Search results for: fire prediction

Authors: Seyed Mohammad Mahdi Zamani, Kamran Behdinan

Abstract:

In this study, a multiscale framework is performed to model the strength of Nextel fibers in presence of an atomistic scale pre-crack at finite temperatures. The bridging cell method (BCM) is the multiscale technique applied in this study, which decomposes the system into the atomistic, bridging and continuum domains; solves the whole system in a finite element framework; and incorporates temperature dependent calculations. Since Nextel is known to be structurally stable and retain 70% of its initial strength up to 1100°C; simulations are conducted at both of the room temperatures, 25°C, and fire temperatures, 1200°C. Two cases are modeled for a pre-crack present in either phases of alumina or mullite of the Nextel structure. The materials’ response is studied with respect to deformation behavior and ultimate tensile strength. Results show different crack growth trends for the two cases, and as the temperature increases, the crack growth resistance and material’s strength decrease.

Keywords: Nextel fibers, multiscale modeling, pre-crack, ultimate tensile strength

Procedia PDF Downloads 419

Authors: Zouari Ezzeddine, Ghraieb Ikram

Abstract:

From the results of the academic literature which evokes the limitations of previous studies, this article shows the reasons for multidimensionality Prediction of financial bubbles. A new framework for modeling study predicting financial bubbles by linking a set of variable presented on several dimensions dictating its multidimensional character. It takes into account the preferences of financial actors. A multicriteria anticipation of the appearance of bubbles in international financial markets helps to fight against a possible crisis.

Keywords: classical measures, predictions, financial bubbles, multidimensional, artificial neural networks

Procedia PDF Downloads 578

Authors: Fatemeh Faramarzi, Hosein Mahjoob

Abstract:

Building dams on rivers for utilization of water resources causes problems in hydrodynamic equilibrium and results in leaving all or part of the sediments carried by water in dam reservoir. This phenomenon has also significant impacts on water and sediment flow regime and in the long term can cause morphological changes in the environment surrounding the river, reducing the useful life of the reservoir which threatens sustainable development through inefficient management of water resources. In the past, empirical methods were used to predict the sedimentation amount in dam reservoirs and to estimate its efficient lifetime. But recently the mathematical and computational models are widely used in sedimentation studies in dam reservoirs as a suitable tool. These models usually solve the equations using finite element method. This study compares the results from tow software packages, GSTARS4 & HEC-6, in the prediction of the sedimentation amount in Dez dam, southern Iran. The model provides a one-dimensional, steady-state simulation of sediment deposition and erosion by solving the equations of momentum, flow and sediment continuity and sediment transport. GSTARS4 (Generalized Sediment Transport Model for Alluvial River Simulation) which is based on a one-dimensional mathematical model that simulates bed changes in both longitudinal and transverse directions by using flow tubes in a quasi-two-dimensional scheme to calibrate a period of 47 years and forecast the next 47 years of sedimentation in Dez Dam, Southern Iran. This dam is among the highest dams all over the world (with its 203 m height), and irrigates more than 125000 square hectares of downstream lands and plays a major role in flood control in the region. The input data including geometry, hydraulic and sedimentary data, starts from 1955 to 2003 on a daily basis. To predict future river discharge, in this research, the time series data were assumed to be repeated after 47 years. Finally, the obtained result was very satisfactory in the delta region so that the output from GSTARS4 was almost identical to the hydrographic profile in 2003. In the Dez dam due to the long (65 km) and a large tank, the vertical currents are dominant causing the calculations by the above-mentioned method to be inaccurate. To solve this problem, we used the empirical reduction method to calculate the sedimentation in the downstream area which led to very good answers. Thus, we demonstrated that by combining these two methods a very suitable model for sedimentation in Dez dam for the study period can be obtained. The present study demonstrated successfully that the outputs of both methods are the same.

Keywords: Dez Dam, prediction, sedimentation, water resources, computational models, finite element method, GSTARS4, HEC-6

Procedia PDF Downloads 313

Authors: Chongyang Ye, Rong Liu

Abstract:

Elastic compression stockings (ECSs) have been widely applied in prophylaxis and treatment of chronic venous insufficiency of lower extremities. The medical function of ECS is to improve venous return and increase muscular pumping action to facilitate blood circulation, which is largely determined by the complex interaction between the ECS and lower limb tissues. Understanding the mechanical transmission of ECS along the skin surface, deeper tissues, and vascular system is essential to assess the effectiveness of the ECSs. In this study, a three-dimensional (3D) finite element (FE) model of the leg-ECS system integrated with a 3D fluid-solid interaction (FSI) model of the leg-vein system was constructed to analyze the biomechanical properties of veins and soft tissues under different ECS compression. The Magnetic Resonance Imaging (MRI) of the human leg was divided into three regions, including soft tissues, bones (tibia and fibula) and veins (peroneal vein, great saphenous vein, and small saphenous vein). The ECSs with pressure ranges from 15 to 26 mmHg (Classes I and II) were adopted in the developed FE-FSI model. The soft tissue was assumed as a Neo-Hookean hyperelastic model with the fixed bones, and the ECSs were regarded as an orthotropic elastic shell. The interfacial pressure and stress transmission were simulated by the FE model, and venous hemodynamics properties were simulated by the FSI model. The experimental validation indicated that the simulated interfacial pressure distributions were in accordance with the pressure measurement results. The developed model can be used to predict interfacial pressure, stress transmission, and venous hemodynamics exerted by ECSs and optimize the structure and materials properties of ECSs design, thus improving the efficiency of compression therapy.

Keywords: elastic compression stockings, fluid-solid interaction, tissue and vein properties, prediction

Procedia PDF Downloads 112

Authors: Lenka Bodnárová, Rudolf Hela, Michala Hubertová, Iveta Nováková

Abstract:

This paper is concerning the issues of behaviour of lightweight expanded clay aggregates concrete exposed to high temperature. Lightweight aggregates from expanded clay are produced by firing of row material up to temperature 1050°C. Lightweight aggregates have suitable properties in terms of volume stability, when exposed to temperatures up to 1050°C, which could indicate their suitability for construction applications with higher risk of fire. The test samples were exposed to heat by using the standard temperature-time curve ISO 834. Negative changes in resulting mechanical properties, such as compressive strength, tensile strength, and flexural strength were evaluated. Also visual evaluation of the specimen was performed. On specimen exposed to excessive heat, an explosive spalling could be observed, due to evaporation of considerable amount of unbounded water from the inner structure of the concrete.

