Search results for: forecasting accuracy

Authors: Sara De Pelsmaeker, Catarina Ferreira da Silva, Janne Prawit

Abstract:

Recent studies show that enzymatically hydrolysed collagen is absorbed and distributed to joint tissues, where it has analgesic and active anti-inflammatory properties. Reviews of the associated relevant literature also support this theory. However, these studies are all using hydrolyzed collagen from animal hide or skin. This study looks into the effect of daily supplementation of hydrolyzed cartilage collagen (HCC), which has a different composition. A consumer study was set up using a double-blind placebo-controlled design with a control group using twice a day 0.5gr of maltodextrin and an experimental group using twice 0.5g of HCC, over a trial period of 12 weeks. A follow-up phase of 4 weeks without supplementation was taken into the experiment to investigate the ‘wash-out’ phase. As this consumer study was conducted during the lockdown periods, a specific app was designed to follow up with the participants. The app had the advantage that in this way, the motivation of the participants was enhanced and the drop-out range of participants was lower than normally seen in consumer studies. Participants were recruited via various sports and health clubs across the UK as we targeted a general population of people that considered themselves in good health. Exclusion criteria were ‘not experiencing any medical conditions’ and ‘not taking any prescribed medication’. A minimum requirement was that they regularly engaged in some level of physical activity. The participants had to log the type of activity that they conducted and the duration of the activity. Weekly, participants were providing feedback on their joint health and subjective pain using the validated pain measuring instrument Visual Analogue Scale (VAS). The weekly repoAbstract Public Health and Wellbeing Conferencerting section in the app was designed with simplicity and based on the accuracy demonstrated in previous similar studies to track subjective pain measures of participants. At the beginning of the trial, each participant indicated their baseline on joint pain. The results of this consumer study indicated that HCC significantly improved joint health and subjective pain scores compared to the placebo group. No significant differences were found between different demographic groups (age or gender). The level of activity, going from high intensive training to regular walking, did not significantly influence the effect of the HCC. The results of the wash-out phase indicated that when the participants stopped the HCC supplementation, their subjective pain scores increased again to the baseline. In conclusion, the results gave a positive indication that the daily supplementation of HCC can contribute to the overall mobility and wellbeing of a general active population

Keywords: VAS-score, food supplement, mobility, joint health

Procedia PDF Downloads 139

Authors: Laura Micheli, Majd Hijazi, Mahmoud Faytarouni

Abstract:

The damage reported by oil and gas industrial facilities revealed the utmost vulnerability of steel liquid storage tanks to seismic events. The failure of steel storage tanks may yield devastating and long-lasting consequences on built and natural environments, including the release of hazardous substances, uncontrolled fires, and soil contamination with hazardous materials. It is, therefore, fundamental to reliably predict the damage that steel liquid storage tanks will likely experience under future seismic hazard events. The seismic performance of steel liquid storage tanks is usually assessed using vulnerability curves obtained from the numerical simulation of a tank under different hazard scenarios. However, the computational demand of high-fidelity numerical simulation models, such as finite element models, makes the vulnerability assessment of liquid storage tanks time-consuming and often impractical. As a solution, this paper presents a surrogate model-based strategy for predicting seismic-induced damage in steel liquid storage tanks. In the proposed strategy, the surrogate model is leveraged to reduce the computational demand of time-consuming numerical simulations. To create the data set for training the surrogate model, field damage data from past earthquakes reconnaissance surveys and reports are collected. Features representative of steel liquid storage tank characteristics (e.g., diameter, height, liquid level, yielding stress) and seismic excitation parameters (e.g., peak ground acceleration, magnitude) are extracted from the field damage data. The collected data are then utilized to train a surrogate model that maps the relationship between tank characteristics, seismic hazard parameters, and seismic-induced damage via a data-driven surrogate model. Different types of surrogate algorithms, including naïve Bayes, k-nearest neighbors, decision tree, and random forest, are investigated, and results in terms of accuracy are reported. The model that yields the most accurate predictions is employed to predict future damage as a function of tank characteristics and seismic hazard intensity level. Results show that the proposed approach can be used to estimate the extent of damage in steel liquid storage tanks, where the use of data-driven surrogates represents a viable alternative to computationally expensive numerical simulation models.

Keywords: damage prediction , data-driven model, seismic performance, steel liquid storage tanks, surrogate model

Procedia PDF Downloads 124

Authors: Shengnan Chen, Shuhua Wang

Abstract:

Successful production of hydrocarbon from unconventional tight oil reserves has changed the energy landscape in North America. The oil contained within these reservoirs typically will not flow to the wellbore at economic rates without assistance from advanced horizontal well and multi-stage hydraulic fracturing. Efficient and economic development of these reserves is a priority of society, government, and industry, especially under the current low oil prices. Meanwhile, society needs technological and process innovations to enhance oil recovery while concurrently reducing environmental impacts. Recently, big data analysis and artificial intelligence become very popular, developing data-driven insights for better designs and decisions in various engineering disciplines. However, the application of data mining in petroleum engineering is still in its infancy. The objective of this research aims to apply intelligent data analysis and data-driven models to exploit unconventional oil reserves both efficiently and economically. More specifically, a comprehensive database including the reservoir geological data, reservoir geophysical data, well completion data and production data for thousands of wells is firstly established to discover the valuable insights and knowledge related to tight oil reserves development. Several data analysis methods are introduced to analysis such a huge dataset. For example, K-means clustering is used to partition all observations into clusters; principle component analysis is applied to emphasize the variation and bring out strong patterns in the dataset, making the big data easy to explore and visualize; exploratory factor analysis (EFA) is used to identify the complex interrelationships between well completion data and well production data. Different data mining techniques, such as artificial neural network, fuzzy logic, and machine learning technique are then summarized, and appropriate ones are selected to analyze the database based on the prediction accuracy, model robustness, and reproducibility. Advanced knowledge and patterned are finally recognized and integrated into a modified self-adaptive differential evolution optimization workflow to enhance the oil recovery and maximize the net present value (NPV) of the unconventional oil resources. This research will advance the knowledge in the development of unconventional oil reserves and bridge the gap between the big data and performance optimizations in these formations. The newly developed data-driven optimization workflow is a powerful approach to guide field operation, which leads to better designs, higher oil recovery and economic return of future wells in the unconventional oil reserves.

Keywords: big data, artificial intelligence, enhance oil recovery, unconventional oil reserves

Procedia PDF Downloads 262

Authors: Mai Abdul Latif, Yuntian Feng

Abstract:

Applied Element Method (AEM) is a method that was developed to aid in the analysis of the collapse of structures. Current available methods cannot deal with structural collapse accurately; however, AEM can simulate the behavior of a structure from an initial state of no loading until collapse of the structure. The elements in AEM are connected with sets of normal and shear springs along the edges of the elements, that represent the stresses and strains of the element in that region. The elements are rigid, and the material properties are introduced through the spring stiffness. Nonlinear dynamic analysis has been widely modelled using the finite element method for analysis of progressive collapse of structures; however, difficulties in the analysis were found at the presence of excessively deformed elements with cracking or crushing, as well as having a high computational cost, and difficulties on choosing the appropriate material models for analysis. The Applied Element method is developed and coded to significantly improve the accuracy and also reduce the computational costs of the method. The scheme works for both linear elastic, and nonlinear cases, including elasto-plastic materials. This paper will focus on elastic and elasto-plastic material behaviour, where the number of springs required for an accurate analysis is tested. A steel cantilever beam is used as the structural element for the analysis. The first modification of the method is based on the Gaussian Quadrature to distribute the springs. Usually, the springs are equally distributed along the face of the element, but it was found that using Gaussian springs, only up to 2 springs were required for perfectly elastic cases, while with equal springs at least 5 springs were required. The method runs on a Newton-Raphson iteration scheme, and quadratic convergence was obtained. The second modification is based on adapting the number of springs required depending on the elasticity of the material. After the first Newton Raphson iteration, Von Mises stress conditions were used to calculate the stresses in the springs, and the springs are classified as elastic or plastic. Then transition springs, springs located exactly between the elastic and plastic region, are interpolated between regions to strictly identify the elastic and plastic regions in the cross section. Since a rectangular cross-section was analyzed, there were two plastic regions (top and bottom), and one elastic region (middle). The results of the present study show that elasto-plastic cases require only 2 springs for the elastic region, and 2 springs for the plastic region. This showed to improve the computational cost, reducing the minimum number of springs in elasto-plastic cases to only 6 springs. All the work is done using MATLAB and the results will be compared to models of structural elements using the finite element method in ANSYS.

