Search results for: memetic algorithms

Authors: Jillian M. Hagel, Joseph E. Tucker, Peter J. Facchini

Abstract:

With the aim of establishing a holistic approach for the treatment of central nervous system (CNS) disorders, we are pursuing a drug development program rapidly progressing through discovery and characterization phases. The drug candidates identified in this program are referred to as neuroplastogens owing to their ability to mediate neuroplasticity, which can be beneficial to patients suffering from anxiety, depression, or posttraumatic stress disorder. These and other related neuropsychiatric conditions are associated with the onset of neuronal atrophy, which is defined as a reduction in the number and/or productivity of neurons. The stimulation of neuroplasticity results in an increase in the connectivity between neurons and promotes the restoration of healthy brain function. We have synthesized a substantial catalogue of proprietary indolethylamine derivatives based on the general structures of serotonin (5-hydroxytryptamine) and psychedelic molecules such as N,N-dimethyltryptamine (DMT) and psilocin (4-hydroxy-DMT) that function as neuroplastogens. A primary objective in our screening protocol is the identification of derivatives associated with a significant reduction in hallucination, which will allow administration of the drug at a dose that induces neuroplasticity and triggers other efficacious outcomes in the treatment of targeted CNS disorders but which does not cause a psychedelic response in the patient. Both neuroplasticity and hallucination are associated with engagement of the 5HT2A receptor, requiring drug candidates differentially coupled to these two outcomes at a molecular level. We use novel and proprietary artificial intelligence algorithms to predict the mode of binding to the 5HT2A receptor, which has been shown to correlate with the hallucinogenic response. Hallucination is tested using the mouse head-twitch response model, whereas mouse marble-burying and sucrose preference assays are used to evaluate anxiolytic and anti-depressive potential. Neuroplasticity is assays using dendritic outgrowth assays and cell-based ELISA analysis. Pharmacokinetics and additional receptor-binding analyses also contribute the selection of lead candidates. A summary of the program is presented.

Keywords: neuroplastogen, non-hallucinogenic, drug development, anxiety, depression, PTSD, indolethylamine derivatives, psychedelic-inspired, 5-HT2A receptor, computational chemistry, head-twitch response behavioural model, neurite outgrowth assay

Procedia PDF Downloads 140

Authors: Misuk Kim, Seokho Kang, Jehyuk Lee, Hyunchang Cho, Sungzoon Cho

Abstract:

In semiconductor manufacturing, virtual metrology (VM) refers to methods to predict properties of a wafer based on machine parameters and sensor data of the production equipment, without performing the (costly) physical measurement of the wafer properties (Wikipedia). Additional benefits include avoidance of human bias and identification of important factors affecting the quality of the process which allow improving the process quality in the future. It is however rare to find VM applied to other areas of manufacturing. In this work, we propose to use VM to copper clad laminate (CCL) manufacturing. CCL is a core element of a printed circuit board (PCB) which is used in smartphones, tablets, digital cameras, and laptop computers. The manufacturing of CCL consists of three processes: Treating, lay-up, and pressing. Treating, the most important process among the three, puts resin on glass cloth, heat up in a drying oven, then produces prepreg for lay-up process. In this process, three important quality factors are inspected: Treated weight (T/W), Minimum Viscosity (M/V), and Gel Time (G/T). They are manually inspected, incurring heavy cost in terms of time and money, which makes it a good candidate for VM application. We developed prediction models of the three quality factors T/W, M/V, and G/T, respectively, with process variables, raw material, and environment variables. The actual process data was obtained from a CCL manufacturer. A variety of variable selection methods and learning algorithms were employed to find the best prediction model. We obtained prediction models of M/V and G/T with a high enough accuracy. They also provided us with information on “important” predictor variables, some of which the process engineers had been already aware and the rest of which they had not. They were quite excited to find new insights that the model revealed and set out to do further analysis on them to gain process control implications. T/W did not turn out to be possible to predict with a reasonable accuracy with given factors. The very fact indicates that the factors currently monitored may not affect T/W, thus an effort has to be made to find other factors which are not currently monitored in order to understand the process better and improve the quality of it. In conclusion, VM application to CCL’s treating process was quite successful. The newly built quality prediction model allowed one to reduce the cost associated with actual metrology as well as reveal some insights on the factors affecting the important quality factors and on the level of our less than perfect understanding of the treating process.

Keywords: copper clad laminate, predictive modeling, quality control, virtual metrology

Procedia PDF Downloads 351

Authors: C. Román, M. Guevara, P. Salas, D. Duclap, J. Houenou, C. Poupon, J. F. Mangin, P. Guevara

Abstract:

Diffusion Magnetic Resonance Imaging (dMRI) allows the non-invasive study of human brain white matter. From diffusion data, it is possible to reconstruct fiber trajectories using tractography algorithms. Our previous work consists in an automatic method for the identification of short association bundles of the superficial white matter (SWM), based on a whole brain inter-subject hierarchical clustering applied to a HARDI database. The method finds representative clusters of similar fibers, belonging to a group of subjects, according to a distance measure between fibers, using a non-linear registration (DTI-TK). The algorithm performs an automatic labeling based on the anatomy, defined by a cortex mesh parcelated with FreeSurfer software. The clustering was applied to two independent groups of 37 subjects. The clusters resulting from both groups were compared using a restrictive threshold of mean distance between each pair of bundles from different groups, in order to keep reproducible connections. In the left hemisphere, 48 reproducible bundles were found, while 43 bundles where found in the right hemisphere. An inter-hemispheric bundle correspondence was then applied. The symmetric horizontal reflection of the right bundles was calculated, in order to obtain the position of them in the left hemisphere. Next, the intersection between similar bundles was calculated. The pairs of bundles with a fiber intersection percentage higher than 50% were considered similar. The similar bundles between both hemispheres were fused and symmetrized. We obtained 30 common bundles between hemispheres. An atlas was created with the resulting bundles and used to segment 78 new subjects from another HARDI database, using a distance threshold between 6-8 mm according to the bundle length. Finally, a laterality index was calculated based on the bundle volume. Seven bundles of the atlas presented right laterality (IP_SP_1i, LO_LO_1i, Op_Tr_0i, PoC_PoC_0i, PoC_PreC_2i, PreC_SM_0i, y RoMF_RoMF_0i) and one presented left laterality (IP_SP_2i), there is no tendency of lateralization according to the brain region. Many factors can affect the results, like tractography artifacts, subject registration, and bundle segmentation. Further studies are necessary in order to establish the influence of these factors and evaluate SWM laterality.