Keywords: expanded clay aggregate, explosive spalling, high temperature, lightweight concrete, temperature-time curve ISO 834

Procedia PDF Downloads 448

Authors: Ana Neves, John Leander, Ignacio Gonzalez, Raid Karoumi

Abstract:

Systematic monitoring and inspection are needed to assess the present state of a structure and predict its future condition. If an irregularity is noticed, repair actions may take place and the adequate intervention will most probably reduce the future costs with maintenance, minimize downtime and increase safety by avoiding the failure of the structure as a whole or of one of its structural parts. For this to be possible decisions must be made at the right time, which implies using systems that can detect abnormalities in their early stage. In this sense, Structural Health Monitoring (SHM) is seen as an effective tool for improving the safety and reliability of infrastructures. This paper explores the decision-making problem in SHM regarding the maintenance of civil engineering structures. The aim is to assess the present condition of a bridge based exclusively on measurements using the suggested method in this paper, such that action is taken coherently with the information made available by the monitoring system. Artificial Neural Networks are trained and their ability to predict structural behavior is evaluated in the light of a case study where acceleration measurements are acquired from a bridge located in Stockholm, Sweden. This relatively old bridge is presently still in operation despite experiencing obvious problems already reported in previous inspections. The prediction errors provide a measure of the accuracy of the algorithm and are subjected to further investigation, which comprises concepts like clustering analysis and statistical hypothesis testing. These enable to interpret the obtained prediction errors, draw conclusions about the state of the structure and thus support decision making regarding its maintenance.

Keywords: artificial neural networks, clustering analysis, model-free damage detection, statistical hypothesis testing, structural health monitoring

Procedia PDF Downloads 209

Authors: Ankita Jadon, Sidi Souvi, Nathalie Girault, Denis Petitprez

Abstract:

Safety requirements for Generation IV nuclear reactor designs, especially the new generation sodium-cooled fast reactors (SFR) require a risk-informed approach to model severe accidents (SA) and their consequences in case of outside release. In SFRs, aerosols are produced during a core disruptive accident when primary system sodium is ejected into the containment and burn in contact with the air; producing sodium aerosols. One of the key aspects of safety evaluation is the in-containment sodium aerosol behavior and their interaction with fission products. The study of the effects of sodium fires is essential for safety evaluation as the fire can both thermally damage the containment vessel and cause an overpressurization risk. Besides, during the fire, airborne fission product first dissolved in the primary sodium can be aerosolized or, as it can be the case for fission products, released under the gaseous form. The objective of this work is to study the interactions between sodium aerosols and fission products (Iodine, toxic and volatile, being the primary concern). Sodium fires resulting from an SA would produce aerosols consisting of sodium peroxides, hydroxides, carbonates, and bicarbonates. In addition to being toxic (in oxide form), this aerosol will then become radioactive. If such aerosols are leaked into the environment, they can pose a danger to the ecosystem. Depending on the chemical affinity of these chemical forms with fission products, the radiological consequences of an SA leading to containment leak tightness loss will also be affected. This work is split into two phases. Firstly, a method to theoretically understand the kinetics and thermodynamics of the heterogeneous reaction between sodium aerosols and fission products: I2 and HI are proposed. Ab-initio, density functional theory (DFT) calculations using Vienna ab-initio simulation package are carried out to develop an understanding of the surfaces of sodium carbonate (Na2CO3) aerosols and hence provide insight on its affinity towards iodine species. A comprehensive study of I2 and HI adsorption, as well as bicarbonate formation on the calculated lowest energy surface of Na2CO3, was performed which provided adsorption energies and description of the optimized configuration of adsorbate on the stable surface. Secondly, the heterogeneous reaction between (I2)g and Na2CO3 aerosols were investigated experimentally. To study this, (I2)g was generated by heating a permeation tube containing solid I2, and, passing it through a reaction chamber containing Na2CO3 aerosol deposit. The concentration of iodine was then measured at the exit of the reaction chamber. Preliminary observations indicate that there is an effective uptake of (I2)g on Na2CO3 surface, as suggested by our theoretical chemistry calculations. This work is the first step in addressing the gaps in knowledge of in-containment and atmospheric source term which are essential aspects of safety evaluation of SFR SA. In particular, this study is aimed to determine and characterize the radiological and chemical source term. These results will then provide useful insights for the developments of new models to be implemented in integrated computer simulation tool to analyze and evaluate SFR safety designs.

Keywords: iodine adsorption, sodium aerosols, sodium cooled reactor, DFT calculations, sodium carbonate

Procedia PDF Downloads 215

Authors: Chandu Rathnayake, Isuri Anuradha

Abstract:

Early detection and accurate diagnosis of lung diseases play a crucial role in improving patient prognosis. However, conventional diagnostic methods heavily rely on subjective symptom assessments and medical imaging, often causing delays in diagnosis and treatment. To overcome this challenge, we propose a novel lung disease prediction system that integrates patient symptoms and X-ray images to provide a comprehensive and reliable diagnosis.In this project, develop a mobile application specifically designed for detecting lung diseases. Our application leverages both patient symptoms and X-ray images to facilitate diagnosis. By combining these two sources of information, our application delivers a more accurate and comprehensive assessment of the patient's condition, minimizing the risk of misdiagnosis. Our primary aim is to create a user-friendly and accessible tool, particularly important given the current circumstances where many patients face limitations in visiting healthcare facilities. To achieve this, we employ several state-of-the-art algorithms. Firstly, the Decision Tree algorithm is utilized for efficient symptom-based classification. It analyzes patient symptoms and creates a tree-like model to predict the presence of specific lung diseases. Secondly, we employ the Random Forest algorithm, which enhances predictive power by aggregating multiple decision trees. This ensemble technique improves the accuracy and robustness of the diagnosis. Furthermore, we incorporate a deep learning model using Convolutional Neural Network (CNN) with the RestNet50 pre-trained model. CNNs are well-suited for image analysis and feature extraction. By training CNN on a large dataset of X-ray images, it learns to identify patterns and features indicative of lung diseases. The RestNet50 architecture, known for its excellent performance in image recognition tasks, enhances the efficiency and accuracy of our deep learning model. By combining the outputs of the decision tree-based algorithms and the deep learning model, our mobile application generates a comprehensive lung disease prediction. The application provides users with an intuitive interface to input their symptoms and upload X-ray images for analysis. The prediction generated by the system offers valuable insights into the likelihood of various lung diseases, enabling individuals to take appropriate actions and seek timely medical attention. Our proposed mobile application has significant potential to address the rising prevalence of lung diseases, particularly among young individuals with smoking addictions. By providing a quick and user-friendly approach to assessing lung health, our application empowers individuals to monitor their well-being conveniently. This solution also offers immense value in the context of limited access to healthcare facilities, enabling timely detection and intervention. In conclusion, our research presents a comprehensive lung disease prediction system that combines patient symptoms and X-ray images using advanced algorithms. By developing a mobile application, we provide an accessible tool for individuals to assess their lung health conveniently. This solution has the potential to make a significant impact on the early detection and management of lung diseases, benefiting both patients and healthcare providers.