Keywords: applied element method, elasto-plastic, Gaussian springs, nonlinear

Procedia PDF Downloads 204

Authors: Tayo P. Ogundunmade, Adedayo A. Adepoju

Abstract:

Terrorist attacks in liberal democracies bring about a few pessimistic results, for example, sabotaged public support in the governments they target, disturbing the peace of a protected environment underwritten by the state, and a limitation of individuals from adding to the advancement of the country, among others. Hence, seeking for techniques to understand the different factors involved in terrorism and how to deal with those factors in order to completely stop or reduce terrorist activities is the topmost priority of the government in every country. This research aim is to develop an efficient deep learning-based predictive model for the prediction of future terrorist activities in Nigeria, addressing low-quality prediction accuracy problems associated with the existing solution methods. The proposed predictive AI-based model as a counterterrorism tool will be useful by governments and law enforcement agencies to protect the lives of individuals in society and to improve the quality of life in general. A Heterogeneous Bayesian Neural Network (HETBNN) model was derived with Gaussian error normal distribution. Three primary transfer functions (HOTTFs), as well as two derived transfer functions (HETTFs) arising from the convolution of the HOTTFs, are namely; Symmetric Saturated Linear transfer function (SATLINS ), Hyperbolic Tangent transfer function (TANH), Hyperbolic Tangent sigmoid transfer function (TANSIG), Symmetric Saturated Linear and Hyperbolic Tangent transfer function (SATLINS-TANH) and Symmetric Saturated Linear and Hyperbolic Tangent Sigmoid transfer function (SATLINS-TANSIG). Data on the Terrorist activities in Nigeria gathered through questionnaires for the purpose of this study were used. Mean Square Error (MSE), Mean Absolute Error (MAE) and Test Error are the forecast prediction criteria. The results showed that the HETFs performed better in terms of prediction and factors associated with terrorist activities in Nigeria were determined. The proposed predictive deep learning-based model will be useful to governments and law enforcement agencies as an effective counterterrorism mechanism to understand the parameters of terrorism and to design strategies to deal with terrorism before an incident actually happens and potentially causes the loss of precious lives. The proposed predictive AI-based model will reduce the chances of terrorist activities and is particularly helpful for security agencies to predict future terrorist activities.

Keywords: activation functions, Bayesian neural network, mean square error, test error, terrorism

Procedia PDF Downloads 138

Authors: Abdolrasoul Daneshjoo

Abstract:

Soccer is a game which draws more attention in different countries especially in Brazil. Kicking among different skills in soccer and soccer players is an excellent role for the success and preference of a team. The way of point gaining in this game is passing the ball over the goal lines which are gained by shoot skill in attack time and or during the penalty kicks.Regarding the above assumption, identifying the effective factors in instep kicking in different distances shoot with maximum force and high accuracy or pass and penalty kick, may assist the coaches and players in raising qualitative level of performing the skill.The aim of the present study was to study of a few kinematical parameters in instep kicking from 5 and 7 meter distance among the male and female elite soccer players.24 right dominant lower limb subjects (12 males and 12 females) among Tehran elite soccer players with average and the standard deviation (22.5 ± 1.5) & (22.08± 1.31) years, height of (179.5 ± 5.81) & (164.3 ± 4.09) cm, weight of (69.66 ± 4.09) & (53.16 ± 3.51) kg, %BMI (21.06 ± .731) & (19.67 ± .709), having playing history of (4 ± .73) & (3.08 ± .66) years respectively participated in this study. They had at least two years of continuous playing experience in Tehran soccer league.For sampling player's kick; Kinemetrix Motion analysis with three cameras with 1000 Hz was used. Five reflective markers were placed laterally on the kicking leg over anatomical points (the iliac crest, major trochanter, lateral epicondyle of femur, lateral malleolus, and lateral aspect of distal head of the fifth metatarsus). Instep kick was filmed, with one step approach and 30 to 45 degrees angle from stationary ball. Three kicks were filmed, one kick selected for further analyses. Using Kinemetrix 3D motion analysis software, the position of the markers was analyzed. Descriptive statistics were used to describe the mean and standard deviation, while the analysis of variance, and independent t-test (P < 0.05) were used to compare the kinematic parameters between two genders.Among the evaluated parameters, the knee acceleration, the thigh angular velocity, the angle of knee proportionately showed significant relationship with consequence of kick. While company performance on 5m in 2 genders, significant differences were observed in internal – external displacement of toe, ankle, hip and the velocity of toe, ankle and the acceleration of toe and the angular velocity of pelvic, thigh and before time contact . Significant differences showed the internal – external displacement of toe, the ankle, the knee and the hip, the iliac crest and the velocity of toe, the ankle and acceleration of ankle and angular velocity of the pelvic and the knee.

Keywords: biomechanics, kinematics, instep kicking, soccer

Procedia PDF Downloads 467

Authors: Alexander Buhr, Klaus Ehrenfried

Abstract:

Trackside induced airflow velocities, also known as slipstream velocities, are an important criterion for the design of high-speed trains. The maximum permitted values are given by the Technical Specifications for Interoperability (TSI) and have to be checked in the approval process. For train manufactures it is of great interest to know in advance, how new train geometries would perform in TSI tests. The Reynolds number in moving model experiments is lower compared to full-scale. Especially the limited model length leads to a thinner boundary layer at the rear end. The hypothesis is that the boundary layer rolls up to characteristic flow structures in the train wake, in which the maximum flow velocities can be observed. The idea is to enlarge the boundary layer using roughness elements at the train model head so that the ratio between the boundary layer thickness and the car width at the rear end is comparable to a full-scale train. This may lead to similar flow structures in the wake and better prediction accuracy for TSI tests. In this case, the design of the roughness elements is limited by the moving model rig. Small rectangular roughness shapes are used to get a sufficient effect on the boundary layer, while the elements are robust enough to withstand the high accelerating and decelerating forces during the test runs. For this investigation, High-Speed Particle Image Velocimetry (HS-PIV) measurements on an ICE3 train model have been realized in the moving model rig of the DLR in Göttingen, the so called tunnel simulation facility Göttingen (TSG). The flow velocities within the boundary layer are analysed in a plain parallel to the ground. The height of the plane corresponds to a test position in the EN standard (TSI). Three different shapes of roughness elements are tested. The boundary layer thickness and displacement thickness as well as the momentum thickness and the form factor are calculated along the train model. Conditional sampling is used to analyse the size and dynamics of the flow structures at the time of maximum velocity in the train wake behind the train. As expected, larger roughness elements increase the boundary layer thickness and lead to larger flow velocities in the boundary layer and in the wake flow structures. The boundary layer thickness, displacement thickness and momentum thickness are increased by using larger roughness especially when applied in the height close to the measuring plane. The roughness elements also cause high fluctuations in the form factors of the boundary layer. Behind the roughness elements, the form factors rapidly are approaching toward constant values. This indicates that the boundary layer, while growing slowly along the second half of the train model, has reached a state of equilibrium.

Keywords: boundary layer, high-speed PIV, ICE3, moving train model, roughness elements

Procedia PDF Downloads 279

Authors: E. Salem

Abstract:

Combustion has been used for a long time as a means of energy extraction. However, in recent years, there has been a further increase in air pollution, through pollutants such as nitrogen oxides, acid etc. In order to solve this problem, there is a need to reduce carbon and nitrogen oxides through learn burning modifying combustors and fuel dilution. A numerical investigation has been done to investigate the effectiveness of several reduced mechanisms in terms of computational time and accuracy, for the combustion of the hydrocarbons/air or diluted with hydrogen in a micro combustor. The simulations were carried out using the ANSYS Fluent 19.1. To validate the results “PREMIX and CHEMKIN” codes were used to calculate 1D premixed flame based on the temperature, composition of burned and unburned gas mixtures. Numerical calculations were carried for several hydrocarbons by changing the equivalence ratios and adding small amounts of hydrogen into the fuel blends then analyzing the flammable limit, the reduction in NOx and CO emissions, then comparing it to experimental data. By solving the conservations equations, several global reduced mechanisms (2-9-12) were obtained. These reduced mechanisms were simulated on a 2D cylindrical tube with dimensions of 40 cm in length and 2.5 cm diameter. The mesh of the model included a proper fine quad mesh, within the first 7 cm of the tube and around the walls. By developing a proper boundary layer, several simulations were performed on hydrocarbon/air blends to visualize the flame characteristics than were compared with experimental data. Once the results were within acceptable range, the geometry of the combustor was modified through changing the length, diameter, adding hydrogen by volume, and changing the equivalence ratios from lean to rich in the fuel blends, the results on flame temperature, shape, velocity and concentrations of radicals and emissions were observed. It was determined that the reduced mechanisms provided results within an acceptable range. The variation of the inlet velocity and geometry of the tube lead to an increase of the temperature and CO2 emissions, highest temperatures were obtained in lean conditions (0.5-0.9) equivalence ratio. Addition of hydrogen blends into combustor fuel blends resulted in; reduction in CO and NOx emissions, expansion of the flammable limit, under the condition of having same laminar flow, and varying equivalence ratio with hydrogen additions. The production of NO is reduced because the combustion happens in a leaner state and helps in solving environmental problems.