Keywords: dMRI, hierarchical clustering, lateralization index, tractography

Procedia PDF Downloads 331

Authors: Javed Mohammed

Abstract:

Right now, bio-medical infrastructure lags well behind the curve. Our healthcare system is dispersed and disjointed; medical records are a bit of a mess; and we do not yet have the capacity to store and process the crazy amounts of data coming our way from widespread whole-genome sequencing. And then there are privacy issues. Despite these infrastructure challenges, some researchers are plunging into bio medical Big Data now, in hopes of extracting new and actionable knowledge. They are doing delving into molecular-level data to discover bio markers that help classify patients based on their response to existing treatments; and pushing their results out to physicians in novel and creative ways. Computer scientists and bio medical researchers are able to transform data into models and simulations that will enable scientists for the first time to gain a profound under-standing of the deepest biological functions. Solving biological problems may require High-Performance Computing HPC due either to the massive parallel computation required to solve a particular problem or to algorithmic complexity that may range from difficult to intractable. Many problems involve seemingly well-behaved polynomial time algorithms (such as all-to-all comparisons) but have massive computational requirements due to the large data sets that must be analyzed. High-throughput techniques for DNA sequencing and analysis of gene expression have led to exponential growth in the amount of publicly available genomic data. With the increased availability of genomic data traditional database approaches are no longer sufficient for rapidly performing life science queries involving the fusion of data types. Computing systems are now so powerful it is possible for researchers to consider modeling the folding of a protein or even the simulation of an entire human body. This research paper emphasizes the computational biology's growing need for high-performance computing and Big Data. It illustrates this article’s indispensability in meeting the scientific and engineering challenges of the twenty-first century, and how Protein Folding (the structure and function of proteins) and Phylogeny Reconstruction (evolutionary history of a group of genes) can use HPC that provides sufficient capability for evaluating or solving more limited but meaningful instances. This article also indicates solutions to optimization problems, and benefits Big Data and Computational Biology. The article illustrates the Current State-of-the-Art and Future-Generation Biology of HPC Computing with Big Data.

Keywords: high performance, big data, parallel computation, molecular data, computational biology

Procedia PDF Downloads 365

Authors: Prince Vivek, Vijay K. Bharti, Manishi Mukesh, Ankita Sharma, Om Prakash Chaurasia, Bhuvnesh Kumar

Abstract:

High performance of endurance in equid requires adaptive changes involving physio-biochemical, and molecular responses in an attempt to regain homeostasis. We hypothesized that the identification of the suitable reference genes might be considered for assessing of endurance related traits in pony at high altitude and may ensure for individuals struggling to potent endurance trait in ponies at high altitude. A total of 12 mares of ponies, Zanskar breed, were divided into three groups, group-A (without load), group-B, (60 Kg) and group-C (80 Kg) on backpack loads were subjected to a load carry protocol, on a steep climb of 4 km uphill, and of gravel, uneven rocky surface track at an altitude of 3292 m to 3500 m (endpoint). Blood was collected before and immediately after the load carry on sodium heparin anticoagulant, and the peripheral blood mononuclear cell was separated for total RNA isolation and thereafter cDNA synthesis. Real time-PCR reactions were carried out to evaluate the mRNAs expression profile of a panel of putative internal control genes (ICGs), related to different functional classes, namely glyceraldehyde 3-phosphate dehydrogenase (GAPDH), β₂ microglobulin (β₂M), β-actin (ACTB), ribosomal protein 18 (RS18), hypoxanthine-guanine phosophoribosyltransferase (HPRT), ubiquitin B (UBB), ribosomal protein L32 (RPL32), transferrin receptor protein (TFRC), succinate dehydrogenase complex subunit A (SDHA) for normalizing the real-time quantitative polymerase chain reaction (qPCR) data of native pony’s. Three different algorithms, geNorm, NormFinder, and BestKeeper software, were used to evaluate the stability of reference genes. The result showed that GAPDH was best stable gene and stability value for the best combination of two genes was observed TFRC and β₂M. In conclusion, the geometric mean of GAPDH, TFRC and β₂M might be used for accurate normalization of transcriptional data for assessing endurance related traits in Zanskar ponies during load carrying.

Keywords: endurance exercise, ubiquitin B (UBB), β₂ microglobulin (β₂M), high altitude, Zanskar ponies, reference gene

Procedia PDF Downloads 132

Authors: Byeongcheol Kang, Junyi Kim, Hyungsik Jung, Hyungjun Yang, Jaewoo An, Jonggeun Choe

Abstract:

It is important to figure out reservoir properties for better production management. Due to the limited information, there are geological uncertainties on very heterogeneous or channel reservoir. One of the solutions is to generate multiple equi-probable realizations using geostatistical methods. However, some models have wrong properties, which need to be excluded for simulation efficiency and reliability. We propose a novel method of model selection scheme, based on distance-based clustering for reliable application of production optimization algorithm. Distance is defined as a degree of dissimilarity between the data. We calculate Hausdorff distance to classify the models based on their similarity. Hausdorff distance is useful for shape matching of the reservoir models. We use multi-dimensional scaling (MDS) to describe the models on two dimensional space and group them by K-means clustering. Rather than simulating all models, we choose one representative model from each cluster and find out the best model, which has the similar production rates with the true values. From the process, we can select good reservoir models near the best model with high confidence. We make 100 channel reservoir models using single normal equation simulation (SNESIM). Since oil and gas prefer to flow through the sand facies, it is critical to characterize pattern and connectivity of the channels in the reservoir. After calculating Hausdorff distances and projecting the models by MDS, we can see that the models assemble depending on their channel patterns. These channel distributions affect operation controls of each production well so that the model selection scheme improves management optimization process. We use one of useful global search algorithms, particle swarm optimization (PSO), for our production optimization. PSO is good to find global optimum of objective function, but it takes too much time due to its usage of many particles and iterations. In addition, if we use multiple reservoir models, the simulation time for PSO will be soared. By using the proposed method, we can select good and reliable models that already matches production data. Considering geological uncertainty of the reservoir, we can get well-optimized production controls for maximum net present value. The proposed method shows one of novel solutions to select good cases among the various probabilities. The model selection schemes can be applied to not only production optimization but also history matching or other ensemble-based methods for efficient simulations.

Keywords: distance-based clustering, geological uncertainty, particle swarm optimization (PSO), production optimization

Procedia PDF Downloads 144

Authors: Hoda Rahmani, Gary Weckman

Abstract:

The body of research examining the integration of drones into various industries is expanding rapidly. Despite progress made in addressing the cybersecurity concerns for commercial drones, knowledge deficits remain in determining potential occupational hazards and risks of drone use to employees’ well-being and health in the workplace. This creates difficulty in identifying key approaches to risk mitigation strategies and thus reflects the need for raising awareness among employers, safety professionals, and policymakers about workplace drone-related accidents. The purpose of this study is to investigate the prevalence of and possible risk factors for drone-related mishaps by comparing the application of drones in construction with manufacturing industries. The chief reason for considering these specific sectors is to ascertain whether there exists any significant difference between indoor and outdoor flights since most construction sites use drones outside and vice versa. Therefore, the current research seeks to examine the causes and patterns of workplace drone-related mishaps and suggest possible ergonomic interventions through data collection. Potential ergonomic practices to mitigate hazards associated with flying drones could include providing operators with professional pieces of training, conducting a risk analysis, and promoting the use of personal protective equipment. For the purpose of data analysis, two data mining techniques, the random forest and association rule mining algorithms, will be performed to find meaningful associations and trends in data as well as influential features that have an impact on the occurrence of drone-related accidents in construction and manufacturing sectors. In addition, Spearman’s correlation and chi-square tests will be used to measure the possible correlation between different variables. Indeed, by recognizing risks and hazards, occupational safety stakeholders will be able to pursue data-driven and evidence-based policy change with the aim of reducing drone mishaps, increasing productivity, creating a safer work environment, and extending human performance in safe and fulfilling ways. This research study was supported by the National Institute for Occupational Safety and Health through the Pilot Research Project Training Program of the University of Cincinnati Education and Research Center Grant #T42OH008432.