Keywords: CNN, random forest, decision tree, machine learning, deep learning

Procedia PDF Downloads 73

Authors: Ning Li, Jiacheng Zhang, Zheng Yang, Sylvia Kang

Abstract:

Introduction: Polycystic ovary syndrome (PCOS) is a common physiological symptom in women of reproductive age. Women with PCOS may have infrequent or prolonged menstrual periods and excess male hormone (androgen) levels. Mira analyzes the cycle profiles and the luteinizing hormone (LH) changes in urine, closely related to the fertility level of healthy women and PCOS women. From the difference between the two groups, Mira helps to understand the physiological state of PCOS women and their hormonal changes in the menstrual cycle. Methods: In this study, data from 1496 cycles and information from 342 women belonging to two groups (181 PCOS and 161 Healthy) were collected and analyzed. Women test their luteinizing hormone (LH) in urine daily with Mira fertility test wand and Mira analyzer, from the day after the menstruation to the starting day of the next menstruation. All the collected data meets Mira’s user agreement and users’ identification was removed. The cycle length, LH peak, and other cycle information of the PCOS group were compared with the Healthy group. Results: The average cycle length of PCOS women is 41 days and of the Healthy women is 33 days. 91.4% of cycle length is within 40 days for the Healthy group, while it decreases to 71.9% for the PCOS group. This means PCOS women have a longer menstrual cycle and more variation during the cycle. With more variation, the ovulation prediction becomes more difficult for the PCOS group. The deviation between the LH surge day and the predicted ovulation day, calculated by the starting day of the next menstruation minus 14 days, is greater in the PCOS group compared with the Healthy group. Also, 46.96% of PCOS women have an irregular cycle, and only 19.25% of healthy women show an irregular cycle. Conclusion: PCOS women have longer menstrual cycles and more variation during the menstrual cycles. The traditional ovulation prediction is not suitable for PCOS women.

Keywords: menstrual cycle, PCOS, urinary luteinizing hormone, Mira

Procedia PDF Downloads 180

Authors: Paul Rowley, Adam Thirkill, Nick Doylend, Philip Leicester, Becky Gough

Abstract:

The dynamic performance of all energy generation technologies is impacted to varying degrees by the stochastic properties of the wider system within which the generation technology is located. This stochasticity can include the varying nature of ambient renewable energy resources such as wind or solar radiation, or unpredicted changes in energy demand which impact upon the operational behaviour of thermal generation technologies. An understanding of these stochastic impacts are especially important in contexts such as highly distributed (or embedded) generation, where an understanding of issues affecting the individual or aggregated performance of high numbers of relatively small generators is especially important, such as in ESCO projects. Probabilistic evaluation of monitored or simulated performance data is one technique which can provide an insight into the dynamic performance characteristics of generating systems, both in a prognostic sense (such as the prediction of future performance at the project’s design stage) as well as in a diagnostic sense (such as in the real-time analysis of underperforming systems). In this work, we describe the development, application and outcomes of a new approach to the acquisition of datasets suitable for use in the subsequent performance and impact analysis (including the use of Bayesian approaches) for a number of distributed generation technologies. The application of the approach is illustrated using a number of case studies involving domestic and small commercial scale photovoltaic, solar thermal and natural gas boiler installations, and the results as presented show that the methodology offers significant advantages in terms of plant efficiency prediction or diagnosis, along with allied environmental and social impacts such as greenhouse gas emission reduction or fuel affordability.

Keywords: renewable energy, dynamic performance simulation, Bayesian analysis, distributed generation

Procedia PDF Downloads 495

Authors: Prashant Verma, Prafulla K. Swain, K. K. Singh, Mukti Khetan

Abstract:

Objective: In India, around 20,000 women die every year due to abortion-related complications. In the modelling of count variables, there is sometimes a preponderance of zero counts. This article concerns the estimation of various count regression models to predict the average number of spontaneous abortion among women in the Punjab state of India. It also assesses the factors associated with the number of spontaneous abortions. Materials and methods: The study included 27,173 married women of Punjab obtained from the DLHS-4 survey (2012-13). Poisson regression (PR), Negative binomial (NB) regression, zero hurdle negative binomial (ZHNB), and zero-inflated negative binomial (ZINB) models were employed to predict the average number of spontaneous abortions and to identify the determinants affecting the number of spontaneous abortions. Results: Statistical comparisons among four estimation methods revealed that the ZINB model provides the best prediction for the number of spontaneous abortions. Antenatal care (ANC) place, place of residence, total children born to a woman, woman's education and economic status were found to be the most significant factors affecting the occurrence of spontaneous abortion. Conclusions: The study offers a practical demonstration of techniques designed to handle count variables. Statistical comparisons among four estimation models revealed that the ZINB model provided the best prediction for the number of spontaneous abortions and is recommended to be used to predict the number of spontaneous abortions. The study suggests that women receive institutional Antenatal care to attain limited parity. It also advocates promoting higher education among women in Punjab, India.

Keywords: count data, spontaneous abortion, Poisson model, negative binomial model, zero hurdle negative binomial, zero-inflated negative binomial, regression

Procedia PDF Downloads 155

Authors: Aschalew Cherie Workneh, K. S. Hari Prasad, Chandra Shekhar Prasad Ojha

Abstract:

Due to the increase in water scarcity, the crop water stress index (CWSI) is receiving significant attention these days, especially in arid and semiarid regions, for quantifying water stress and effective irrigation scheduling. Nowadays, machine learning techniques such as neural networks are being widely used to determine CWSI. In the present study, the performance of two artificial neural networks, namely, Self-Organizing Maps (SOM) and Feed Forward-Back Propagation Artificial Neural Networks (FF-BP-ANN), are compared while determining the CWSI of rice crop. Irrigation field experiments with varying degrees of irrigation were conducted at the irrigation field laboratory of the Indian Institute of Technology, Roorkee, during the growing season of the rice crop. The CWSI of rice was computed empirically by measuring key meteorological variables (relative humidity, air temperature, wind speed, and canopy temperature) and crop parameters (crop height and root depth). The empirically computed CWSI was compared with SOM and FF-BP-ANN predicted CWSI. The upper and lower CWSI baselines are computed using multiple regression analysis. The regression analysis showed that the lower CWSI baseline for rice is a function of crop height (h), air vapor pressure deficit (AVPD), and wind speed (u), whereas the upper CWSI baseline is a function of crop height (h) and wind speed (u). The performance of SOM and FF-BP-ANN were compared by computing Nash-Sutcliffe efficiency (NSE), index of agreement (d), root mean squared error (RMSE), and coefficient of correlation (R²). It is found that FF-BP-ANN performs better than SOM while predicting the CWSI of rice crops.