Keywords: combustor, equivalence-ratio, hydrogenation, premixed flames

Procedia PDF Downloads 96

Authors: Jutta Ransmayr

Abstract:

The digital age has long arrived at Austrian schools, so both society and educationalists demand that digital means should be integrated accordingly to day-to-day school routines. Therefore, the Austrian school-leaving exam (A-levels) can now be written either by hand or by using a computer. However, the choice of writing medium (pen and paper or computer) for written examination papers, which are considered 'high-stakes' exams, raises a number of questions that have not yet been adequately investigated and answered until recently, such as: What effects do the different conditions of text production in the written German A-levels have on the component of normative linguistic accuracy? How do the spelling skills of German A-level papers written with a pen differ from those that the students wrote on the computer? And how is the teacher's assessment related to this? Which practical desiderata for German didactics can be derived from this? In a trilateral pilot project of the Austrian Center for Digital Humanities (ACDH) of the Austrian Academy of Sciences and the University of Vienna in cooperation with the Austrian Ministry of Education and the Council for German Orthography, these questions were investigated. A representative Austrian learner corpus, consisting of around 530 German A-level papers from all over Austria (pen and computer written), was set up in order to subject it to a quantitative (corpus-linguistic and statistical) and qualitative investigation with regard to the spelling and punctuation performance of the high school graduates and the differences between pen- and computer-written papers and their assessments. Relevant studies are currently available mainly from the Anglophone world. These have shown that writing on the computer increases the motivation to write, has positive effects on the length of the text, and, in some cases, also on the quality of the text. Depending on the writing situation and other technical aids, better results in terms of spelling and punctuation could also be found in the computer-written texts as compared to the handwritten ones. Studies also point towards a tendency among teachers to rate handwritten texts better than computer-written texts. In this paper, the first comparable results from the German-speaking area are to be presented. Research results have shown that, on the one hand, there are significant differences between handwritten and computer-written work with regard to performance in orthography and punctuation. On the other hand, the corpus linguistic investigation and the subsequent statistical analysis made it clear that not only the teachers' assessments of the students’ spelling performance vary enormously but also the overall assessments of the exam papers – the factor of the production medium (pen and paper or computer) also seems to play a decisive role.

Keywords: exam paper assessment, pen and paper or computer, learner corpora, linguistics

Procedia PDF Downloads 139

Authors: W. Meron Mebrahtu, R. Absi

Abstract:

Velocity distribution in turbulent open-channel flows is organized in a complex manner. This is due to the large spatial and temporal variability of fluid motion resulting from the free-surface turbulent flow condition. This phenomenon is complicated further due to the complex geometry of channels and the presence of solids transported. Thus, several efforts were made to understand the phenomenon and obtain accurate mathematical models that are suitable for engineering applications. However, predictions are inaccurate because oversimplified assumptions are involved in modeling this complex phenomenon. Therefore, the aim of this work is to study velocity distribution profiles and obtain simple, more accurate, and predictive mathematical models. Particular focus will be made on the acceptable simplification of the general transport equations and an accurate representation of eddy viscosity. Wide rectangular open-channel seems suitable to begin the study; other assumptions are smooth-wall, and sediment-free flow under steady and uniform flow conditions. These assumptions will allow examining the effect of the bottom wall and the free surface only, which is a necessary step before dealing with more complex flow scenarios. For this flow condition, two ordinary differential equations are obtained for velocity profiles; from the Reynolds-averaged Navier-Stokes (RANS) equation and equilibrium consideration between turbulent kinetic energy (TKE) production and dissipation. Then different analytic models for eddy viscosity, TKE, and mixing length were assessed. Computation results for velocity profiles were compared to experimental data for different flow conditions and the well-known linear, log, and log-wake laws. Results show that the model based on the RANS equation provides more accurate velocity profiles. In the viscous sublayer and buffer layer, the method based on Prandtl’s eddy viscosity model and Van Driest mixing length give a more precise result. For the log layer and outer region, a mixing length equation derived from Von Karman’s similarity hypothesis provides the best agreement with measured data except near the free surface where an additional correction based on a damping function for eddy viscosity is used. This method allows more accurate velocity profiles with the same value of the damping coefficient that is valid under different flow conditions. This work continues with investigating narrow channels, complex geometries, and the effect of solids transported in sewers.

Keywords: accuracy, eddy viscosity, sewers, velocity profile

Procedia PDF Downloads 85

Authors: Eliane G. Tótoli, Hérida Regina N. Salgado

Abstract:

Daptomycin is an important antimicrobial agent used in clinical practice nowadays, since it is very active against some Gram-positive bacteria that are particularly challenges for the medicine, such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococci (VRE). The importance of environmental preservation has receiving special attention since last years. Considering the evident need to protect the natural environment and the introduction of strict quality requirements regarding analytical procedures used in pharmaceutical analysis, the industries must seek environmentally friendly alternatives in relation to the analytical methods and other processes that they follow in their routine. In view of these factors, green analytical chemistry is prevalent and encouraged nowadays. In this context, infrared spectroscopy stands out. This is a method that does not use organic solvents and, although it is formally accepted for the identification of individual compounds, also allows the quantitation of substances. Considering that there are few green analytical methods described in literature for the analysis of daptomycin, the aim of this work was the development and validation of a green analytical method for the quantification of this drug in lyophilized powder for injectable solution, by Fourier-transform infrared spectroscopy (FT-IR). Method: Translucent potassium bromide pellets containing predetermined amounts of the drug were prepared and subjected to spectrophotometric analysis in the mid-infrared region. After obtaining the infrared spectrum and with the assistance of the IR Solution software, quantitative analysis was carried out in the spectral region between 1575 and 1700 cm-1, related to a carbonyl band of the daptomycin molecule, and this band had its height analyzed in terms of absorbance. The method was validated according to ICH guidelines regarding linearity, precision (repeatability and intermediate precision), accuracy and robustness. Results and discussion: The method showed to be linear (r = 0.9999), precise (RSD% < 2.0), accurate and robust, over a concentration range from 0.2 to 0.6 mg/pellet. In addition, this technique does not use organic solvents, which is one great advantage over the most common analytical methods. This fact contributes to minimize the generation of organic solvent waste by the industry and thereby reduces the impact of its activities on the environment. Conclusion: The validated method proved to be adequate to quantify daptomycin in lyophilized powder for injectable solution and can be used for its routine analysis in quality control. In addition, the proposed method is environmentally friendly, which is in line with the global trend.

Keywords: daptomycin, Fourier-transform infrared spectroscopy, green analytical chemistry, quality control, spectrometry in IR region

Procedia PDF Downloads 356

Authors: Shubham Gogri, Lucia Florescu

Abstract:

Medical diagnosis and radiotherapy treatment planning using Computed Tomography (CT) rely on the quantitative accuracy and quality of the CT images. At the same time, requirements for CT imaging include reducing the radiation dose exposure to patients and minimizing scanning time. A solution to this is the sparse-view CT technique, based on a reduced number of projection views. This, however, introduces a new problem— the incomplete projection data results in lower quality of the reconstructed images. To tackle this issue, deep learning methods have been applied to enhance the quality of the sparse-view CT images. A first approach involved employing Mir-Net, a dedicated deep neural network designed for image enhancement. This showed promise, utilizing an intricate architecture comprising encoder and decoder networks, along with the incorporation of the Charbonnier Loss. However, this approach was computationally demanding. Subsequently, a specialized Generative Adversarial Network (GAN) architecture, rooted in the Pix2Pix framework, was implemented. This GAN framework involves a U-Net-based Generator and a Discriminator based on Convolutional Neural Networks. To bolster the GAN's performance, both Charbonnier and Wasserstein loss functions were introduced, collectively focusing on capturing minute details while ensuring training stability. The integration of the perceptual loss, calculated based on feature vectors extracted from the VGG16 network pretrained on the ImageNet dataset, further enhanced the network's ability to synthesize relevant images. A series of comprehensive experiments with clinical CT data were conducted, exploring various GAN loss functions, including Wasserstein, Charbonnier, and perceptual loss. The outcomes demonstrated significant image quality improvements, confirmed through pertinent metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) between the corrected images and the ground truth. Furthermore, learning curves and qualitative comparisons added evidence of the enhanced image quality and the network's increased stability, while preserving pixel value intensity. The experiments underscored the potential of deep learning frameworks in enhancing the visual interpretation of CT scans, achieving outcomes with SSIM values close to one and PSNR values reaching up to 76.