Keywords: commercial drones, ergonomic interventions, occupational safety, pattern recognition

Procedia PDF Downloads 212

Authors: Rui Fonseca-Pinto, Emanuel Silva, Rui Rijo, Ricardo Martinho, Bruno Carreira

Abstract:

The sedentary lifestyle is associated to the development of chronic noncommunicable diseases (obesity, hypertension, Diabetes Mellitus Type 2) and the World Health Organization, given the evidence that physical activity is determinant for individual and collective health, defined the Physical Activity Level (PAL) as a vital signal. Strategies for increasing the practice of physical activity in all age groups have emerged from the various social organizations (municipalities, universities, health organizations, companies, social groups) by increasingly developing innovative strategies to promote motivation strategies and conditions to the practice of physical activity. The adaptation of cities to the new paradigms of sustainable mobility has provided the adaptation of urban training circles and mobilized citizens to combat sedentarism. This adaptation has accompanied the technological evolution and makes possible the use of mobile technology to monitor outdoor training programs and also, through the network connection (IoT), use the training data to make personalized recommendations. This work presents a physical activity counseling platform to be used in the physical maintenance circuits of urban centers, the MOVIDA.polis. The platform consists of a back office for the management of circuits and training stations, and for a mobile application for monitoring the user performance during workouts. Using a QRcode, each training station is recognized by the App and based on the individual performance records (effort perception, heart rate variation) artificial intelligence algorithms are used to make a new personalized recommendation. The results presented in this work were obtained during the proof of concept phase, which was carried out in the PolisLeiria training circuit in the city of Leiria (Portugal). It was possible to verify the increase in adherence to the practice of physical activity, as well as to decrease the interval between training days. Moreover, the AI-based recommendation acts as a partner in the training and an additional challenging factor. The platform is ready to be used by other municipalities in order to reduce the levels of sedentarism and approach the weekly goal of 150 minutes of moderate physical activity. Acknowledgments: This work was supported by Fundação para a Ciência e Tecnologia FCT- Portugal and CENTRO2020 under the scope of MOVIDA project: 02/SAICT/2016 – 23878.

Keywords: physical activity, mHealth, urban training circuits, health promotion

Procedia PDF Downloads 173

Authors: Paolo Guzzini, Emad Samuel M. Ebeid

Abstract:

Bridges and Railways are critical infrastructures. Ensuring safety for transports using such assets is a primary goal as it directly impacts the lives of people. By the way, improving safety could require increased investments in O&M, and therefore optimizing resource usage for asset maintenance becomes crucial. Drones4Safety (D4S), a European project funded under the H2020 Research and Innovation Action (RIA) program, aims to increase the safety of the European civil transport by building a system that relies on 3 main pillars: • Drones operating autonomously in swarm mode; • Drones able to recharge themselves using inductive phenomena produced by transmission lines in the nearby of bridges and railways assets to be inspected; • Data acquired that are analyzed with AI-empowered algorithms for defect detection This paper describes the business model behind this disruptive project. The Business Model is structured in 2 parts: • The first part is focused on the design of the business model Canvas, to explain the value provided by the Drone4safety project; • The second part aims at defining a detailed financial analysis, with the target of calculating the IRR (Internal Return rate) and the NPV (Net Present Value) of the investment in a 7 years plan (2 years to run the project + 5 years post-implementation). As to the financial analysis 2 different points of view are assumed: • Point of view of the Drones4safety company in charge of designing, producing, and selling the new system; • Point of view of the Utility company that will adopt the new system in its O&M practices; Assuming the point of view of the Drones4safety company 3 scenarios were considered: • Selling the drones > revenues will be produced by the drones’ sales; • Renting the drones > revenues will be produced by the rental of the drones (with a time-based model); • Selling the data acquisition service > revenues will be produced by the sales of pictures acquired by drones; Assuming the point of view of a utility adopting the D4S system, a 4th scenario was analyzed taking into account the decremental costs related to the change of operation and maintenance practices. The paper will show, for both companies, what are the key parameters affecting most of the business model and which are the sustainable scenarios.

Keywords: a swarm of drones, AI, bridges, railways, drones4safety company, utility companies

Procedia PDF Downloads 141

Authors: Minh Huynh, Aaron Heuser, Luke Patterson, Chris Zhang, Mason Miller, Daniel Wang, Sandeep Shetty, Mike Trinh, Abigail Miller, Adaeze Enekwechi, Tenille Daniels, Lu Huynh

Abstract:

Machine learning and text analytics have been used to analyze child abuse, cyberbullying, domestic abuse and domestic violence, and hate speech. However, to the authors’ knowledge, no research to date has used these methods to study elder abuse in nursing homes or skilled nursing facilities from field inspection reports. We used machine learning and text analytics methods to analyze 356,000 inspection reports, which have been extracted from CMS Form-2567 field inspections of US nursing homes and skilled nursing facilities between 2016 and 2021. Our algorithm detected occurrences of the various types of abuse, including physical abuse, psychological abuse, verbal abuse, sexual abuse, and passive and active neglect. For example, to detect physical abuse, our algorithms search for combinations or phrases and words suggesting willful infliction of damage (hitting, pinching or burning, tethering, tying), or consciously ignoring an emergency. To detect occurrences of elder neglect, our algorithm looks for combinations or phrases and words suggesting both passive neglect (neglecting vital needs, allowing malnutrition and dehydration, allowing decubiti, deprivation of information, limitation of freedom, negligence toward safety precautions) and active neglect (intimidation and name-calling, tying the victim up to prevent falls without consent, consciously ignoring an emergency, not calling a physician in spite of indication, stopping important treatments, failure to provide essential care, deprivation of nourishment, leaving a person alone for an inappropriate amount of time, excessive demands in a situation of care). We further compare the prevalence of abuse before and after Covid-19 related restrictions on nursing home visits. We also identified the facilities with the most number of cases of abuse with no abuse facilities within a 25-mile radius as most likely candidates for additional inspections. We also built an interactive display to visualize the location of these facilities.