Keywords: artificial neural networks; crop water stress index; canopy temperature, prediction capability

Procedia PDF Downloads 117

Authors: Rashid S. Mohammad, Shicheng Zhang, Xinzhe Zhao

Abstract:

In tight oil reservoirs, wettability alteration has generally been considered as an effective way to remove fracturing fluid retention on the surface of the fracture and consequently improved oil production. However, there is a lack of a reliable productivity prediction model to show the relationship between the wettability and oil production in tight oil well. In this paper, a new oil productivity prediction model of immiscible oil-water flow and miscible CO₂-oil flow accounting for wettability is developed. This mathematical model is established by considering two different length scales: nonporous network and propped fractures. CO₂ flow diffuses in the nonporous network and high velocity non-Darcy flow in propped fractures are considered by taking into account the effect of wettability alteration on capillary pressure and relative permeability. A laboratory experiment is also conducted here to validate this model. Laboratory experiments have been designed to compare the water saturation profiles for different contact angle, revealing the fluid retention in rock pores that affects capillary force and relative permeability. Four kinds of brines with different concentrations are selected here to create different contact angles. In water-wet porous media, as the system becomes more oil-wet, water saturation decreases. As a result, oil relative permeability increases. On the other hand, capillary pressure which is the resistance for the oil flow increases as well. The oil production change due to wettability alteration is the result of the comprehensive changes of oil relative permeability and capillary pressure. The results indicate that wettability is a key factor for fracturing fluid retention removal and oil enhancement in tight reservoirs. By incorporating laboratory test into a mathematical model, this work shows the relationship between wettability and oil production is not a simple linear pattern but a parabolic one. Additionally, it can be used for a better understanding of optimization design of fracturing fluids.

Keywords: wettability, relative permeability, fluid retention, oil production, unconventional and tight reservoirs

Procedia PDF Downloads 236

Authors: Ginger Egberts, Marianne Schaaphok, Fred Vermolen, Paul van Zuijlen

Abstract:

A few years ago, a promising morphoelastic model was developed for the simulation of contraction formation after burn injuries. Contraction can lead to a serious reduction in physical mobility, like a reduction in the range-of-motion of joints. If this is the case in a healing burn wound, then this is referred to as a contracture that needs medical intervention. The morphoelastic model consists of a set of partial differential equations describing both a chemical part and a mechanical part in dermal wound healing. These equations are solved with the numerical finite element method (FEM). In this method, many calculations are required on each of the chosen elements. In general, the more elements, the more accurate the solution. However, the number of elements increases rapidly if simulations are performed in 2D and 3D. In that case, it not only takes longer before a prediction is available, the computation also becomes more expensive. It is therefore important to investigate alternative possibilities to generate the same results, based on the input parameters only. In this study, a surrogate neural network has been designed to mimic the results of the one-dimensional morphoelastic model. The neural network generates predictions quickly, is easy to implement, and there is freedom in the choice of input and output. Because a neural network requires extensive training and a data set, it is ideal that the one-dimensional FEM code generates output quickly. These feed-forward-type neural network results are very promising. Not only can the network give faster predictions, but it also has a performance of over 99%. It reports on the relative surface area of the wound/scar, the total strain energy density, and the evolutions of the densities of the chemicals and mechanics. It is, therefore, interesting to investigate the applicability of a neural network for the two- and three-dimensional morphoelastic model for contraction after burn injuries.

Keywords: biomechanics, burns, feasibility, feed-forward NN, morphoelasticity, neural network, relative surface area wound

Procedia PDF Downloads 55

Authors: Parul Jindal, Shubhi Singh, Priya Ramakrishnan, Shailender Raghuvanshi

Abstract:

Background: Knowledge of the influence of the age of the child on laryngeal dimensions is essential for all practitioners who are dealing with paediatric airway. Choosing the correct endotracheal tube (ETT) size is a crucial step in pediatric patients because a large-sized tube may cause complications like post-extubation stridor and subglottic stenosis. On the other hand with a smaller tube, there will be increased gas flow resistance, aspiration risk, poor ventilation, inaccurate monitoring of end-tidal gases and reintubation may also be required with a different size of the tracheal tube. Recent advancement in ultrasonography (USG) techniques should now allow for accurate and descriptive evaluation of pediatric airway. Aims and objectives: This study was planned to determine the accuracy of Ultrasonography (USG) to assess the appropriate ETT size and compare it with physical indices based formulae. Methods: After obtaining approval from Institute’s Ethical and Research committee, and parental written and informed consent, the study was conducted on 100 subjects of either sex between 12-60 months of age, undergoing various elective surgeries under general anesthesia requiring endotracheal intubation. The same experienced radiologist performed ultrasonography. The transverse diameter was measured at the level of cricoids cartilage by USG. After USG, general anesthesia was administered using standard techniques followed by the institute. An experienced anesthesiologist performed the endotracheal intubations with uncuffed endotracheal tube (Portex Tracheal Tube Smiths Medical India Pvt. Ltd.) with Murphy’s eye. He was unaware of the finding of the ultrasonography. The tracheal tube was considered best fit if air leak was satisfactory at 15-20 cm H₂O of airway pressure. The obtained values were compared with the values of endotracheal tube size calculated by ultrasonography, various age, height, weight-based formulas and diameter of right and left little finger. The correlation of the size of the endotracheal tube by different modalities was done and Pearson's correlation coefficient was obtained. The comparison of the mean size of the endotracheal tube by ultrasonography and by traditional formula was done by the Friedman’s test and Wilcoxon sign-rank test. Results: The predicted tube size was equal to best fit and best determined by ultrasonography (100%) followed by comparison to left little finger (98%) and right little finger (97%) and age-based formula (95%) followed by multivariate formula (83%) and body length (81%) formula. According to Pearson`s correlation, there was a moderate correlation of best fit endotracheal tube with endotracheal tube size by age-based formula (r=0.743), body length based formula (r=0.683), right little finger based formula (r=0.587), left little finger based formula (r=0.587) and multivariate formula (r=0.741). There was a strong correlation with ultrasonography (r=0.943). Ultrasonography was the most sensitive (100%) method of prediction followed by comparison to left (98%) and right (97%) little finger and age-based formula (95%), the multivariate formula had an even lesser sensitivity (83%) whereas body length based formula was least sensitive with a sensitivity of 78%. Conclusion: USG is a reliable method of estimation of subglottic diameter and for prediction of ETT size in children.