Keywords: generative adversarial networks, sparse view computed tomography, CT image correction, Mir-Net

Procedia PDF Downloads 120

Authors: Jun Won Kim

Abstract:

Objective: Theta-phase gamma-amplitude coupling (TGC) was used as a novel evidence-based tool to reflect the dysfunctional cortico-thalamic interaction in patients with schizophrenia. However, to our best knowledge, no studies have reported the diagnostic utility of the TGC in the resting-state electroencephalographic (EEG) of neuroleptic-naive patients with schizophrenia compared to healthy controls. Thus, the purpose of this EEG study was to understand the underlying mechanisms in patients with schizophrenia by comparing the TGC at rest between two groups and to evaluate the diagnostic utility of TGC. Method: The subjects included 90 patients with schizophrenia and 90 healthy controls. All patients were diagnosed with schizophrenia according to the criteria of Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) by two independent psychiatrists using semi-structured clinical interviews. Because patients were either drug-naïve (first episode) or had not been taking psychoactive drugs for one month before the study, we could exclude the influence of medications. Five frequency bands were defined for spectral analyses: delta (1–4 Hz), theta (4–8 Hz), slow alpha (8–10 Hz), fast alpha (10–13.5 Hz), beta (13.5–30 Hz), and gamma (30-80 Hz). The spectral power of the EEG data was calculated with fast Fourier Transformation using the 'spectrogram.m' function of the signal processing toolbox in Matlab. An analysis of covariance (ANCOVA) was performed to compare the TGC results between the groups, which were adjusted using a Bonferroni correction (P < 0.05/19 = 0.0026). Receiver operator characteristic (ROC) analysis was conducted to examine the discriminating ability of the TGC data for schizophrenia diagnosis. Results: The patients with schizophrenia showed a significant increase in the resting-state TGC at all electrodes. The delta, theta, slow alpha, fast alpha, and beta powers showed low accuracies of 62.2%, 58.4%, 56.9%, 60.9%, and 59.0%, respectively, in discriminating the patients with schizophrenia from the healthy controls. The ROC analysis performed on the TGC data generated the most accurate result among the EEG measures, displaying an overall classification accuracy of 92.5%. Conclusion: As TGC includes phase, which contains information about neuronal interactions from the EEG recording, TGC is expected to be useful for understanding the mechanisms the dysfunctional cortico-thalamic interaction in patients with schizophrenia. The resting-state TGC value was increased in the patients with schizophrenia compared to that in the healthy controls and had a higher discriminating ability than the other parameters. These findings may be related to the compensatory hyper-arousal patterns of the dysfunctional default-mode network (DMN) in schizophrenia. Further research exploring the association between TGC and medical or psychiatric conditions that may confound EEG signals will help clarify the potential utility of TGC.

Keywords: quantitative electroencephalography (QEEG), theta-phase gamma-amplitude coupling (TGC), schizophrenia, diagnostic utility

Procedia PDF Downloads 117

Authors: Ciara Pienaar, J. W. Odendaal, J. Joubert, J. C. Smit

Abstract:

Radar cross section (RCS) of dielectric objects play an important role in many applications, such as low observability technology development, drone detection, and monitoring as well as coastal surveillance. Various materials are used to construct the targets of interest such as metal, wood, composite materials, radar absorbent materials, and other dielectrics. Since simulated datasets are increasingly being used to supplement infield measurements, as it is more cost effective and a larger variety of targets can be simulated, it is important to have a high level of confidence in the predicted results. Confidence can be attained through validation. Various computational electromagnetic (CEM) methods are capable of predicting the RCS of dielectric targets. This study will extend previous studies by validating full-wave and asymptotic RCS simulations of dielectric targets with measured data. The paper will provide measured RCS data of a number of canonical dielectric targets exhibiting different material properties. As stated previously, these measurements are used to validate numerous CEM methods. The dielectric properties are accurately characterized to reduce the uncertainties in the simulations. Finally, an analysis of the sensitivity of oblique and normal incidence scattering predictions to material characteristics is also presented. In this paper, the ability of several CEM methods, including method of moments (MoM), and physical optics (PO), to calculate the RCS of dielectrics were validated with measured data. A few dielectrics, exhibiting different material properties, were selected and several canonical targets, such as flat plates and cylinders, were manufactured. The RCS of these dielectric targets were measured in a compact range at the University of Pretoria, South Africa, over a frequency range of 2 to 18 GHz and a 360° azimuth angle sweep. This study also investigated the effect of slight variations in the material properties on the calculated RCS results, by varying the material properties within a realistic tolerance range and comparing the calculated RCS results. Interesting measured and simulated results have been obtained. Large discrepancies were observed between the different methods as well as the measured data. It was also observed that the accuracy of the RCS data of the dielectrics can be frequency and angle dependent. The simulated RCS for some of these materials also exhibit high sensitivity to variations in the material properties. Comparison graphs between the measured and simulation RCS datasets will be presented and the validation thereof will be discussed. Finally, the effect that small tolerances in the material properties have on the calculated RCS results will be shown. Thus the importance of accurate dielectric material properties for validation purposes will be discussed.

Keywords: asymptotic, CEM, dielectric scattering, full-wave, measurements, radar cross section, validation

Procedia PDF Downloads 219

Authors: Xiangtuo Chen, Paul-Henry Cournéde

Abstract:

Crop yield prediction is a paramount issue in agriculture. The main idea of this paper is to find out efficient way to predict the yield of corn based meteorological records. The prediction models used in this paper can be classified into model-driven approaches and data-driven approaches, according to the different modeling methodologies. The model-driven approaches are based on crop mechanistic modeling. They describe crop growth in interaction with their environment as dynamical systems. But the calibration process of the dynamic system comes up with much difficulty, because it turns out to be a multidimensional non-convex optimization problem. An original contribution of this paper is to propose a statistical methodology, Multi-Scenarios Parameters Estimation (MSPE), for the parametrization of potentially complex mechanistic models from a new type of datasets (climatic data, final yield in many situations). It is tested with CORNFLO, a crop model for maize growth. On the other hand, the data-driven approach for yield prediction is free of the complex biophysical process. But it has some strict requirements about the dataset. A second contribution of the paper is the comparison of these model-driven methods with classical data-driven methods. For this purpose, we consider two classes of regression methods, methods derived from linear regression (Ridge and Lasso Regression, Principal Components Regression or Partial Least Squares Regression) and machine learning methods (Random Forest, k-Nearest Neighbor, Artificial Neural Network and SVM regression). The dataset consists of 720 records of corn yield at county scale provided by the United States Department of Agriculture (USDA) and the associated climatic data. A 5-folds cross-validation process and two accuracy metrics: root mean square error of prediction(RMSEP), mean absolute error of prediction(MAEP) were used to evaluate the crop prediction capacity. The results show that among the data-driven approaches, Random Forest is the most robust and generally achieves the best prediction error (MAEP 4.27%). It also outperforms our model-driven approach (MAEP 6.11%). However, the method to calibrate the mechanistic model from dataset easy to access offers several side-perspectives. The mechanistic model can potentially help to underline the stresses suffered by the crop or to identify the biological parameters of interest for breeding purposes. For this reason, an interesting perspective is to combine these two types of approaches.