Keywords: machine learning, text analytics, elder abuse, elder neglect, nursing home abuse

Procedia PDF Downloads 148

Authors: Nathaniel Bayode Eniolorunda, Murtala Abubakar Gada, Sheikh Danjuma Abubakar

Abstract:

The challenge of erosion in the study area is persistent, suggesting the need for a better understanding of the mechanisms that drive it. Thus, the study evolved a predictive erosion model (RUSLE_Sok), deploying Remote Sensing (RS) and Geographical Information System (GIS) tools. The nature and pattern of the factors of erosion were characterized, while soil losses were quantified. Factors’ impacts were also measured, and the morphometry of gullies was described. Data on the five factors of RUSLE and distances to settlements, rivers and roads (K, R, LS, P, C, DS DRd and DRv) were combined and processed following standard RS and GIS algorithms. Harmonized World Soil Data (HWSD), Shuttle Radar Topographical Mission (SRTM) image, Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), Sentinel-2 image accessed and processed within the Google Earth Engine, road network and settlements were the data combined and calibrated into the factors for erosion modeling. A gully morphometric study was conducted at some purposively selected sites. Factors of soil erosion showed low, moderate, to high patterns. Soil losses ranged from 0 to 32.81 tons/ha/year, classified into low (97.6%), moderate (0.2%), severe (1.1%) and very severe (1.05%) forms. The multiple regression analysis shows that factors statistically significantly predicted soil loss, F (8, 153) = 55.663, p < .0005. Except for the C-Factor with a negative coefficient, all other factors were positive, with contributions in the order of LS>C>R>P>DRv>K>DS>DRd. Gullies are generally from less than 100m to about 3km in length. Average minimum and maximum depths at gully heads are 0.6 and 1.2m, while those at mid-stream are 1 and 1.9m, respectively. The minimum downstream depth is 1.3m, while that for the maximum is 4.7m. Deeper gullies exist in proximity to rivers. With minimum and maximum gully elevation values ranging between 229 and 338m and an average slope of about 3.2%, the study area is relatively flat. The study concluded that major erosion influencers in the study area are topography and vegetation cover and that the RUSLE_Sok well predicted soil loss more effectively than ordinary RUSLE. The adoption of conservation measures such as tree planting and contour ploughing on sloppy farmlands was recommended.

Keywords: RUSLE_Sok, Sokoto, google earth engine, sentinel-2, erosion

Procedia PDF Downloads 78

Authors: Noah Odion, Honest Jimu, Blessing Atinuke Afuape

Abstract:

Introduction: Motor control and gait abnormalities are commonly observed in individuals with autism spectrum disorder (ASD), affecting their mobility and coordination. Understanding the underlying neurological and biomechanical factors is essential for designing effective interventions. This study focuses on developing microchip-integrated wearable devices to capture real-time movement data from individuals with autism. By applying computational models to the collected data, we aim to analyze motor control patterns and gait abnormalities, bridging a crucial knowledge gap in autism-related motor dysfunction. Methods: We designed microchip-enabled wearable devices capable of capturing precise kinematic data, including joint angles, acceleration, and velocity during movement. A cross-sectional study was conducted on individuals with ASD and a control group to collect comparative data. Computational modelling was applied using machine learning algorithms to analyse motor control patterns, focusing on gait variability, balance, and coordination. Finite element models were also used to simulate muscle and joint dynamics. The study employed descriptive and analytical methods to interpret the motor data. Results: The wearable devices effectively captured detailed movement data, revealing significant gait variability in the ASD group. For example, gait cycle time was 25% longer, and stride length was reduced by 15% compared to the control group. Motor control analysis showed a 30% reduction in balance stability in individuals with autism. Computational models successfully predicted movement irregularities and helped identify motor control deficits, particularly in the lower limbs. Conclusions: The integration of microchip-based wearable devices with computational models offers a powerful tool for diagnosing and treating motor control deficits in autism. These results have significant implications for patient care, providing objective data to guide personalized therapeutic interventions. The findings also contribute to the broader field of neuroscience by improving our understanding of the motor dysfunctions associated with ASD and other neurodevelopmental disorders.

Keywords: motor control, gait abnormalities, autism, wearable devices, microchips, computational modeling, kinematic analysis, neurodevelopmental disorders

Procedia PDF Downloads 25

Authors: Stephanie Chen, Mitja Echim, Christof Büskens

Abstract:

To describe and propagate the behavior of a system mathematical models are formulated. Parameter identification is used to adapt the coefficients of the underlying laws of science. For complex systems this approach can be incomplete and hence imprecise and moreover too slow to be computed efficiently. Therefore, these models might be not applicable for the numerical optimization of real systems, since these techniques require numerous evaluations of the models. Moreover not all quantities necessary for the identification might be available and hence the system must be adapted manually. Therefore, an approach is described that generates models that overcome the before mentioned limitations by not focusing on physical laws, but on measured (sensor) data of real systems. The approach is more general since it generates models for every system detached from the scientific background. Additionally, this approach can be used in a more general sense, since it is able to automatically identify correlations in the data. The method can be classified as a multivariate data regression analysis. In contrast to many other data regression methods this variant is also able to identify correlations of products of variables and not only of single variables. This enables a far more precise and better representation of causal correlations. The basis and the explanation of this method come from an analytical background: the series expansion. Another advantage of this technique is the possibility of real-time adaptation of the generated models during operation. Herewith system changes due to aging, wear or perturbations from the environment can be taken into account, which is indispensable for realistic scenarios. Since these data driven models can be evaluated very efficiently and with high precision, they can be used in mathematical optimization algorithms that minimize a cost function, e.g. time, energy consumption, operational costs or a mixture of them, subject to additional constraints. The proposed method has successfully been tested in several complex applications and with strong industrial requirements. The generated models were able to simulate the given systems with an error in precision less than one percent. Moreover the automatic identification of the correlations was able to discover so far unknown relationships. To summarize the above mentioned approach is able to efficiently compute high precise and real-time-adaptive data-based models in different fields of industry. Combined with an effective mathematical optimization algorithm like WORHP (We Optimize Really Huge Problems) several complex systems can now be represented by a high precision model to be optimized within the user wishes. The proposed methods will be illustrated with different examples.

Keywords: adaptive modeling, automatic identification of correlations, data based modeling, optimization

Procedia PDF Downloads 409

Authors: Yangzhi Li

Abstract:

Network transfer of information and performance customization is now a viable method of digital industrial production in the era of Industry 4.0. Robot platforms and network platforms have grown more important in digital design and construction. The pressing need for novel building techniques is driven by the growing labor scarcity problem and increased awareness of construction safety. Robotic approaches in construction research are regarded as an extension of operational and production tools. Several technological theories related to robot autonomous recognition, which include high-performance computing, physical system modeling, extensive sensor coordination, and dataset deep learning, have not been explored using intelligent construction. Relevant transdisciplinary theory and practice research still has specific gaps. Optimizing high-performance computing and autonomous recognition visual guidance technologies improves the robot's grasp of the scene and capacity for autonomous operation. Intelligent vision guidance technology for industrial robots has a serious issue with camera calibration, and the use of intelligent visual guiding and identification technologies for industrial robots in industrial production has strict accuracy requirements. It can be considered that visual recognition systems have challenges with precision issues. In such a situation, it will directly impact the effectiveness and standard of industrial production, necessitating a strengthening of the visual guiding study on positioning precision in recognition technology. To best facilitate the handling of complicated components, an approach for the visual recognition of parts utilizing machine learning algorithms is proposed. This study will identify the position of target components by detecting the information at the boundary and corner of a dense point cloud and determining the aspect ratio in accordance with the guidelines for the modularization of building components. To collect and use components, operational processing systems assign them to the same coordinate system based on their locations and postures. The RGB image's inclination detection and the depth image's verification will be used to determine the component's present posture. Finally, a virtual environment model for the robot's obstacle-avoidance route will be constructed using the point cloud information.