Keywords: endotracheal intubation, pediatric airway, subglottic diameter, traditional formulas, ultrasonography

Procedia PDF Downloads 240

Authors: K. Sangeeth, P. Rekha, P. Preeja, P. Divya, R. Arya, R. Maneesha

Abstract:

Wireless sensor networks (WSN) are small or tiny devices that consists of different sensors to sense physical parameters like air pressure, temperature, vibrations, movement etc., process these data and sends it to the central data center to take decisions. The WSN domain, has wide range of applications such as monitoring and detecting natural hazards like landslides, forest fire, avalanche, flood monitoring and also in healthcare applications. With such different applications, it is being taught in undergraduate/post graduate level in many universities under department of computer science. But the cost and infrastructure required to purchase WSN nodes for having the students getting hands on expertise on these devices is expensive. This paper gives overview about the remote triggered lab that consists of more than 100 WSN nodes that helps the students to remotely login from anywhere in the world using the World Wide Web, configure the nodes and learn the WSN concepts in intuitive way. It proposes new way called over the air programming (OTAP) and its internals that program the 100 nodes simultaneously and view the results without the nodes being physical connected to the computer system, thereby allowing for sparse deployment.

Keywords: WSN, over the air programming, virtual lab, AT45DB

Procedia PDF Downloads 377

Authors: Maryam Kheirollahpour, Mahmoud Danaee, Amir Faisal Merican, Asma Ahmad Shariff

Abstract:

Background: The presence of nonlinearity among the risk factors of eating behavior causes a bias in the prediction models. The accuracy of estimation of eating behaviors risk factors in the primary prevention of obesity has been established. Objective: The aim of this study was to explore the potential of a hybrid model of structural equation modeling (SEM) and Artificial Neural Networks (ANN) to predict eating behaviors. Methods: The Partial Least Square-SEM (PLS-SEM) and a hybrid model (SEM-Artificial Neural Networks (SEM-ANN)) were applied to evaluate the factors affecting eating behavior patterns among university students. 340 university students participated in this study. The PLS-SEM analysis was used to check the effect of emotional eating scale (EES), body shape concern (BSC), and body appreciation scale (BAS) on different categories of eating behavior patterns (EBP). Then, the hybrid model was conducted using multilayer perceptron (MLP) with feedforward network topology. Moreover, Levenberg-Marquardt, which is a supervised learning model, was applied as a learning method for MLP training. The Tangent/sigmoid function was used for the input layer while the linear function applied for the output layer. The coefficient of determination (R²) and mean square error (MSE) was calculated. Results: It was proved that the hybrid model was superior to PLS-SEM methods. Using hybrid model, the optimal network happened at MPLP 3-17-8, while the R² of the model was increased by 27%, while, the MSE was decreased by 9.6%. Moreover, it was found that which one of these factors have significantly affected on healthy and unhealthy eating behavior patterns. The p-value was reported to be less than 0.01 for most of the paths. Conclusion/Importance: Thus, a hybrid approach could be suggested as a significant methodological contribution from a statistical standpoint, and it can be implemented as software to be able to predict models with the highest accuracy.

Keywords: hybrid model, structural equation modeling, artificial neural networks, eating behavior patterns

Procedia PDF Downloads 156

Authors: Danny Barash

Abstract:

Riboswitches are RNA genetic control elements that were originally discovered in bacteria and provide a unique mechanism of gene regulation. They work without the participation of proteins and are believed to represent ancient regulatory systems in the evolutionary timescale. One of the biggest challenges in riboswitch research is that many are found in prokaryotes but only a small percentage of known riboswitches have been found in certain eukaryotic organisms. The few examples of eukaryotic riboswitches were identified using sequence-based bioinformatics search methods that include some slight structural considerations. These pattern-matching methods were the first ones to be applied for the purpose of riboswitch detection and they can also be programmed very efficiently using a data structure called affix arrays, making them suitable for genome-wide searches of riboswitch patterns. However, they are limited by their ability to detect harder to find riboswitches that deviate from the known patterns. Several methods have been developed since then to tackle this problem. The most commonly used by practitioners is Infernal that relies on Hidden Markov Models (HMMs) and Covariance Models (CMs). Profile Hidden Markov Models were also carried out in the pHMM Riboswitch Scanner web application, independently from Infernal. Other computational approaches that have been developed include RMDetect by the use of 3D structural modules and RNAbor that utilizes Boltzmann probability of structural neighbors. We have tried to incorporate more sophisticated secondary structure considerations based on RNA folding prediction using several strategies. The first idea was to utilize window-based methods in conjunction with folding predictions by energy minimization. The moving window approach is heavily geared towards secondary structure consideration relative to sequence that is treated as a constraint. However, the method cannot be used genome-wide due to its high cost because each folding prediction by energy minimization in the moving window is computationally expensive, enabling to scan only at the vicinity of genes of interest. The second idea was to remedy the inefficiency of the previous approach by constructing a pipeline that consists of inverse RNA folding considering RNA secondary structure, followed by a BLAST search that is sequence-based and highly efficient. This approach, which relies on inverse RNA folding in general and our own in-house fragment-based inverse RNA folding program called RNAfbinv in particular, shows capability to find attractive candidates that are missed by Infernal and other standard methods being used for riboswitch detection. We demonstrate attractive candidates found by both the moving-window approach and the inverse RNA folding approach performed together with BLAST. We conclude that structure-based methods like the two strategies outlined above hold considerable promise in detecting riboswitches and other conserved RNAs of functional importance in a variety of organisms.

Keywords: riboswitches, RNA folding prediction, RNA structure, structure-based methods

Procedia PDF Downloads 234

Authors: Adel Khorramrouz, Monireh Ahmadi Bani, Ehsan Norouzi, Morvarid Lalenoor

Abstract:

Background: Plantar pressure distribution measurement has been used for a long time to assess foot disorders. Plantar pressure is an important component affecting the foot and ankle function and Changes in plantar pressure distribution could indicate various foot and ankle disorders. Morphologic and mechanical properties of the foot may be important factors affecting the plantar pressure distribution. Accurate and early measurement may help to reduce the prevalence of pes planus. With recent developments in technology, new techniques such as machine learning have been used to assist clinicians in predicting patients with foot disorders. Significance of the study: This study proposes a neural network learning-based flat foot classification methodology using static foot pressure distribution. Methodologies: Data were collected from 895 patients who were referred to a foot clinic due to foot disorders. Patients with pes planus were labeled by an experienced physician based on clinical examination. Then all subjects (with and without pes planus) were evaluated for static plantar pressures distribution. Patients who were diagnosed with the flat foot in both feet were included in the study. In the next step, the leg length was normalized and the network was trained for plantar pressure mapping images. Findings: From a total of 895 image data, 581 were labeled as pes planus. A computational neural network (CNN) ran to evaluate the performance of the proposed model. The prediction accuracy of the basic CNN-based model was performed and the prediction model was derived through the proposed methodology. In the basic CNN model, the training accuracy was 79.14%, and the test accuracy was 72.09%. Conclusion: This model can be easily and simply used by patients with pes planus and doctors to predict the classification of pes planus and prescreen for possible musculoskeletal disorders related to this condition. However, more models need to be considered and compared for higher accuracy.