Keywords: crop yield prediction, crop model, sensitivity analysis, paramater estimation, particle swarm optimization, random forest

Procedia PDF Downloads 208

Authors: Kien Vu, Jien Min Soh, Mohamed Jahangir Abubacker, Piyawut Pattamanon, Soojin Lee, Suvro Banerjee

Abstract:

Recent advances in Generative Artificial Intelligence (GenAI), in particular Large Language Models (LLMs) have shown promise to disrupt multiple industries at scale. However, LLMs also present unique challenges, notably, these so-called "hallucination" which is the generation of outputs that are not grounded in the input data that hinders its adoption into production. Common practice to mitigate hallucination problem is utilizing Retrieval Agmented Generation (RAG) system to ground LLMs'response to ground truth. RAG converts the grounding documents into embeddings, retrieve the relevant parts with vector similarity between user's query and documents, then generates a response that is not only based on its pre-trained knowledge but also on the specific information from the retrieved documents. However, the RAG system is not suitable for tabular data and subsequent data analysis tasks due to multiple reasons such as information loss, data format, and retrieval mechanism. In this study, we have explored a novel methodology that combines planning-and-execution and code generation agents to enhance LLMs' data analysis capabilities. The approach enables LLMs to autonomously dissect a complex analytical task into simpler sub-tasks and requirements, then convert them into executable segments of code. In the final step, it generates the complete response from output of the executed code. When deployed beta version on DataSense, the property insight tool of PropertyGuru, the approach yielded promising results, as it was able to provide market insights and data visualization needs with high accuracy and extensive coverage by abstracting the complexities for real-estate agents and developers from non-programming background. In essence, the methodology not only refines the analytical process but also serves as a strategic tool for real estate professionals, aiding in market understanding and enhancement without the need for programming skills. The implication extends beyond immediate analytics, paving the way for a new era in the real estate industry characterized by efficiency and advanced data utilization.

Keywords: large language model, reasoning, planning and execution, code generation, natural language processing, prompt engineering, data analysis, real estate, data sense, PropertyGuru

Procedia PDF Downloads 45

Authors: Pooria Norouzi, Nicholas Tsang, Adam van der Goes, Joseph Yu, Douglas Zheng, Sirine Maleej

Abstract:

In this study, virtual meters will be designed and used for energy balance measurements of an air handling unit (AHU). The method aims to replace traditional physical sensors in heating, ventilation, and air conditioning (HVAC) systems with simulated virtual meters. Due to the inability to manage and monitor these systems, many HVAC systems have a high level of inefficiency and energy wastage. Virtual meters are implemented and applied in an actual HVAC system, and the result confirms the practicality of mathematical sensors for alternative energy measurement. While most residential buildings and offices are commonly not equipped with advanced sensors, adding, exploiting, and monitoring sensors and measurement devices in the existing systems can cost thousands of dollars. The first purpose of this study is to provide an energy consumption rate based on available sensors and without any physical energy meters. It proves the performance of virtual meters in HVAC systems as reliable measurement devices. To demonstrate this concept, mathematical models are created for AHU-07, located in building NE01 of the British Columbia Institute of Technology (BCIT) Burnaby campus. The models will be created and integrated with the system’s historical data and physical spot measurements. The actual measurements will be investigated to prove the models' accuracy. Based on preliminary analysis, the resulting mathematical models are successful in plotting energy consumption patterns, and it is concluded confidently that the results of the virtual meter will be close to the results that physical meters could achieve. In the second part of this study, the use of virtual meters is further assisted by artificial intelligence (AI) in the HVAC systems of building to improve energy management and efficiency. By the data mining approach, virtual meters’ data is recorded as historical data, and HVAC system energy consumption prediction is also implemented in order to harness great energy savings and manage the demand and supply chain effectively. Energy prediction can lead to energy-saving strategies and considerations that can open a window in predictive control in order to reach lower energy consumption. To solve these challenges, the energy prediction could optimize the HVAC system and automates energy consumption to capture savings. This study also investigates AI solutions possibility for autonomous HVAC efficiency that will allow quick and efficient response to energy consumption and cost spikes in the energy market.

Keywords: virtual meters, HVAC, artificial intelligence, energy consumption prediction

Procedia PDF Downloads 77

Authors: Paramita Banerjee, K. R. S. Chandrakumar, G. P. Das

Abstract:

Bulk MgH2 has drawn much attention for the purpose of hydrogen storage because of its high hydrogen storage capacity (~7.7 wt %) as well as low cost and abundant availability. However, its practical usage has been hindered because of its high hydrogen desorption enthalpy (~0.8 eV/H2 molecule), which results in an undesirable desorption temperature of 3000C at 1 bar H2 pressure. To surmount the limitations of bulk MgH2 for the purpose of hydrogen storage, a detailed first-principles density functional theory (DFT) based study on the structure and stability of neutral (Mgm) and positively charged (Mgm+) Mg nanoclusters of different sizes (m = 2, 4, 8 and 12), as well as their interaction with molecular hydrogen (H2), is reported here. It has been found that due to the absence of d-electrons within the Mg atoms, hydrogen remained in molecular form even after its interaction with neutral and charged Mg nanoclusters. Interestingly, the H2 molecules do not enter into the interstitial positions of the nanoclusters. Rather, they remain on the surface by ornamenting these nanoclusters and forming new structures with a gravimetric density higher than 15 wt %. Our observation is that the inclusion of Grimme’s DFT-D3 dispersion correction in this weakly interacting system has a significant effect on binding of the H2 molecules with these nanoclusters. The dispersion corrected interaction energy (IE) values (0.1-0.14 eV/H2 molecule) fall in the right energy window, that is ideal for hydrogen storage. These IE values are further verified by using high-level coupled-cluster calculations with non-iterative triples corrections i.e. CCSD(T), (which has been considered to be a highly accurate quantum chemical method) and thereby confirming the accuracy of our ‘dispersion correction’ incorporated DFT calculations. The significance of the polarization and dispersion energy in binding of the H2 molecules are confirmed by performing energy decomposition analysis (EDA). A total of 16, 24, 32 and 36 H2 molecules can be attached to the neutral and charged nanoclusters of size m = 2, 4, 8 and 12 respectively. Ab-initio molecular dynamics (AIMD) simulation shows that the outermost H2 molecules are desorbed at a rather low temperature viz. 150 K (-1230C) which is expected. However, complete dehydrogenation of these nanoclusters occur at around 1000C. Most importantly, the host nanoclusters remain stable up to ~500 K (2270C). All these results on the adsorption and desorption of molecular hydrogen with neutral and charged Mg nanocluster systems indicate towards the possibility of reducing the dehydrogenation temperature of bulk MgH2 by designing new Mg-based nano materials which will be able to adsorb molecular hydrogen via this weak Mg-H2 interaction, rather than the strong Mg-H bonding. Notwithstanding the fact that in practical applications, these interactions will be further complicated by the effect of substrates as well as interactions with other clusters, the present study has implications on our fundamental understanding to this problem.

Keywords: density functional theory, DFT, hydrogen storage, molecular dynamics, molecular hydrogen adsorption, nanoclusters, physisorption

Procedia PDF Downloads 398

Authors: Angela T. H. Kwan, Saman Arfaie, Joseph Therriault, Zahra Azizi, Firoza Z. Lussier, Cecile Tissot, Mira Chamoun, Gleb Bezgin, Stijn Servaes, Jenna Stevenon, Nesrine Rahmouni, Vanessa Pallen, Serge Gauthier, Pedro Rosa-Neto

Abstract:

Alzheimer’s disease (AD) can be detected in living people using in vivo biomarkers of amyloid-β (Aβ) and tau, even in the absence of cognitive impairment during the preclinical phase. [¹⁸F]-MK-6420 is a high affinity positron emission tomography (PET) tracer that quantifies tau neurofibrillary tangles, but its ability to predict cognitive changes associated with early AD symptoms, such as memory decline, is unclear. Here, we assess the prognostic accuracy of baseline [18F]-MK-6420 tau PET for predicting longitudinal memory decline in asymptomatic elderly individuals. In a longitudinal observational study, we evaluated a cohort of cognitively normal elderly participants (n = 111) from the Translational Biomarkers in Aging and Dementia (TRIAD) study (data collected between October 2017 and July 2020, with a follow-up period of 12 months). All participants underwent tau PET with [¹⁸F]-MK-6420 and Aβ PET with [¹⁸F]-AZD-4694. The exclusion criteria included the presence of head trauma, stroke, or other neurological disorders. There were 111 eligible participants who were chosen based on the availability of Aβ PET, tau PET, magnetic resonance imaging (MRI), and APOEε4 genotyping. Among these participants, the mean (SD) age was 70.1 (8.6) years; 20 (18%) were tau PET positive, and 71 of 111 (63.9%) were women. A significant association between baseline Braak I-II [¹⁸F]-MK-6240 SUVR positivity and change in composite memory score was observed at the 12-month follow-up, after correcting for age, sex, and years of education (Logical Memory and RAVLT, standardized beta = -0.52 (-0.82-0.21), p < 0.001, for dichotomized tau PET and -1.22 (-1.84-(-0.61)), p < 0.0001, for continuous tau PET). Moderate cognitive decline was observed for A+T+ over the follow-up period, whereas no significant change was observed for A-T+, A+T-, and A-T-, though it should be noted that the A-T+ group was small.Our results indicate that baseline tau neurofibrillary tangle pathology is associated with longitudinal changes in memory function, supporting the use of [¹⁸F]-MK-6420 PET to predict the likelihood of asymptomatic elderly individuals experiencing future memory decline. Overall, [¹⁸F]-MK-6420 PET is a promising tool for predicting memory decline in older adults without cognitive impairment at baseline. This is of critical relevance as the field is shifting towards a biological model of AD defined by the aggregation of pathologic tau. Therefore, early detection of tau pathology using [¹⁸F]-MK-6420 PET provides us with the hope that living patients with AD may be diagnosed during the preclinical phase before it is too late.