Keywords: robotic construction, robotic assembly, visual guidance, machine learning

Procedia PDF Downloads 87

Authors: Vinicius G. Goecks, Gregory M. Gremillion, William D. Nothwang

Abstract:

Modern approaches to train intelligent agents rely on prolonged training sessions, high amounts of input data, and multiple interactions with the environment. This restricts the application of these learning algorithms in robotics and real-world applications, in which there is low tolerance to inadequate actions, interactions are expensive, and real-time processing and action are required. This paper addresses this issue introducing CyberSteer, a novel approach to efficiently design intrinsic reward functions based on human intention to guide deep reinforcement learning agents with no environment-dependent rewards. CyberSteer uses non-expert human operators for initial demonstration of a given task or desired behavior. The trajectories collected are used to train a behavior cloning deep neural network that asynchronously runs in the background and suggests actions to the deep reinforcement learning module. An intrinsic reward is computed based on the similarity between actions suggested and taken by the deep reinforcement learning algorithm commanding the agent. This intrinsic reward can also be reshaped through additional human demonstration or critique. This approach removes the need for environment-dependent or hand-engineered rewards while still being able to safely shape the behavior of autonomous robotic agents, in this case, based on human intention. CyberSteer is tested in a high-fidelity unmanned aerial vehicle simulation environment, the Microsoft AirSim. The simulated aerial robot performs collision avoidance through a clustered forest environment using forward-looking depth sensing and roll, pitch, and yaw references angle commands to the flight controller. This approach shows that the behavior of robotic systems can be shaped in a reduced amount of time when guided by a non-expert human, who is only aware of the high-level goals of the task. Decreasing the amount of training time required and increasing safety during training maneuvers will allow for faster deployment of intelligent robotic agents in dynamic real-world applications.

Keywords: human-robot interaction, intelligent robots, robot learning, semisupervised learning, unmanned aerial vehicles

Procedia PDF Downloads 259

Authors: Soheila Sadeghi

Abstract:

Construction projects are complex and often subject to significant cost overruns due to the multifaceted nature of the activities involved. Accurate cost estimation is crucial for effective budget planning and resource allocation. Traditional methods for predicting overruns often rely on expert judgment or analysis of historical data, which can be time-consuming, subjective, and may fail to consider important factors. However, with the increasing availability of data from construction projects, machine learning techniques can be leveraged to improve the accuracy of overrun predictions. This study applied machine learning algorithms to enhance the prediction of cost overruns in a case study of a construction project. The methodology involved the development and evaluation of two machine learning models: Random Forest and Neural Networks. Random Forest can handle high-dimensional data, capture complex relationships, and provide feature importance estimates. Neural Networks, particularly Deep Neural Networks (DNNs), are capable of automatically learning and modeling complex, non-linear relationships between input features and the target variable. These models can adapt to new data, reduce human bias, and uncover hidden patterns in the dataset. The findings of this study demonstrate that both Random Forest and Neural Networks can significantly improve the accuracy of cost overrun predictions compared to traditional methods. The Random Forest model also identified key cost drivers and risk factors, such as changes in the scope of work and delays in material delivery, which can inform better project risk management. However, the study acknowledges several limitations. First, the findings are based on a single construction project, which may limit the generalizability of the results to other projects or contexts. Second, the dataset, although comprehensive, may not capture all relevant factors influencing cost overruns, such as external economic conditions or political factors. Third, the study focuses primarily on cost overruns, while schedule overruns are not explicitly addressed. Future research should explore the application of machine learning techniques to a broader range of projects, incorporate additional data sources, and investigate the prediction of both cost and schedule overruns simultaneously.

Keywords: cost prediction, machine learning, project management, random forest, neural networks

Procedia PDF Downloads 60

Authors: Jaehyung An, Sungjoo Lee

Abstract:

Patent documents constitute an up-to-date and reliable source of knowledge for reflecting technological advance, so patent analysis has been widely used for identification of technological trends and formulation of technology strategies. But, identifying technological information from patent data entails some limitations such as, high cost, complexity, and inconsistency because it rely on the expert’ knowledge. To overcome these limitations, researchers have applied to a quantitative analysis based on the keyword technique. By using this method, you can include a technological implication, particularly patent documents, or extract a keyword that indicates the important contents. However, it only uses the simple-counting method by keyword frequency, so it cannot take into account the sematic relationship with the keywords and sematic information such as, how the technologies are used in their technology area and how the technologies affect the other technologies. To automatically analyze unstructured technological information in patents to extract the semantic information, it should be transformed into an abstracted form that includes the technological key concepts. Specific sentence structure ‘SAO’ (subject, action, object) is newly emerged by representing ‘key concepts’ and can be extracted by NLP (Natural language processor). An SAO structure can be organized in a problem-solution format if the action-object (AO) states that the problem and subject (S) form the solution. In this paper, we propose the new methodology that can extract the SAO structure through technical elements extracting rules. Although sentence structures in the patents text have a unique format, prior studies have depended on general NLP (Natural language processor) applied to the common documents such as newspaper, research paper, and twitter mentions, so it cannot take into account the specific sentence structure types of the patent documents. To overcome this limitation, we identified a unique form of the patent sentences and defined the SAO structures in the patents text data. There are four types of technical elements that consist of technology adoption purpose, application area, tool for technology, and technical components. These four types of sentence structures from patents have their own specific word structure by location or sequence of the part of speech at each sentence. Finally, we developed algorithms for extracting SAOs and this result offer insight for the technology innovation process by providing different perspectives of technology.

Keywords: NLP, patent analysis, SAO, semantic-analysis

Procedia PDF Downloads 262

Authors: Kumar Shashvat, Amol P. Bhondekar

Abstract:

In the midst of the five senses, odor is the most reminiscent and least understood. Odor testing has been mysterious and odor data fabled to most practitioners. The delinquent of recognition and classification of odor is important to achieve. The facility to smell and predict whether the artifact is of further use or it has become undesirable for consumption; the imitation of this problem hooked on a model is of consideration. The general industrial standard for this classification is color based anyhow; odor can be improved classifier than color based classification and if incorporated in machine will be awfully constructive. For cataloging of odor for peas, trees and cashews various discriminative approaches have been used Discriminative approaches offer good prognostic performance and have been widely used in many applications but are incapable to make effectual use of the unlabeled information. In such scenarios, generative approaches have better applicability, as they are able to knob glitches, such as in set-ups where variability in the series of possible input vectors is enormous. Generative models are integrated in machine learning for either modeling data directly or as a transitional step to form an indeterminate probability density function. The algorithms or models Linear Discriminant Analysis and Naive Bayes Classifier have been used for classification of the odor of cashews. Linear Discriminant Analysis is a method used in data classification, pattern recognition, and machine learning to discover a linear combination of features that typifies or divides two or more classes of objects or procedures. The Naive Bayes algorithm is a classification approach base on Bayes rule and a set of qualified independence theory. Naive Bayes classifiers are highly scalable, requiring a number of restraints linear in the number of variables (features/predictors) in a learning predicament. The main recompenses of using the generative models are generally a Generative Models make stronger assumptions about the data, specifically, about the distribution of predictors given the response variables. The Electronic instrument which is used for artificial odor sensing and classification is an electronic nose. This device is designed to imitate the anthropological sense of odor by providing an analysis of individual chemicals or chemical mixtures. The experimental results have been evaluated in the form of the performance measures i.e. are accuracy, precision and recall. The investigational results have proven that the overall performance of the Linear Discriminant Analysis was better in assessment to the Naive Bayes Classifier on cashew dataset.