Keywords: foot disorder, machine learning, neural network, pes planus

Procedia PDF Downloads 361

Authors: Junfeng Yan, Wenkui Hou

Abstract:

Prognostics and Health Management (PHM) is a technology to monitor the equipment status and predict impending faults. It is used to predict the potential fault and provide fault information and track trends of system degradation by capturing characteristics signals. So how to detect characteristics signals is very important. The select of test point plays a very important role in detecting characteristics signal. Traditionally, we use dependency model to select the test point containing the most detecting information. But, facing the large complicated system, the dependency model is not built so easily sometimes and the greater trouble is how to calculate the matrix. Rely on this premise, the paper provide a highly effective method to select test point without dependency model. Because signal flow model is a diagnosis model based on failure mode, which focuses on system’s failure mode and the dependency relationship between the test points and faults. In the signal flow model, a fault information can flow from the beginning to the end. According to the signal flow model, we can find out location and structure information of every test point and module. We break the signal flow model up into serial and parallel parts to obtain the final relationship function between the system’s testability or prediction metrics and test points. Further, through the partial derivatives operation, we can obtain every test point’s importance degree in determining the testability metrics, such as undetected rate, false alarm rate, untrusted rate. This contributes to installing the test point according to the real requirement and also provides a solid foundation for the Prognostics and Health Management. According to the real effect of the practical engineering application, the method is very efficient.

Keywords: false alarm rate, importance degree, signal flow model, undetected rate, untrusted rate

Procedia PDF Downloads 377

Authors: Shahrzad Zolfagharnassab, Abdul Rashid Mohamed Shariff, Siti Khairunniza Bejo

Abstract:

Conventional methods for nutrient soil mapping are based on laboratory tests of samples that are obtained from surveys. The time and cost involved in gathering and analyzing soil samples are the reasons that researchers use Predictive Soil Mapping (PSM). PSM can be defined as the development of a numerical or statistical model of the relationship among environmental variables and soil properties, which is then applied to a geographic database to create a predictive map. Kriging is a group of geostatistical techniques to spatially interpolate point values at an unobserved location from observations of values at nearby locations. The main problem with using kriging as an interpolator is that it is excessively data-dependent and requires a large number of closely spaced data points. Hence, there is a need to minimize the number of data points without sacrificing the accuracy of the results. In this paper, an Artificial Neural Networks (ANN) scheme was used to predict macronutrient values at un-sampled points. ANN has become a popular tool for prediction as it eliminates certain difficulties in soil property prediction, such as non-linear relationships and non-normality. Back-propagation multilayer feed-forward network structures were used to predict nitrogen, phosphorous and potassium values in the soil of the study area. A limited number of samples were used in the training, validation and testing phases of ANN (pattern reconstruction structures) to classify soil properties and the trained network was used for prediction. The soil analysis results of samples collected from the soil survey of block C of Sawah Sempadan, Tanjung Karang rice irrigation project at Selangor of Malaysia were used. Soil maps were produced by the Kriging method using 236 samples (or values) that were a combination of actual values (obtained from real samples) and virtual values (neural network predicted values). For each macronutrient element, three types of maps were generated with 118 actual and 118 virtual values, 59 actual and 177 virtual values, and 30 actual and 206 virtual values, respectively. To evaluate the performance of the proposed method, for each macronutrient element, a base map using 236 actual samples and test maps using 118, 59 and 30 actual samples respectively produced by the Kriging method. A set of parameters was defined to measure the similarity of the maps that were generated with the proposed method, termed the sample reduction method. The results show that the maps that were generated through the sample reduction method were more accurate than the corresponding base maps produced through a smaller number of real samples. For example, nitrogen maps that were produced from 118, 59 and 30 real samples have 78%, 62%, 41% similarity, respectively with the base map (236 samples) and the sample reduction method increased similarity to 87%, 77%, 71%, respectively. Hence, this method can reduce the number of real samples and substitute ANN predictive samples to achieve the specified level of accuracy.

Keywords: artificial neural network, kriging, macro nutrient, pattern recognition, precision farming, soil mapping

Procedia PDF Downloads 70

Authors: S. Mousavian, F. Mousavian, V. Nikkhah Rashidabad

Abstract:

Cubic equations of state like Redlich–Kwong (RK) EOS have been proved to be very reliable tools in the prediction of phase behavior. Despite their good performance in compositional calculations, they usually suffer from weaknesses in the predictions of saturated liquid density. In this research, RK equation was modified. The result of this study shows that modified equation has good agreement with experimental data.

Keywords: equation of state, modification, ammonia, genetic algorithm

Procedia PDF Downloads 382

Authors: Selim M. Khan, Aaron A. Goodarzi, Joshua M. Taron, Tryggve Rönnqvist

Abstract:

Indoor air quality, a prime determinant of health, is strongly influenced by the presence of hazardous radon gas within the built environment. As a health issue, dangerously high indoor radon arose within the 20th century to become the 2nd leading cause of lung cancer. While the 21st century building metrics and human behaviors have captured, contained, and concentrated radon to yet higher and more hazardous levels, the issue is rapidly worsening in Canada. It is established that Canadians in the Prairies are the 2nd highest radon-exposed population in the world, with 1 in 6 residences experiencing 0.2-6.5 millisieverts (mSv) radiation per week, whereas the Canadian Nuclear Safety Commission sets maximum 5-year occupational limits for atomic workplace exposure at only 20 mSv. This situation is also deteriorating over time within newer housing stocks containing higher levels of radon. Deep machine learning (LSTM) algorithms were applied to analyze multiple quantitative and qualitative features, determine the most important contributory factors, and predicted radon levels in the known past (1990-2020) and projected future (2021-2050). The findings showed gradual downwards patterns in Sweden, whereas it would continue to go from high to higher levels in Canada over time. The contributory factors found to be the basement porosity, roof insulation depthness, R-factor, and air dynamics of the indoor environment related to human window opening behaviour. Building codes must consider including these factors to ensure adequate indoor ventilation and healthy living that can prevent lung cancer in non-smokers.