Keywords: alzheimer’s disease, braak I-II, in vivo biomarkers, memory, PET, tau

Procedia PDF Downloads 55

Authors: Marlin Mubarak

Abstract:

Background: Endometriosis is a complex disabling disease affecting young females in the reproductive period mainly. The aim of this project is to generate a diagnostic model to predict severity and stage of endometriosis prior to Laparoscopic surgery. This will help to improve the pre-operative diagnostic accuracy of stage 3 & 4 endometriosis and as a result, refer relevant women to a specialist centre for complex Laparoscopic surgery. The model is based on the British Society of Gynaecological Endoscopy (BSGE) pain questionnaire, clinical examination and ultrasound scan. Design: This is a prospective, observational, study, in which women completed the BSGE pain questionnaire, a BSGE requirement. Also, as part of the routine preoperative assessment patient had a routine ultrasound scan and when recto-vaginal and deep infiltrating endometriosis was suspected an MRI was performed. Setting: Luton & Dunstable University Hospital. Patients: Symptomatic women (n = 56) scheduled for laparoscopy due to pelvic pain. The age ranged between 17 – 52 years of age (mean 33.8 years, SD 8.7 years). Interventions: None outside the recognised and established endometriosis centre protocol set up by BSGE. Main Outcome Measure(s): Sensitivity and specificity of endometriosis diagnosis predicted by symptoms based on BSGE pain questionnaire, clinical examinations and imaging. Findings: The prevalence of diagnosed endometriosis was calculated to be 76.8% and the prevalence of advanced stage was 55.4%. Deep infiltrating endometriosis in various locations was diagnosed in 32/56 women (57.1%) and some had DIE involving several locations. Logistic regression analysis was performed on 36 clinical variables to create a simple clinical prediction model. After creating the scoring system using variables with P < 0.05, the model was applied to the whole dataset. The sensitivity was 83.87% and specificity 96%. The positive likelihood ratio was 20.97 and the negative likelihood ratio was 0.17, indicating that the model has a good predictive value and could be useful in predicting advanced stage endometriosis. Conclusions: This is a hypothesis-generating project with one operator, but future proposed research would provide validation of the model and establish its usefulness in the general setting. Predictive tools based on such model could help organise the appropriate investigation in clinical practice, reduce risks associated with surgery and improve outcome. It could be of value for future research to standardise the assessment of women presenting with pelvic pain. The model needs further testing in a general setting to assess if the initial results are reproducible.

Keywords: deep endometriosis, endometriosis, minimally invasive, MRI, ultrasound.

Procedia PDF Downloads 320

Authors: Asma Omri, Noureddine Benothman, Sofiane Sayahi, Fethi Tlili, Hichem Besbes

Abstract:

Cycling, a popular mode of transportation, can also be perilous due to cyclists' vulnerability to collisions with vehicles and obstacles. This paper presents an innovative cyclist safety system based on radar technology designed to offer real-time collision risk warnings to cyclists. The system incorporates a low-power radar sensor affixed to the bicycle and connected to a microcontroller. It leverages radar point cloud detections, a clustering algorithm, and a supervised classifier. These algorithms are optimized for efficiency to run on the TI’s AWR 1843 BOOST radar, utilizing a coarse classification approach distinguishing between cars, trucks, two-wheeled vehicles, and other objects. To enhance the performance of clustering techniques, we propose a 2-Level clustering approach. This approach builds on the state-of-the-art Density-based spatial clustering of applications with noise (DBSCAN). The objective is to first cluster objects based on their velocity, then refine the analysis by clustering based on position. The initial level identifies groups of objects with similar velocities and movement patterns. The subsequent level refines the analysis by considering the spatial distribution of these objects. The clusters obtained from the first level serve as input for the second level of clustering. Our proposed technique surpasses the classical DBSCAN algorithm in terms of geometrical metrics, including homogeneity, completeness, and V-score. Relevant cluster features are extracted and utilized to classify objects using an SVM classifier. Potential obstacles are identified based on their velocity and proximity to the cyclist. To optimize the system, we used the View of Delft dataset for hyperparameter selection and SVM classifier training. The system's performance was assessed using our collected dataset of radar point clouds synchronized with a camera on an Nvidia Jetson Nano board. The radar-based cyclist safety system is a practical solution that can be easily installed on any bicycle and connected to smartphones or other devices, offering real-time feedback and navigation assistance to cyclists. We conducted experiments to validate the system's feasibility, achieving an impressive 85% accuracy in the classification task. This system has the potential to significantly reduce the number of accidents involving cyclists and enhance their safety on the road.

Keywords: 2-level clustering, coarse classification, cyclist safety, warning system based on radar technology

Procedia PDF Downloads 49

Authors: Jose A. Diaz-Olivares, Nafise Mahdavian, Farhad Abtahi, Kaj Lindecrantz, Abdelakram Hafid, Fernando Seoane

Abstract:

Despite new recommendations and guidelines for the evaluation of occupational risk assessments and their prevention, work-related musculoskeletal disorders are still one of the biggest causes of work activity disruption, productivity loss, sick leave and chronic work disability. It affects millions of workers throughout Europe, with a large-scale economic and social burden. These specific efforts have failed to produce significant results yet, probably due to the limited availability and high costs of occupational risk assessment at work, especially when the methods are complex, consume excessive resources or depend on self-evaluations and observations of poor accuracy. To overcome these limitations, a pervasive system of risk assessment tools in real time has been developed, which has the characteristics of a systematic approach, with good precision, usability and resource efficiency, essential to facilitate the prevention of musculoskeletal disorders in the long term. The system allows the combination of different wearable sensors, placed on different limbs, to be used for data collection and evaluation by a software solution, according to the needs and requirements in each individual working environment. This is done in a non-disruptive manner for both the occupational health expert and the workers. The creation of this solution allows us to attend different research activities that require, as an essential starting point, the recording of data with ergonomic value of very diverse origin, especially in real work environments. The software platform is here presented with a complimentary smart clothing system for data acquisition, comprised of a T-shirt containing inertial measurement units (IMU), a vest sensorized with textile electronics, a wireless electrocardiogram (ECG) and thoracic electrical bio-impedance (TEB) recorder and a glove sensorized with variable resistors, dependent on the angular position of the wrist. The collected data is processed in real-time through a mobile application software solution, implemented in commercially available Android-based smartphones and tablet platforms. Based on the collection of this information and its analysis, real-time risk assessment and feedback about postural improvement is possible, adapted to different contexts. The result is a tool which provides added value to ergonomists and occupational health agents, as in situ analysis of postural behavior can assist in a quantitative manner in the evaluation of work techniques and the occupational environment.

Keywords: ergonomics, mobile technologies, risk assessment, smart textiles

Procedia PDF Downloads 95

Authors: Man Fung Ho, Lap So, Jiaqi Zhang, Yuheng Zhao, Huiyang Lu, Tat Shing Choi, K. Y. Michael Wong

Abstract:

Nowadays, a tremendous amount of data is available in the transportation system, enabling the development of various machine learning approaches to make short-term latency predictions. A natural question is then the choice of relevant information to enable accurate predictions. Using traffic data collected from the Taiwan Freeway System, we consider the prediction of short-term latency of a freeway segment with a length of 17 km covering 5 measurement points, each collecting vehicle-by-vehicle data through the electronic toll collection system. The processed data include the past latencies of the freeway segment with different time lags, the traffic conditions of the individual segments (the accumulations, the traffic fluxes, the entrance and exit rates), the total accumulations, and the weekday latency profiles obtained by Gaussian process regression of past data. We arrive at several important conclusions about how data should be refined to obtain accurate predictions, which have implications for future system-wide latency predictions. (1) We find that the prediction of median latency is much more accurate and meaningful than the prediction of average latency, as the latter is plagued by outliers. This is verified by machine-learning prediction using XGBoost that yields a 35% improvement in the mean square error of the 5-minute averaged latencies. (2) We find that the median latency of the segment 15 minutes ago is a very good baseline for performance comparison, and we have evidence that further improvement is achieved by machine learning approaches such as XGBoost and Long Short-Term Memory (LSTM). (3) By analyzing the feature importance score in XGBoost and calculating the mutual information between the inputs and the latencies to be predicted, we identify a sequence of inputs ranked in importance. It confirms that the past latencies are most informative of the predicted latencies, followed by the total accumulation, whereas inputs such as the entrance and exit rates are uninformative. It also confirms that the inputs are much less informative of the average latencies than the median latencies. (4) For predicting the latencies of segments composed of two or three sub-segments, summing up the predicted latencies of each sub-segment is more accurate than the one-step prediction of the whole segment, especially with the latency prediction of the downstream sub-segments trained to anticipate latencies several minutes ahead. The duration of the anticipation time is an increasing function of the traveling time of the upstream segment. The above findings have important implications to predicting the full set of latencies among the various locations in the freeway system.