Keywords: odor classification, generative models, naive bayes, linear discriminant analysis

Procedia PDF Downloads 390

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 120

Authors: Soheila Sadeghi

Abstract:

Construction projects are complex and often subject to significant cost overruns due to the multifaceted nature of the activities involved. Accurate cost estimation is crucial for effective budget planning and resource allocation. Traditional methods for predicting overruns often rely on expert judgment or analysis of historical data, which can be time-consuming, subjective, and may fail to consider important factors. However, with the increasing availability of data from construction projects, machine learning techniques can be leveraged to improve the accuracy of overrun predictions. This study applied machine learning algorithms to enhance the prediction of cost overruns in a case study of a construction project. The methodology involved the development and evaluation of two machine learning models: Random Forest and Neural Networks. Random Forest can handle high-dimensional data, capture complex relationships, and provide feature importance estimates. Neural Networks, particularly Deep Neural Networks (DNNs), are capable of automatically learning and modeling complex, non-linear relationships between input features and the target variable. These models can adapt to new data, reduce human bias, and uncover hidden patterns in the dataset. The findings of this study demonstrate that both Random Forest and Neural Networks can significantly improve the accuracy of cost overrun predictions compared to traditional methods. The Random Forest model also identified key cost drivers and risk factors, such as changes in the scope of work and delays in material delivery, which can inform better project risk management. However, the study acknowledges several limitations. First, the findings are based on a single construction project, which may limit the generalizability of the results to other projects or contexts. Second, the dataset, although comprehensive, may not capture all relevant factors influencing cost overruns, such as external economic conditions or political factors. Third, the study focuses primarily on cost overruns, while schedule overruns are not explicitly addressed. Future research should explore the application of machine learning techniques to a broader range of projects, incorporate additional data sources, and investigate the prediction of both cost and schedule overruns simultaneously.

Keywords: resource allocation, machine learning, optimization, data-driven decision-making, project management

Procedia PDF Downloads 40

Authors: Mariamawit T. Belete

Abstract:

Cereal production and marketing are the means of livelihood for millions of households in Ethiopia. However, cereal production is constrained by technical and socio-economic factors. Among the technical factors, cereal crop diseases are the major contributing factors to the low yield. The aim of this research is to develop an integration of data mining and knowledge based system for sorghum anthracnose disease diagnosis that assists agriculture experts and development agents to make timely decisions. Anthracnose diagnosing systems gather information from Melkassa agricultural research center and attempt to score anthracnose severity scale. Empirical research is designed for data exploration, modeling, and confirmatory procedures for testing hypothesis and prediction to draw a sound conclusion. WEKA (Waikato Environment for Knowledge Analysis) was employed for the modeling. Knowledge based system has come across a variety of approaches based on the knowledge representation method; case-based reasoning (CBR) is one of the popular approaches used in knowledge-based system. CBR is a problem solving strategy that uses previous cases to solve new problems. The system utilizes hidden knowledge extracted by employing clustering algorithms, specifically K-means clustering from sampled anthracnose dataset. Clustered cases with centroid value are mapped to jCOLIBRI, and then the integrator application is created using NetBeans with JDK 8.0.2. The important part of a case based reasoning model includes case retrieval; the similarity measuring stage, reuse; which allows domain expert to transfer retrieval case solution to suit for the current case, revise; to test the solution, and retain to store the confirmed solution to the case base for future use. Evaluation of the system was done for both system performance and user acceptance. For testing the prototype, seven test cases were used. Experimental result shows that the system achieves an average precision and recall values of 70% and 83%, respectively. User acceptance testing also performed by involving five domain experts, and an average of 83% acceptance is achieved. Although the result of this study is promising, however, further study should be done an investigation on hybrid approach such as rule based reasoning, and pictorial retrieval process are recommended.

Keywords: sorghum anthracnose, data mining, case based reasoning, integration

Procedia PDF Downloads 82

Authors: Mohamad R. Moshtagh, Ahmad Bagheri

Abstract:

Accurate fault detection in rotating machinery is of utmost importance to ensure optimal performance and prevent costly downtime in industrial applications. This study presents a robust fault detection system based on vibration data collected from rotating gears under various operating conditions. The considered scenarios include: (1) both gears being healthy, (2) one healthy gear and one faulty gear, and (3) introducing an imbalanced condition to a healthy gear. Vibration data was acquired using a Hentek 1008 device and stored in a CSV file. Python code implemented in the Spider environment was used for data preprocessing and analysis. Winner features were extracted using the Wiener feature selection method. These features were then employed in multiple machine learning algorithms, including Convolutional Neural Networks (CNN), Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN), and Random Forest, to evaluate their performance in detecting and classifying faults in both the training and validation datasets. The comparative analysis of the methods revealed the superior performance of the Wiener-CNN approach. The Wiener-CNN method achieved a remarkable accuracy of 100% for both the two-class (healthy gear and faulty gear) and three-class (healthy gear, faulty gear, and imbalanced) scenarios in the training and validation datasets. In contrast, the other methods exhibited varying levels of accuracy. The Wiener-MLP method attained 100% accuracy for the two-class training dataset and 100% for the validation dataset. For the three-class scenario, the Wiener-MLP method demonstrated 100% accuracy in the training dataset and 95.3% accuracy in the validation dataset. The Wiener-KNN method yielded 96.3% accuracy for the two-class training dataset and 94.5% for the validation dataset. In the three-class scenario, it achieved 85.3% accuracy in the training dataset and 77.2% in the validation dataset. The Wiener-Random Forest method achieved 100% accuracy for the two-class training dataset and 85% for the validation dataset, while in the three-class training dataset, it attained 100% accuracy and 90.8% accuracy for the validation dataset. The exceptional accuracy demonstrated by the Wiener-CNN method underscores its effectiveness in accurately identifying and classifying fault conditions in rotating machinery. The proposed fault detection system utilizes vibration data analysis and advanced machine learning techniques to improve operational reliability and productivity. By adopting the Wiener-CNN method, industrial systems can benefit from enhanced fault detection capabilities, facilitating proactive maintenance and reducing equipment downtime.