Keywords: radon, building metrics, deep learning, LSTM prediction model, lung cancer, canada, sweden

Procedia PDF Downloads 112

Authors: Muhamad Aris Burhanudin, Angga Firmansyas, Bagus Jaya Santosa

Abstract:

Earthquakes are natural hazard that can trigger the most dangerous hazard, tsunami. 26 December 2004, a giant earthquake occurred in north-west Andalas Island. It made giant tsunami which crushed Sumatra, Bangladesh, India, Sri Lanka, Malaysia and Singapore. More than twenty thousand people dead. The occurrence of earthquake and tsunami can not be avoided. But this hazard can be mitigated by earthquake forecasting. Early preparation is the key factor to reduce its damages and consequences. We aim to investigate quantitatively on pattern of earthquake. Then, we can know the trend. We study about earthquake which has happened in Andalas island, Indonesia one last decade. Andalas is island which has high seismicity, more than a thousand event occur in a year. It is because Andalas island is in tectonic subduction zone of Hindia sea plate and Eurasia plate. A tsunami forecasting is needed to mitigation action. Thus, a Tsunami Forecasting Method is presented in this work. Neutral Network has used widely in many research to estimate earthquake and it is convinced that by using Backpropagation Method, earthquake can be predicted. At first, ANN is trained to predict Tsunami 26 December 2004 by using earthquake data before it. Then after we get trained ANN, we apply to predict the next earthquake. Not all earthquake will trigger Tsunami, there are some characteristics of earthquake that can cause Tsunami. Wrong decision can cause other problem in the society. Then, we need a method to reduce possibility of wrong decision. Fuzzy TOPSIS is a statistical method that is widely used to be decision seconder referring to given parameters. Fuzzy TOPSIS method can make the best decision whether it cause Tsunami or not. This work combines earthquake prediction using neural network method and using Fuzzy TOPSIS to determine the decision that the earthquake triggers Tsunami wave or not. Neural Network model is capable to capture non-linear relationship and Fuzzy TOPSIS is capable to determine the best decision better than other statistical method in tsunami prediction.

Keywords: earthquake, fuzzy TOPSIS, neural network, tsunami

Procedia PDF Downloads 496

Authors: Sanela Slavkovic, Congor Nad, Spela Golubovic

Abstract:

Background: The diagnosis of multiple sclerosis (MS) is a major life challenge and has repercussions on all aspects of the daily functioning of those attained by it – personal activities, social participation, and quality of life. Regular follow-up of only the neurological status is not informative enough so that it could provide data on the sort of support and rehabilitation that is required. Objective: The aim of this study was to establish the current level of functioning of persons attained by MS and the factors that influence it. Methods: The study was conducted in Serbia, on a sample of 108 persons with relapse-remitting form of MS, aged 20 to 53 (mean 39.86 years; SD 8.20 years). All participants were fully ambulatory. Methods applied in the study include Expanded Disability Status Scale-EDSS and World Health Organization Disability Assessment Schedule, WHODAS 2.0 (36-item version, self-administered). Results: Participants were found to experience the most problems in the domains of Participation, Mobility, Life activities and Cognition. The least difficulties were found in the domain of Self-care. Symptom duration was the only control variable with a significant partial contribution to the prediction of the WHODAS scale score (β=0.30, p < 0.05). The total EDSS score correlated with the total WHODAS 2.0 score (r=0.34, p=0.00). Statistically significant differences in the domain of EDSS 0-5.5 were found within categories (0-1.5; 2-3.5; 4-5.5). The more pronounced a participant’s EDSS score was, although not indicative of large changes in the neurological status, the more apparent the changes in the functional domain, i.e. in all areas covered by WHODAS 2.0. Pyramidal (β=0.34, p < 0.05) and Bowel and bladder (β=0.24, p < 0.05) functional systems were found to have a significant partial contribution to the prediction of the WHODAS score. Conclusion: Measuring functioning and disability is important in the follow-up of persons suffering from MS in order to plan rehabilitation and define areas in which additional support is needed.

Keywords: disability, functionality, multiple sclerosis, rehabilitation

Procedia PDF Downloads 122

Authors: E. A. Asabina, V. I. Pet'kov, P. A. Mayorov, A. V. Markin, N. N. Smirnova, A. M. Kovalskii, A. A. Usenko

Abstract:

The purpose of this work was to synthesize and investigate phase formation, structure and thermophysical properties of the phosphates M_0.5+xM'_xZr_2–x(PO₄)₃ (M – Cd, Sr, Pb; M' – Mg, Co, Mn). The compounds were synthesized by sol-gel method. The results showed formation of limited solid solutions of NZP/NASICON type. The crystal structures of triple phosphates of the compositions MMg_0.5Zr_1.5(PO₄)₃ were refined by the Rietveld method using XRD data. Heat capacity (8–660 K) of the phosphates Pb_0.5+xMg_xZr_2-x(PO₄)₃ (x = 0, 0.5) was measured, and reversible polymorphic transitions were found at temperatures, close to the room temperature. The results of Rietveld structure refinement showed the polymorphism caused by disordering of lead cations in the cavities of NZP/NASICON structure. Thermal expansion (298−1073 K) of the phosphates MMg_0.5Zr_1.5(PO₄)₃ was studied by XRD method, and the compounds were found to belong to middle and low-expanding materials. Thermal diffusivity (298–573 K) of the ceramic samples of phosphates slightly decreased with temperature increasing. As was demonstrated, the studied phosphates are characterized by the better thermophysical characteristics than widespread fire-resistant materials, such as zirconia and etc.

Keywords: NASICON, NZP, phosphate, structure, synthesis, thermophysical properties

Procedia PDF Downloads 141

Authors: Etienne Gaborit, Dorothy Durnford, Daniel Deacu, Marco Carrera, Nathalie Gauthier, Camille Garnaud, Vincent Fortin

Abstract:

A new experimental streamflow forecasting system was recently implemented at the Environment and Climate Change Canada’s (ECCC) Canadian Centre for Meteorological and Environmental Prediction (CCMEP). It relies on CaLDAS (Canadian Land Data Assimilation System) for the assimilation of surface variables, and on a surface prediction system that feeds a routing component. The surface energy and water budgets are simulated with the SVS (Soil, Vegetation, and Snow) Land-Surface Scheme (LSS) at 2.5-km grid spacing over Canada. The routing component is based on the Watroute routing scheme at 1-km grid spacing for the Great Lakes and Nelson River watersheds. The system is run in two distinct phases: an analysis part and a forecast part. During the analysis part, CaLDAS outputs are used to force the routing system, which performs streamflow assimilation. In forecast mode, the surface component is forced with the Canadian GEM atmospheric forecasts and is initialized with a CaLDAS analysis. Streamflow performances of this new system are presented over 2019. Performances are compared to the current ECCC’s operational streamflow forecasting system, which is different from the new experimental system in many aspects. These new streamflow forecasts are also compared to persistence. Overall, the new streamflow forecasting system presents promising results, highlighting the need for an elaborated assimilation phase before performing the forecasts. However, the system is still experimental and is continuously being improved. Some major recent improvements are presented here and include, for example, the assimilation of snow cover data from remote sensing, a backward propagation of assimilated flow observations, a new numerical scheme for the routing component, and a new reservoir model.

