapist. EEG signals are measured at baseline, during session and post intervention to bring about effective epileptic seizure control or its elimination altogether.

Keywords: EEG signal, Reiki, time consuming, epileptic seizure

Procedia PDF Downloads 406

Authors: Jonathan Gong

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deep learning, image processing, machine learning

Procedia PDF Downloads 130

Authors: Sudha T., Naveen C.

Abstract:

Dengue is a serious public health issue that causes significant annual economic and welfare burdens on nations. However, enhanced optimization techniques and quantitative modeling approaches can predict the incidence of dengue. By advocating for a data-driven approach, public health officials can make informed decisions, thereby improving the overall effectiveness of sudden disease outbreak control efforts. The National Oceanic and Atmospheric Administration and the Centers for Disease Control and Prevention are two of the U.S. Federal Government agencies from which this study uses environmental data. Based on environmental data that describe changes in temperature, precipitation, vegetation, and other factors known to affect dengue incidence, many predictive models are constructed that use different machine learning methods to estimate weekly dengue cases. The first step involves preparing the data, which includes handling outliers and missing values to make sure the data is prepared for subsequent processing and the creation of an accurate forecasting model. In the second phase, multiple feature selection procedures are applied using various machine learning models and optimization techniques. During the third phase of the research, machine learning models like the Huber Regressor, Support Vector Machine, Gradient Boosting Regressor (GBR), and Support Vector Regressor (SVR) are compared with several optimization techniques for feature selection, such as Harmony Search and Genetic Algorithm. In the fourth stage, the model's performance is evaluated using Mean Square Error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) as assistance. Selecting an optimization strategy with the least number of errors, lowest price, biggest productivity, or maximum potential results is the goal. In a variety of industries, including engineering, science, management, mathematics, finance, and medicine, optimization is widely employed. An effective optimization method based on harmony search and an integrated genetic algorithm is introduced for input feature selection, and it shows an important improvement in the model's predictive accuracy. The predictive models with Huber Regressor as the foundation perform the best for optimization and also prediction.

Keywords: deep learning model, dengue fever, prediction, optimization

Procedia PDF Downloads 65

Authors: Maxence Queyrel, Edi Prifti, Alexandre Templier, Jean-Daniel Zucker

Abstract:

Analysis of the human microbiome using metagenomic sequencing data has demonstrated high ability in discriminating various human diseases. Raw metagenomic sequencing data require multiple complex and computationally heavy bioinformatics steps prior to data analysis. Such data contain millions of short sequences read from the fragmented DNA sequences and stored as fastq files. Conventional processing pipelines consist in multiple steps including quality control, filtering, alignment of sequences against genomic catalogs (genes, species, taxonomic levels, functional pathways, etc.). These pipelines are complex to use, time consuming and rely on a large number of parameters that often provide variability and impact the estimation of the microbiome elements. Training Deep Neural Networks directly from raw sequencing data is a promising approach to bypass some of the challenges associated with mainstream bioinformatics pipelines. Most of these methods use the concept of word and sentence embeddings that create a meaningful and numerical representation of DNA sequences, while extracting features and reducing the dimensionality of the data. In this paper we present an end-to-end approach that classifies patients into disease groups directly from raw metagenomic reads: metagenome2vec. This approach is composed of four steps (i) generating a vocabulary of k-mers and learning their numerical embeddings; (ii) learning DNA sequence (read) embeddings; (iii) identifying the genome from which the sequence is most likely to come and (iv) training a multiple instance learning classifier which predicts the phenotype based on the vector representation of the raw data. An attention mechanism is applied in the network so that the model can be interpreted, assigning a weight to the influence of the prediction for each genome. Using two public real-life data-sets as well a simulated one, we demonstrated that this original approach reaches high performance, comparable with the state-of-the-art methods applied directly on processed data though mainstream bioinformatics workflows. These results are encouraging for this proof of concept work. We believe that with further dedication, the DNN models have the potential to surpass mainstream bioinformatics workflows in disease classification tasks.

Keywords: deep learning, disease prediction, end-to-end machine learning, metagenomics, multiple instance learning, precision medicine

Procedia PDF Downloads 125

Authors: Morteza Esfandiari, Shabnam Jabari, Heather MacGrath, David Coleman

Abstract:

Flood is a severe issue in different places in the world as well as the city of Fredericton, New Brunswick, Canada. The downtown area of Fredericton is close to the Saint John River, which is susceptible to flood around May every year. Recently, the frequency of flooding seems to be increased, especially after the fact that the downtown area and surrounding urban/agricultural lands got flooded in two consecutive years in 2018 and 2019. In order to have an explicit vision of flood span and damage to affected areas, it is necessary to use either flood inundation modelling or satellite data. Due to contingent availability and weather dependency of optical satellites, and limited existing data for the high cost of hydrodynamic models, it is not always feasible to rely on these sources of data to generate quality flood maps after or during the catastrophe. Height Above the Nearest Drainage (HAND), a state-of-the-art topo-hydrological index, normalizes the height of a basin based on the relative elevation along with the stream network and specifies the gravitational or the relative drainage potential of an area. HAND is a relative height difference between the stream network and each cell on a Digital Terrain Model (DTM). The stream layer is provided through a multi-step, time-consuming process which does not always result in an optimal representation of the river centerline depending on the topographic complexity of that region. HAND is used in numerous case studies with quite acceptable and sometimes unexpected results because of natural and human-made features on the surface of the earth. Some of these features might cause a disturbance in the generated model, and consequently, the model might not be able to predict the flow simulation accurately. We propose to include a previously existing stream layer generated by the province of New Brunswick and benefit from culvert maps to improve the water flow simulation and accordingly the accuracy of HAND model. By considering these parameters in our processing, we were able to increase the accuracy of the model from nearly 74% to almost 92%. The improved model can be used for generating highly accurate flood maps, which is necessary for future urban planning and flood damage estimation without any need for satellite imagery or hydrodynamic computations.