Keywords: fault detection, gearbox, machine learning, wiener method

Procedia PDF Downloads 81

Authors: Oluwafunmibi Omotayo Fasanya, Augustine Kena Adjei

Abstract:

Diabetic nephropathy (DN) is a major complication of diabetes, with fibrosis and propionate metabolism playing critical roles in its progression. Identifying biomarkers linked to these pathways may provide novel insights into DN diagnosis and treatment. This study aims to identify biomarkers associated with fibrosis and propionate metabolism in DN. Analyze the biological pathways and regulatory mechanisms of these biomarkers. Develop a machine learning model to predict DN-related biomarkers and validate their functional roles. Publicly available transcriptome datasets related to DN (GSE96804 and GSE104948) were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/gds), and 924 propionate metabolism-related genes (PMRGs) and 656 fibrosis-related genes (FRGs) were identified. The analysis began with the extraction of DN-differentially expressed genes (DN-DEGs) and propionate metabolism-related DEGs (PM-DEGs), followed by the intersection of these with fibrosis-related genes to identify key intersected genes. Instead of relying on traditional models, we employed a combination of deep neural networks (DNNs) and ensemble methods such as Gradient Boosting Machines (GBM) and XGBoost to enhance feature selection and biomarker discovery. Recursive feature elimination (RFE) was coupled with these advanced algorithms to refine the selection of the most critical biomarkers. Functional validation was conducted using convolutional neural networks (CNN) for gene set enrichment and immunoinfiltration analysis, revealing seven significant biomarkers—SLC37A4, ACOX2, GPD1, ACE2, SLC9A3, AGT, and PLG. These biomarkers are involved in critical biological processes such as fatty acid metabolism and glomerular development, providing a mechanistic link to DN progression. Furthermore, a TF–miRNA–mRNA regulatory network was constructed using natural language processing models to identify 8 transcription factors and 60 miRNAs that regulate these biomarkers, while a drug–gene interaction network revealed potential therapeutic targets such as UROKINASE–PLG and ATENOLOL–AGT. This integrative approach, leveraging deep learning and ensemble models, not only enhances the accuracy of biomarker discovery but also offers new perspectives on DN diagnosis and treatment, specifically targeting fibrosis and propionate metabolism pathways.

Keywords: diabetic nephropathy, deep neural networks, gradient boosting machines (GBM), XGBoost

Procedia PDF Downloads 12

Authors: J. D. García Espinel, Paula Pérez Sánchez, Carlos Egea Ruiz, Carlos Lardín Mifsut, Andrés López-Aranguren Oliver

Abstract:

This paper shortly describes three main actuations over a Waste Water Treatment Plant (WWTP) for reducing its energy consumption: Optimization of the biological reactor in the aeration stage by including new control algorithms and introducing new efficient equipment, the installation of an innovative hybrid system with zero Grid injection (formed by 100kW of PV energy and 5 kW of mini-wind energy generation) and an intelligent management system for load consumption and energy generation control in the most optimum way. This project called RENEWAT, involved in the European Commission call LIFE 2013, has the main objective of reducing the energy consumptions through different actions on the processes which take place in a WWTP and introducing renewable energies on these treatment plants, with the purpose of promoting the usage of treated waste water for irrigation and decreasing the C02 gas emissions. WWTP is always required before waste water can be reused for irrigation or discharged in water bodies. However, the energetic demand of the treatment process is high enough for making the price of treated water to exceed the one for drinkable water. This makes any policy very difficult to encourage the re-use of treated water, with a great impact on the water cycle, particularly in those areas suffering hydric stress or deficiency. The cost of treating waste water involves another climate-change related burden: the energy necessary for the process is obtained mainly from the electric network, which is, in most of the cases in Europe, energy obtained from the burning of fossil fuels. The innovative part of this project is based on the implementation, adaptation and integration of solutions for this problem, together with a new concept of the integration of energy input and operative energy demand. Moreover, there is an important qualitative jump between the technologies used and the alleged technologies to use in the project which give it an innovative character, due to the fact that there are no similar previous experiences of a WWTP including an intelligent discrimination of energy sources, integrating renewable ones (PV and Wind) and the grid.

Keywords: aeration system, biological reactor, CO2 emissions, energy efficiency, hybrid systems, LIFE 2013 call, process optimization, renewable energy sources, wasted water treatment plants

Procedia PDF Downloads 352

Authors: Gautier Viaud, Paul-Henry Cournède

Abstract:

Plant growth models have been used extensively for the prediction of the phenotypic performance of plants. However, they remain most often calibrated for a given genotype and therefore do not take into account genotype by environment interactions. One way of achieving such an objective is to consider Bayesian hierarchical models. Three levels can be identified in such models: The first level describes how a given growth model describes the phenotype of the plant as a function of individual parameters, the second level describes how these individual parameters are distributed within a plant population, the third level corresponds to the attribution of priors on population parameters. Thanks to the Bayesian framework, choosing appropriate priors for the population parameters permits to derive analytical expressions for the full conditional distributions of these population parameters. As plant growth models are of a nonlinear nature, individual parameters cannot be sampled explicitly, and a Metropolis step must be performed. This allows for the use of a hybrid Gibbs--Metropolis sampler. A generic approach was devised for the implementation of both general state space models and estimation algorithms within a programming platform. It was designed using the Julia language, which combines an elegant syntax, metaprogramming capabilities and exhibits high efficiency. Results were obtained for Arabidopsis thaliana on both simulated and real data. An organ-scale Greenlab model for the latter is thus presented, where the surface areas of each individual leaf can be simulated. It is assumed that the error made on the measurement of leaf areas is proportional to the leaf area itself; multiplicative normal noises for the observations are therefore used. Real data were obtained via image analysis of zenithal images of Arabidopsis thaliana over a period of 21 days using a two-step segmentation and tracking algorithm which notably takes advantage of the Arabidopsis thaliana phyllotaxy. Since the model formulation is rather flexible, there is no need that the data for a single individual be available at all times, nor that the times at which data is available be the same for all the different individuals. This allows to discard data from image analysis when it is not considered reliable enough, thereby providing low-biased data in large quantity for leaf areas. The proposed model precisely reproduces the dynamics of Arabidopsis thaliana’s growth while accounting for the variability between genotypes. In addition to the estimation of the population parameters, the level of variability is an interesting indicator of the genotypic stability of model parameters. A promising perspective is to test whether some of the latter should be considered as fixed effects.

Keywords: bayesian, genotypic differentiation, hierarchical models, plant growth models

Procedia PDF Downloads 304

Authors: Samuel Munaf, Kevin Swingler, Franz Brülisauer, Anthony O’Hare, George Gunn, Aaron Reeves

Abstract:

Background: Avian influenza, more commonly known as Bird flu, is a viral zoonotic respiratory disease stemming from various species of poultry, including pets and migratory birds. Researchers have purported that the accessibility of health information online, in addition to the low-cost data collection methods the internet provides, has revolutionized the methods in which epidemiological and disease surveillance data is utilized. This paper examines the feasibility of using internet data sources, such as Twitter and livestock forums, for the early detection of the avian flu outbreak, through the use of text mining algorithms and social network analysis. Methods: Social media mining was conducted on Twitter between the period of 01/01/2021 to 31/12/2021 via the Twitter API in Python. The results were filtered firstly by hashtags (#avianflu, #birdflu), word occurrences (avian flu, bird flu, H5N1), and then refined further by location to include only those results from within the UK. Analysis was conducted on this text in a time-series manner to determine keyword frequencies and topic modeling to uncover insights in the text prior to a confirmed outbreak. Further analysis was performed by examining clinical signs (e.g., swollen head, blue comb, dullness) within the time series prior to the confirmed avian flu outbreak by the Animal and Plant Health Agency (APHA). Results: The increased search results in Google and avian flu-related tweets showed a correlation in time with the confirmed cases. Topic modeling uncovered clusters of word occurrences relating to livestock biosecurity, disposal of dead birds, and prevention measures. Conclusions: Text mining social media data can prove to be useful in relation to analysing discussed topics for epidemiological surveillance purposes, especially given the lack of applied research in the veterinary domain. The small sample size of tweets for certain weekly time periods makes it difficult to provide statistically plausible results, in addition to a great amount of textual noise in the data.