Keywords: assimilation system, distributed physical model, offline hydro-meteorological chain, short-term streamflow forecasts

Procedia PDF Downloads 130

Authors: Xiaomin Zhong, Alexander Pate, Ya-Ting Yang, Ali Fahmi, Darren M. Ashcroft, Ben Goldacre, Brian Mackenna, Amir Mehrkar, Sebastian C. J. Bacon, Jon Massey, Louis Fisher, Peter Inglesby, Kieran Hand, Tjeerd van Staa, Victoria Palin

Abstract:

Background: This study aimed to predict risks of potentially inappropriate antibiotic type and repeat prescribing and assess changes during COVID-19. Methods: With the approval of NHS England, we used the OpenSAFELY platform to access the TPP SystmOne electronic health record (EHR) system and selected patients prescribed antibiotics from 2019 to 2021. Multinomial logistic regression models predicted the patient’s probability of receiving an inappropriate antibiotic type or repeating the antibiotic course for each common infection. Findings: The population included 9.1 million patients with 29.2 million antibiotic prescriptions. 29.1% of prescriptions were identified as repeat prescribing. Those with same-day incident infection coded in the EHR had considerably lower rates of repeat prescribing (18.0%), and 8.6% had a potentially inappropriate type. No major changes in the rates of repeat antibiotic prescribing during COVID-19 were found. In the ten risk prediction models, good levels of calibration and moderate levels of discrimination were found. Important predictors included age, prior antibiotic prescribing, and region. Patients varied in their predicted risks. For sore throat, the range from 2.5 to 97.5th percentile was 2.7 to 23.5% (inappropriate type) and 6.0 to 27.2% (repeat prescription). For otitis externa, these numbers were 25.9 to 63.9% and 8.5 to 37.1%, respectively. Interpretation: Our study found no evidence of changes in the level of inappropriate or repeat antibiotic prescribing after the start of COVID-19. Repeat antibiotic prescribing was frequent and varied according to regional and patient characteristics. There is a need for treatment guidelines to be developed around antibiotic failure and clinicians provided with individualised patient information.

Keywords: antibiotics, infection, COVID-19 pandemic, antibiotic stewardship, primary care

Procedia PDF Downloads 120

Authors: Dongping Fang, Lian Duan, Xiaojing Yuan, Mike Xu, Allyn Klunder, Kevin Tan, Suiting Cao, Yeqing Ji

Abstract:

Accurate prediction of a medical condition with straight clinical evidence is a long-sought topic in the medical management and health insurance field. Although great progress has been made with machine learning algorithms, the medical community is still, to a certain degree, suspicious about the model's accuracy and interpretability. This paper presents an innovative hierarchical attention deep learning model to achieve good prediction and clear interpretability that can be easily understood by medical professionals. This deep learning model uses a hierarchical attention structure that matches naturally with the medical history data structure and reflects the member’s encounter (date of service) sequence. The model attention structure consists of 3 levels: (1) attention on the medical code types (diagnosis codes, procedure codes, lab test results, and prescription drugs), (2) attention on the sequential medical encounters within a type, (3) attention on the medical codes within an encounter and type. This model is applied to predict the occurrence of stage 3 chronic kidney disease (CKD3), using three years’ medical history of Medicare Advantage (MA) members from a top health insurance company. The model takes members’ medical events, both claims and electronic medical record (EMR) data, as input, makes a prediction of CKD3 and calculates the contribution from individual events to the predicted outcome. The model outcome can be easily explained with the clinical evidence identified by the model algorithm. Here are examples: Member A had 36 medical encounters in the past three years: multiple office visits, lab tests and medications. The model predicts member A has a high risk of CKD3 with the following well-contributed clinical events - multiple high ‘Creatinine in Serum or Plasma’ tests and multiple low kidneys functioning ‘Glomerular filtration rate’ tests. Among the abnormal lab tests, more recent results contributed more to the prediction. The model also indicates regular office visits, no abnormal findings of medical examinations, and taking proper medications decreased the CKD3 risk. Member B had 104 medical encounters in the past 3 years and was predicted to have a low risk of CKD3, because the model didn’t identify diagnoses, procedures, or medications related to kidney disease, and many lab test results, including ‘Glomerular filtration rate’ were within the normal range. The model accurately predicts members A and B and provides interpretable clinical evidence that is validated by clinicians. Without extra effort, the interpretation is generated directly from the model and presented together with the occurrence date. Our model uses the medical data in its most raw format without any further data aggregation, transformation, or mapping. This greatly simplifies the data preparation process, mitigates the chance for error and eliminates post-modeling work needed for traditional model explanation. To our knowledge, this is the first paper on an interpretable deep-learning model using a 3-level attention structure, sourcing both EMR and claim data, including all 4 types of medical data, on the entire Medicare population of a big insurance company, and more importantly, directly generating model interpretation to support user decision. In the future, we plan to enrich the model input by adding patients’ demographics and information from free-texted physician notes.

Keywords: deep learning, interpretability, attention, big data, medical conditions

Procedia PDF Downloads 91

Authors: Panaya Sudta, Wanchalerm Patanacharoenwong, Prachya Bumrungkun

Abstract:

As With the evolution of smart grids, ensuring the reliability and efficiency of electrical systems in smart cities has become crucial. This paper proposes a distinct approach that combines advanced machine learning techniques to accurately predict electrical failures and address voltage reliability issues. This approach aims to improve the accuracy and efficiency of reliability evaluations in smart cities. The aim of this research is to develop a comprehensive predictive model that accurately predicts electrical failures and voltage reliability in smart cities. This model integrates RFM analysis, K-means clustering, and LSTM networks to achieve this objective. The research utilizes RFM analysis, traditionally used in customer value assessment, to categorize and analyze electrical components based on their failure recency, frequency, and monetary impact. K-means clustering is employed to segment electrical components into distinct groups with similar characteristics and failure patterns. LSTM networks are used to capture the temporal dependencies and patterns in customer data. This integration of RFM, K-means, and LSTM results in a robust predictive tool for electrical failures and voltage reliability. The proposed model has been tested and validated on diverse electrical utility datasets. The results show a significant improvement in prediction accuracy and reliability compared to traditional methods, achieving an accuracy of 92.78% and an F1-score of 0.83. This research contributes to the proactive maintenance and optimization of electrical infrastructures in smart cities. It also enhances overall energy management and sustainability. The integration of advanced machine learning techniques in the predictive model demonstrates the potential for transforming the landscape of electrical system management within smart cities. The research utilizes diverse electrical utility datasets to develop and validate the predictive model. RFM analysis, K-means clustering, and LSTM networks are applied to these datasets to analyze and predict electrical failures and voltage reliability. The research addresses the question of how accurately electrical failures and voltage reliability can be predicted in smart cities. It also investigates the effectiveness of integrating RFM analysis, K-means clustering, and LSTM networks in achieving this goal. The proposed approach presents a distinct, efficient, and effective solution for predicting and mitigating electrical failures and voltage issues in smart cities. It significantly improves prediction accuracy and reliability compared to traditional methods. This advancement contributes to the proactive maintenance and optimization of electrical infrastructures, overall energy management, and sustainability in smart cities.

Keywords: electrical state prediction, smart grids, data-driven method, long short-term memory, RFM, k-means, machine learning

Procedia PDF Downloads 56