Keywords: HAND, DTM, rapid floodplain, simplified conceptual models

Procedia PDF Downloads 151

Authors: Majid Eslami

Abstract:

Functionally graded materials (FGMs) exhibit heterogeneous microstructures in which the composition and properties gently change in specified directions. The common type of FGMs consist of a metal in which ceramic particles are distributed with a graded concentration. There are many processing routes for FGMs. An important group of these methods is casting techniques (gravity or centrifugal). However, the main problem of casting molten metal slurry with dispersed ceramic particles is a destructive chemical reaction between these two phases which deteriorates the properties of the materials. In order to overcome this problem, in the present investigation a suspension of 6061 aluminum and alumina powders in a liquid polymer was used as the starting material and subjected to centrifugal force for making FGMs. The size rang of these powders was 45-63 and 106-125 μm. The volume percent of alumina in the Al/Al2O3 powder mixture was in the range of 5 to 20%. PMMA (Plexiglas) in different concentrations (20-50 g/lit) was dissolved in toluene and used as the suspension liquid. The glass mold contaning the suspension of Al/Al2O3 powders in the mentioned liquid was rotated at 1700 rpm for different times (4-40 min) while the arm length was kept constant (10 cm) for all the experiments. After curing the polymer, burning out the binder, cold pressing and sintering , cylindrical samples (φ=22 mm h=20 mm) were produced. The density of samples before and after sintering was quantified by Archimedes method. The results indicated that by using the same sized alumina and aluminum powders particles, FGM sample can be produced by rotation times exceeding 7 min. However, by using coarse alumina and fine alumina powders the sample exhibits step concentration. On the other hand, using fine alumina and coarse alumina results in a relatively uniform concentration of Al2O3 along the sample height. These results are attributed to the effects of size and density of different powders on the centrifugal force induced on the powders during rotation. The PMMA concentration and the vol.% of alumina in the suspension did not have any considerable effect on the distribution of alumina particles in the samples. The hardness profiles along the height of samples were affected by both the alumina vol.% and porosity content. The presence of alumina particles increased the hardness while increased porosity reduced the hardness. Therefore, the hardness values did not show the expected gradient in same sample. The sintering resulted in decreased porosity for all the samples investigated.

Keywords: FGM, powder metallurgy, centrifugal method, polymeric suspension

Procedia PDF Downloads 211

Authors: Soon Young Choi, Chan Hwan Kim, Chan Hong Park, Hyung Rae Kim, Myoung Hoon Lee, Hyeon-Yeong Park

Abstract:

We performed the marine magnetic, bathymetry and seismic survey at the TA22 seamount (in the Lau basin, SW Pacific) for finding the submarine hydrothermal deposits in October 2009. We acquired magnetic and bathymetry data sets by suing Overhouser Proton Magnetometer SeaSPY (Marine Magnetics Co.), Multi-beam Echo Sounder EM120 (Kongsberg Co.). We conducted the data processing to obtain detailed seabed topography, magnetic anomaly, reduction to the pole (RTP) and magnetization. Based on the magnetic properties result, we analyzed submarine geology structure of TA22 seamount with post-processed seismic profile. The detailed bathymetry of the TA22 seamount showed the left and right crest parts that have caldera features in each crest central part. The magnetic anomaly distribution of the TA22 seamount regionally displayed high magnetic anomalies in northern part and the low magnetic anomalies in southern part around the caldera features. The RTP magnetic anomaly distribution of the TA22 seamount presented commonly high magnetic anomalies in the each caldera central part. Also, it represented strong anomalies at the inside of caldera rather than outside flank of the caldera. The magnetization distribution of the TA22 seamount showed the low magnetization zone in the center of each caldera, high magnetization zone in the southern and northern east part. From analyzed the seismic profile map, The TA22 seamount area is showed for the inferred small mounds inside each caldera central part and it assumes to make possibility of sills by the magma in cases of the right caldera. Taking into account all results of this study (bathymetry, magnetic anomaly, RTP, magnetization, seismic profile) with rock samples at the left caldera area in 2009 survey, we suppose the possibility of hydrothermal deposits at mounds in each caldera central part and at outside flank of the caldera representing the low magnetization zone. We expect to have the better results by combined modeling from this study data with the other geological data (ex. detailed gravity, 3D seismic, petrologic study results and etc).

Keywords: detailed bathymetry, magnetic anomaly, seamounts, seismic profile, SW Pacific

Procedia PDF Downloads 403