Keywords: data refinement, machine learning, mutual information, short-term latency prediction

Procedia PDF Downloads 148

Authors: Eugen Rusu

Abstract:

The objective of the work proposed is to perform an analysis of the wave conditions in the Black Sea basin. This is especially focused on the spatial and temporal occurrences and on the dynamics of the most extreme storms in the context of the climate changes. A numerical modelling system, based on the spectral phase averaged wave model SWAN, has been implemented and validated against both in situ measurements and remotely sensed data, all along the sea. Moreover, a successive correction method for the assimilation of the satellite data has been associated with the wave modelling system. This is based on the optimal interpolation of the satellite data. Previous studies show that the process of data assimilation improves considerably the reliability of the results provided by the modelling system. This especially concerns the most sensitive cases from the point of view of the accuracy of the wave predictions, as the extreme storm situations are. Following this numerical approach, it has to be highlighted that the results provided by the wave modelling system above described are in general in line with those provided by some similar wave prediction systems implemented in enclosed or semi-enclosed sea basins. Simulations of this wave modelling system with data assimilation have been performed for the 30-year period 1987-2016. Considering this database, the next step was to analyze the intensity and the dynamics of the higher storms encountered in this period. According to the data resulted from the model simulations, the western side of the sea is considerably more energetic than the rest of the basin. In this western region, regular strong storms provide usually significant wave heights greater than 8m. This may lead to maximum wave heights even greater than 15m. Such regular strong storms may occur several times in one year, usually in the wintertime, or in late autumn, and it can be noticed that their frequency becomes higher in the last decade. As regards the case of the most extreme storms, significant wave heights greater than 10m and maximum wave heights close to 20m (and even greater) may occur. Such extreme storms, which in the past were noticed only once in four or five years, are more recent to be faced almost every year in the Black Sea, and this seems to be a consequence of the climate changes. The analysis performed included also the dynamics of the monthly and annual significant wave height maxima as well as the identification of the most probable spatial and temporal occurrences of the extreme storm events. Finally, it can be concluded that the present work provides valuable information related to the characteristics of the storm conditions and on their dynamics in the Black Sea. This environment is currently subjected to high navigation traffic and intense offshore and nearshore activities and the strong storms that systematically occur may produce accidents with very serious consequences.

Keywords: Black Sea, extreme storms, SWAN simulations, waves

Procedia PDF Downloads 216

Authors: Kitty Kioskli, Theofanis Fotis, Nineta Polemi

Abstract:

The European Union (EU) directive Artificial Intelligence (AI) Act in Article 9 requires that risk management of AI systems includes both technical and human oversight, while according to NIST_AI_RFM (Appendix C) and ENISA AI Framework recommendations, claim that further research is needed to understand the current limitations of social threats and human-AI interaction. AI threats within social contexts significantly affect the security and trustworthiness of the AI systems; they are interrelated and trigger technical threats as well. For example, lack of explainability (e.g. the complexity of models can be challenging for stakeholders to grasp) leads to misunderstandings, biases, and erroneous decisions. Which in turn impact the privacy, security, accountability of the AI systems. Based on the NIST four fundamental criteria for explainability it can also classify the explainability threats into four (4) sub-categories: a) Lack of supporting evidence: AI systems must provide supporting evidence or reasons for all their outputs. b) Lack of Understandability: Explanations offered by systems should be comprehensible to individual users. c) Lack of Accuracy: The provided explanation should accurately represent the system's process of generating outputs. d) Out of scope: The system should only function within its designated conditions or when it possesses sufficient confidence in its outputs. Biases may also stem from historical data reflecting undesired behaviors. When present in the data, biases can permeate the models trained on them, thereby influencing the security and trustworthiness of the of AI systems. Social related AI threats are recognized by various initiatives (e.g., EU Ethics Guidelines for Trustworthy AI), standards (e.g. ISO/IEC TR 24368:2022 on AI ethical concerns, ISO/IEC AWI 42105 on guidance for human oversight of AI systems) and EU legislation (e.g. the General Data Protection Regulation 2016/679, the NIS 2 Directive 2022/2555, the Directive on the Resilience of Critical Entities 2022/2557, the EU AI Act, the Cyber Resilience Act). Measuring social threats, estimating the risks to AI systems associated to these threats and mitigating them is a research challenge. In this paper it will present the efforts of two European Commission Projects (FAITH and THEMIS) from the HorizonEurope programme that analyse the social threats by building cyber-social exercises in order to study human behaviour, traits, cognitive ability, personality, attitudes, interests, and other socio-technical profile characteristics. The research in these projects also include the development of measurements and scales (psychometrics) for human-related vulnerabilities that can be used in estimating more realistically the vulnerability severity, enhancing the CVSS4.0 measurement.

Keywords: social threats, artificial Intelligence, mitigation, social experiment

Procedia PDF Downloads 29

Authors: Netasha Shaikh, Bryony Wood, Demitra Tsivos, Michael Knight, Risto Kauppinen, Elizabeth Coulthard

Abstract:

Background: Effective treatment of dementia will require early diagnosis, before significant brain damage has accumulated. Memory loss is an early symptom of Alzheimer’s disease (AD). The hippocampus, a brain area critical for memory, degenerates early in the course of AD. The hippocampus comprises several subfields. In contrast to healthy aging where CA3 and dentate gyrus are the hippocampal subfields with most prominent atrophy, in AD the CA1 and subiculum are thought to be affected early. Conventional clinical structural neuroimaging is not sufficiently sensitive to identify preferential atrophy in individual subfields. Here, we will explore the sensitivity of new magnetic resonance imaging (MRI) sequences designed to interrogate medial temporal regions as an early marker of Alzheimer’s. As it is likely a combination of tests may predict early Alzheimer’s disease (AD) better than any single test, we look at the potential efficacy of such imaging alone and in combination with standard and novel cognitive tasks of hippocampal dependent memory. Methods: 20 patients with mild cognitive impairment (MCI), 20 with mild-moderate AD and 20 age-matched healthy elderly controls (HC) are being recruited to undergo 3T MRI (with sequences designed to allow volumetric analysis of hippocampal subfields) and a battery of cognitive tasks (including Paired Associative Learning from CANTAB, Hopkins Verbal Learning Test and a novel hippocampal-dependent abstract word memory task). AD participants and healthy controls are being tested just once whereas patients with MCI will be tested twice a year apart. We will compare subfield size between groups and correlate subfield size with cognitive performance on our tasks. In the MCI group, we will explore the relationship between subfield volume, cognitive test performance and deterioration in clinical condition over a year. Results: Preliminary data (currently on 16 participants: 2 AD; 4 MCI; 9 HC) have revealed subfield size differences between subject groups. Patients with AD perform with less accuracy on tasks of hippocampal-dependent memory, and MCI patient performance and reaction times also differ from healthy controls. With further testing, we hope to delineate how subfield-specific atrophy corresponds with changes in cognitive function, and characterise how this progresses over the time course of the disease. Conclusion: Novel sequences on a MRI scanner such as those in route in clinical use can be used to delineate hippocampal subfields in patients with and without dementia. Preliminary data suggest that such subfield analysis, perhaps in combination with cognitive tasks, may be an early marker of AD.