Keywords: veterinary epidemiology, disease surveillance, infodemiology, infoveillance, avian influenza, social media

Procedia PDF Downloads 106

Authors: Erva Akin

Abstract:

– The use of copyrighted material for machine learning purposes is a challenging issue in the field of artificial intelligence (AI). While machine learning algorithms require large amounts of data to train and improve their accuracy and creativity, the use of copyrighted material without permission from the authors may infringe on their intellectual property rights. In order to overcome copyright legal hurdle against the data sharing, access and re-use of data, the use of copyrighted material for machine learning purposes may be considered permissible under certain circumstances. For example, if the copyright holder has given permission to use the data through a licensing agreement, then the use for machine learning purposes may be lawful. It is also argued that copying for non-expressive purposes that do not involve conveying expressive elements to the public, such as automated data extraction, should not be seen as infringing. The focus of such ‘copy-reliant technologies’ is on understanding language rules, styles, and syntax and no creative ideas are being used. However, the non-expressive use defense is within the framework of the fair use doctrine, which allows the use of copyrighted material for research or educational purposes. The questions arise because the fair use doctrine is not available in EU law, instead, the InfoSoc Directive provides for a rigid system of exclusive rights with a list of exceptions and limitations. One could only argue that non-expressive uses of copyrighted material for machine learning purposes do not constitute a ‘reproduction’ in the first place. Nevertheless, the use of machine learning with copyrighted material is difficult because EU copyright law applies to the mere use of the works. Two solutions can be proposed to address the problem of copyright clearance for AI training data. The first is to introduce a broad exception for text and data mining, either mandatorily or for commercial and scientific purposes, or to permit the reproduction of works for non-expressive purposes. The second is that copyright laws should permit the reproduction of works for non-expressive purposes, which opens the door to discussions regarding the transposition of the fair use principle from the US into EU law. Both solutions aim to provide more space for AI developers to operate and encourage greater freedom, which could lead to more rapid innovation in the field. The Data Governance Act presents a significant opportunity to advance these debates. Finally, issues concerning the balance of general public interests and legitimate private interests in machine learning training data must be addressed. In my opinion, it is crucial that robot-creation output should fall into the public domain. Machines depend on human creativity, innovation, and expression. To encourage technological advancement and innovation, freedom of expression and business operation must be prioritised.

Keywords: artificial intelligence, copyright, data governance, machine learning

Procedia PDF Downloads 85

Authors: Mina Sadat Orooje, Mohammad Mehdi Latifi, Behnam Fereydooni Eftekhari

Abstract:

The integration of cutting-edge Internet of Things (IoT) technologies with advanced Artificial Intelligence (AI) systems is revolutionizing healthcare facility management. However, the current landscape of hospital building maintenance suffers from slow, repetitive, and disjointed processes, leading to significant financial, resource, and time losses. Additionally, the potential of Building Information Modeling (BIM) in facility maintenance is hindered by a lack of data within digital models of built environments, necessitating a more streamlined data collection process. This paper presents a robust framework that harmonizes AI with BIM-IoT technology to elevate healthcare Facility Maintenance Management (FMM) and address these pressing challenges. The methodology begins with a thorough literature review and requirements analysis, providing insights into existing technological landscapes and associated obstacles. Extensive data collection and analysis efforts follow to deepen understanding of hospital infrastructure and maintenance records. Critical AI algorithms are identified to address predictive maintenance, anomaly detection, and optimization needs alongside integration strategies for BIM and IoT technologies, enabling real-time data collection and analysis. The framework outlines protocols for data processing, analysis, and decision-making. A prototype implementation is executed to showcase the framework's functionality, followed by a rigorous validation process to evaluate its efficacy and gather user feedback. Refinement and optimization steps are then undertaken based on evaluation outcomes. Emphasis is placed on the scalability of the framework in real-world scenarios and its potential applications across diverse healthcare facility contexts. Finally, the findings are meticulously documented and shared within the healthcare and facility management communities. This framework aims to significantly boost maintenance efficiency, cut costs, provide decision support, enable real-time monitoring, offer data-driven insights, and ultimately enhance patient safety and satisfaction. By tackling current challenges in healthcare facility maintenance management it paves the way for the adoption of smarter and more efficient maintenance practices in healthcare facilities.

Keywords: artificial intelligence, building information modeling, healthcare facility maintenance, internet of things integration, maintenance efficiency

Procedia PDF Downloads 61

Authors: Francesco Lubrano, Fabrizio Bertone, Federico Stirano

Abstract:

Modern societies strongly depend on Critical Infrastructures (CI). Hospitals, power supplies, water supplies, telecommunications are just few examples of CIs that provide vital functions to societies. CIs like hospitals are very complex environments, characterized by a huge number of cyber and physical systems that are becoming increasingly integrated. Ensuring a high level of security within such critical infrastructure requires a deep knowledge of vulnerabilities, threats, and potential attacks that may occur, as well as defence and prevention or mitigation strategies. The possibility to remotely monitor and control almost everything is pushing the adoption of network-connected devices. This implicitly introduces new threats and potential vulnerabilities, posing a risk, especially to those devices connected to the Internet. Modern medical devices used in hospitals are not an exception and are more and more being connected to enhance their functionalities and easing the management. Moreover, hospitals are environments with high flows of people, that are difficult to monitor and can somehow easily have access to the same places used by the staff, potentially creating damages. It is therefore clear that physical and cyber threats should be considered, analysed, and treated together as cyber-physical threats. This means that an integrated approach is required. SAFECARE, an integrated cyber-physical security solution, tries to respond to the presented issues within healthcare infrastructures. The challenge is to bring together the most advanced technologies from the physical and cyber security spheres, to achieve a global optimum for systemic security and for the management of combined cyber and physical threats and incidents and their interconnections. Moreover, potential impacts and cascading effects are evaluated through impact propagation models that rely on modular ontologies and a rule-based engine. Indeed, SAFECARE architecture foresees i) a macroblock related to cyber security field, where innovative tools are deployed to monitor network traffic, systems and medical devices; ii) a physical security macroblock, where video management systems are coupled with access control management, building management systems and innovative AI algorithms to detect behavior anomalies; iii) an integration system that collects all the incoming incidents, simulating their potential cascading effects, providing alerts and updated information regarding assets availability.

Keywords: cyber security, defence strategies, impact propagation, integrated security, physical security

Procedia PDF Downloads 166

Authors: Adel Edwar Waheeb Louka

Abstract:

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

Keywords: artificial intelligence, convolutional neural networks, deeplearning, image processing, machine learningSarapin, intraarticular, chronic knee pain, osteoarthritisFNS, trauma, hip, neck femur fracture, minimally invasive surgery

Procedia PDF Downloads 74