Keywords: Alzheimer's disease, dementia, memory, cognition, hippocampus

Procedia PDF Downloads 552

Authors: Nijanthan Mohan, Rolf Gross, Fabian Theis

Abstract:

The conventional method used in the construction industries often resulted in significant rework since most of the decisions were taken onsite under the pressure of project deadlines and also due to the improper information flow, which results in ineffective coordination. However, today’s architecture, engineering, and construction (AEC) stakeholders demand faster and accurate deliverables, efficient buildings, and smart processes, which turns out to be a tall order. Hence, the building information modelling (BIM) concept was developed as a solution to fulfill the above-mentioned necessities. Even though BIM is successfully implemented in most of the world, it is still in the early stages in Germany, since the stakeholders are sceptical of its reliability and efficiency. Due to the huge capital requirement, the small and medium-sized construction companies are still reluctant to implement BIM workflow in their projects. The purpose of this paper is to analyse the opportunities and obstacles to implementing BIM for prefabrication. Among all other advantages of BIM, pre-fabrication is chosen for this paper because it plays a vital role in creating an impact on time as well as cost factors of a construction project. The positive impact of prefabrication can be explicitly observed by the project stakeholders and participants, which enables the breakthrough of the skepticism factor among the small scale construction companies. The analysis consists of the development of a process workflow for implementing prefabrication in building construction, followed by a practical approach, which was executed with two case studies. The first case study represents on-site prefabrication, and the second was done for off-site prefabrication. It was planned in such a way that the first case study gives a first-hand experience for the workers at the site on the BIM model so that they can make much use of the created BIM model, which is a better representation compared to the traditional 2D plan. The main aim of the first case study is to create a belief in the implementation of BIM models, which was succeeded by the execution of offshore prefabrication in the second case study. Based on the case studies, the cost and time analysis was made, and it is inferred that the implementation of BIM for prefabrication can reduce construction time, ensures minimal or no wastes, better accuracy, less problem-solving at the construction site. It is also observed that this process requires more planning time, better communication, and coordination between different disciplines such as mechanical, electrical, plumbing, architecture, etc., which was the major obstacle for successful implementation. This paper was carried out in the perspective of small and medium-sized mechanical contracting companies for the private building sector in Germany.

Keywords: building information modelling, construction wastes, pre-fabrication, small and medium sized company

Procedia PDF Downloads 89

Authors: Leonard Tiong, Sonit Singh, Kevin Ho Shon, Sarah Lewis

Abstract:

Investigation into the use of artificial intelligence in radiology continues to develop at a rapid rate. During the coronavirus pandemic, the combination of an exponential increase in chest x-rays and unpredictable staff shortages resulted in a huge strain on the department's workload. There is a World Health Organisation estimate that two-thirds of the global population does not have access to diagnostic radiology. Therefore, there could be demand for a program that could detect acute changes in imaging compatible with infection to assist with screening. We generated a conventional neural network and tested its efficacy in recognizing changes compatible with coronavirus infection. Following ethics approval, a deidentified set of 77 normal and 77 abnormal chest x-rays in patients with confirmed coronavirus infection were used to generate an algorithm that could train, validate and then test itself. DICOM and PNG image formats were selected due to their lossless file format. The model was trained with 100 images (50 positive, 50 negative), validated against 28 samples (14 positive, 14 negative), and tested against 26 samples (13 positive, 13 negative). The initial training of the model involved training a conventional neural network in what constituted a normal study and changes on the x-rays compatible with coronavirus infection. The weightings were then modified, and the model was executed again. The training samples were in batch sizes of 8 and underwent 25 epochs of training. The results trended towards an 85.71% true positive/true negative detection rate and an area under the curve trending towards 0.95, indicating approximately 95% accuracy in detecting changes on chest X-rays compatible with coronavirus infection. Study limitations include access to only a small dataset and no specificity in the diagnosis. Following a discussion with our programmer, there are areas where modifications in the weighting of the algorithm can be made in order to improve the detection rates. Given the high detection rate of the program, and the potential ease of implementation, this would be effective in assisting staff that is not trained in radiology in detecting otherwise subtle changes that might not be appreciated on imaging. Limitations include the lack of a differential diagnosis and application of the appropriate clinical history, although this may be less of a problem in day-to-day clinical practice. It is nonetheless our belief that implementing this program and widening its scope to detecting multiple pathologies such as lung masses will greatly assist both the radiology department and our colleagues in increasing workflow and detection rate.

Keywords: artificial intelligence, COVID, neural network, machine learning

Procedia PDF Downloads 66

Authors: Vibha Devi, Shabina Khanam

Abstract:

Hemp (Cannabis sativa) possesses a rich content of ω-6 linoleic and ω-3 linolenic essential fatty acid in the ratio of 3:1, which is a rare and most desired ratio that enhances the quality of hemp oil. These components are beneficial for the development of cell and body growth, strengthen the immune system, possess anti-inflammatory action, lowering the risk of heart problem owing to its anti-clotting property and a remedy for arthritis and various disorders. The present study employs supercritical fluid extraction (SFE) approach on hemp seed at various conditions of parameters; temperature (40 - 80) °C, pressure (200 - 350) bar, flow rate (5 - 15) g/min, particle size (0.430 - 1.015) mm and amount of co-solvent (0 - 10) % of solvent flow rate through central composite design (CCD). CCD suggested 32 sets of experiments, which was carried out. As SFE process includes large number of variables, the present study recommends the application of resampling techniques for cross-validation of the obtained data. Cross-validation refits the model on each data to achieve the information regarding the error, variability, deviation etc. Bootstrap and jackknife are the most popular resampling techniques, which create a large number of data through resampling from the original dataset and analyze these data to check the validity of the obtained data. Jackknife resampling is based on the eliminating one observation from the original sample of size N without replacement. For jackknife resampling, the sample size is 31 (eliminating one observation), which is repeated by 32 times. Bootstrap is the frequently used statistical approach for estimating the sampling distribution of an estimator by resampling with replacement from the original sample. For bootstrap resampling, the sample size is 32, which was repeated by 100 times. Estimands for these resampling techniques are considered as mean, standard deviation, variation coefficient and standard error of the mean. For ω-6 linoleic acid concentration, mean value was approx. 58.5 for both resampling methods, which is the average (central value) of the sample mean of all data points. Similarly, for ω-3 linoleic acid concentration, mean was observed as 22.5 through both resampling. Variance exhibits the spread out of the data from its mean. Greater value of variance exhibits the large range of output data, which is 18 for ω-6 linoleic acid (ranging from 48.85 to 63.66 %) and 6 for ω-3 linoleic acid (ranging from 16.71 to 26.2 %). Further, low value of standard deviation (approx. 1 %), low standard error of the mean (< 0.8) and low variance coefficient (< 0.2) reflect the accuracy of the sample for prediction. All the estimator value of variance coefficients, standard deviation and standard error of the mean are found within the 95 % of confidence interval.

Keywords: resampling, supercritical fluid extraction, hemp oil, cross-validation

Procedia PDF Downloads 121

Authors: B. Salehi, D. Andrews, I. Chaer, A. Gillich, A. Chalk, D. Bush

Abstract:

With the rapid increase of energy consumption in buildings in recent years, especially with the rise in population and growing economies, the importance of energy savings in buildings becomes more critical. One of the key factors in ensuring energy consumption is controlled and kept at a minimum is to utilise building energy modelling at the very early stages of the design. So, building modelling and simulation is a growing discipline. During the design phase of construction, modelling software can be used to estimate a building’s projected energy consumption, as well as building performance. The growth in the use of building modelling software packages opens the door for improvements in the design and also in the modelling itself by introducing novel methods such as building information modelling-based software packages which promote conventional building energy modelling into the digital building design process. To understand the most effective implementation tools, research projects undertaken should include elements of real-world experiments and not just rely on theoretical and simulated approaches. Upon review of the related studies undertaken, it’s evident that they are mostly based on modelling and simulation, which can be due to various reasons such as the more expensive and time-consuming nature of real-time data-based studies. Taking in to account the recent rise of building energy software modelling packages and the increasing number of studies utilising these methods in their projects and research, the accuracy and reliability of these modelling software packages has become even more crucial and critical. This Energy Performance Gap refers to the discrepancy between the predicted energy savings and the realised actual savings, especially after buildings implement energy-efficient technologies. There are many different software packages available which are either free or have commercial versions. In this study, IES VE (Integrated Environmental Solutions Virtual Environment) is used as it is a common Building Energy Modeling and Simulation software in the UK. This paper describes a study that compares real time results with those in a virtual model to illustrate this gap. The subject of the study is a north west facing north-west (345°) facing, naturally ventilated, conservatory within a domestic building in London is monitored during summer to capture real-time data. Then these results are compared to the virtual results of IES VE, which is a commonly used building energy modelling and simulation software in the UK. In this project, the effect of the wrong position of blinds on overheating is studied as well as providing new evidence of Performance Gap. Furthermore, the challenges of drawing the input of solar shading products in IES VE will be considered.

Keywords: building energy modelling and simulation, integrated environmental solutions virtual environment, IES VE, performance gap, real time data, solar shading products

Procedia PDF Downloads 108