Search results for: train routing

Authors: Anita Paddy, Millicent Obodai, Lebbaeus Asamani

Abstract:

The management of sickle cell disease requires a multidisciplinary team for better outcomes. Thus, literature on the application of the biopsychosocial model for the management and explanation of chronic pain in sickle cell disease (SCD) and other chronic diseases abound. However, there is limited research on the use of the biopsychosocial model, together with a spiritual component (biopsychosocial-spiritual model). The study investigated the extent to which healthcare providers utilized the biopsychosocial-spiritual model in the management of chronic pain to improve the quality of life (QoL) of patients with SCD. This study employed the descriptive survey design involving a consecutive sampling of 261 patients with SCD who were between the ages of 18 to 79 years and were accessing hematological services at the Clinical Genetics Department of the Korle Bu Teaching Hospital. These patients willingly consented to participate in the study by appending their signatures. The theory of integrated quality of life, the gate control theory of pain and the biopsychosocial(spiritual) model were tested. An instrument for the biopsychosocial-spiritual model was developed, with a basis from the literature reviewed, while the World Health Organisation Quality of Life BREF (WHOQoLBref) and the spirituality rating scale were adapted and used for data collection. Data were analyzed using descriptive statistics (means, standard deviations, frequencies, and percentages) and partial least square structural equation modeling. The study revealed that healthcare providers had a great leaning toward the biological domain of the model compared to the other domains. Hence, participants’ QoL was not fully improved as suggested by the biopsychosocial(spiritual) model. Again, the QoL and spirituality of patients with SCD were quite high. A significant negative impact of spirituality on QoL was also found. Finally, the biosocial domain of the biopsychosocial-spiritual model was the most significant predictor of QoL. It was recommended that policymakers train healthcare providers to integrate the psychosocial-spiritual component in health services. Also, education on SCD and its resultant impact from the domains of the model should be intensified while health practitioners consider utilizing these components fully in the management of the condition.

Keywords: biopsychosocial (spritual), sickle cell disease, quality of life, healthcare, accra

Procedia PDF Downloads 73

Authors: Jabulani Mtshiya, Yasmin Sultana-Muchindu

Abstract:

Technical and Vocational Education (TVE) is the platform on which developed nations have built their economic foundations, which have led them to attain high standards of living. Zambia has put up educational systems aimed at empowering the citizens and building the economy. Nations such as China, the United States America, and several other European nations are such examples. Despite having programs in Technical and Vocations Education, the Zambian economy still lags, and the industries contributing merger to Gross Domestic Product. This study addresses the significance of Technical and Vocational Education and how it can improve the livelihood of citizens. It addresses aspects of development and productivity and highlights the problems faced by learners in Lusaka and Southern provinces in Zambia. The study employed qualitative research design in data collection and a method of descriptive data analysis was used in order to bring out the description of the prevailing state of affairs in TVE in the perspective of learners. This meant that the respondents indicated their views and thoughts toward TVE. The study collected information through research questionnaires. The findings showed that TVE is regarded important by government and various stakeholders and that it is also regarded important by learners. The findings also showed that stakeholders and society need to pay particular attention to the development of TVE in order to improve the livelihood of citizens and to improve the national economy. Just like any other developed nation that used TVE to develop their industries, Zambia also has the potential to train its youth and to equip them with the necessary skills required for them to contribute positively to the growth of industries and the growth of the economy. Deliberate steps need to be taken by the government and stakeholders to apply and make firm the TVE policies that were laid. At the end of the study recommendations were made; that government should put in the right measures in order to harness the potential at hand. Further on, recommendations were made to carry out this research at the national level and also to conduct it using the quantitative research method, and that government should be consistent to its obligations of funding and maintaining TVE institutions in order for them to be able to operate effectively.

Keywords: education, technical, training, vocational

Procedia PDF Downloads 162

Authors: Rosa Figueroa, Christopher Flores

Abstract:

Text categorization is the problem of assigning a new document to a set of predetermined categories, on the basis of a training set of free-text data that contains documents whose category membership is known. To train a classification model, it is necessary to extract characteristics in the form of tokens that facilitate the learning and classification process. In text categorization, the feature extraction process involves the use of word sequences also known as N-grams. In general, it is expected that documents belonging to the same category share similar features. The Smith-Waterman (SW) algorithm is a dynamic programming algorithm that performs a local sequence alignment in order to determine similar regions between two strings or protein sequences. This work explores the use of SW algorithm as an alternative to feature extraction in text categorization. The dataset used for this purpose, contains 2,610 annotated documents with the classes Obese/Non-Obese. This dataset was represented in a matrix form using the Bag of Word approach. The score selected to represent the occurrence of the tokens in each document was the term frequency-inverse document frequency (TF-IDF). In order to extract features for classification, four experiments were conducted: the first experiment used SW to extract features, the second one used unigrams (single word), the third one used bigrams (two word sequence) and the last experiment used a combination of unigrams and bigrams to extract features for classification. To test the effectiveness of the extracted feature set for the four experiments, a Support Vector Machine (SVM) classifier was tuned using 20% of the dataset. The remaining 80% of the dataset together with 5-Fold Cross Validation were used to evaluate and compare the performance of the four experiments of feature extraction. Results from the tuning process suggest that SW performs better than the N-gram based feature extraction. These results were confirmed by using the remaining 80% of the dataset, where SW performed the best (accuracy = 97.10%, weighted average F-measure = 97.07%). The second best was obtained by the combination of unigrams-bigrams (accuracy = 96.04, weighted average F-measure = 95.97) closely followed by the bigrams (accuracy = 94.56%, weighted average F-measure = 94.46%) and finally unigrams (accuracy = 92.96%, weighted average F-measure = 92.90%).

Keywords: comorbidities, machine learning, obesity, Smith-Waterman algorithm

Procedia PDF Downloads 297

Authors: Achim Kampker, Heiner H. Heimes, Mathias Ordung, Christoph Lienemann, Ansgar Hollah, Nemanja Sarovic

Abstract:

Electric cars with their fast innovation cycles and their disruptive character offer a high degree of freedom regarding innovative design for remanufacturing. Remanufacturing increases not only the resource but also the economic efficiency by a prolonged product life time. The reduced power train wear of electric cars combined with high manufacturing costs for batteries allow new business models and even second life applications. Modular and intermountable designed battery packs enable the replacement of defective or outdated battery cells, allow additional cost savings and a prolongation of life time. This paper discusses opportunities for future remanufacturing value chains of electric cars and their battery components and how to address their potentials with elaborate designs. Based on a brief overview of implemented remanufacturing structures in different industries, opportunities of transferability are evaluated. In addition to an analysis of current and upcoming challenges, promising perspectives for a sustainable electric car circular economy enabled by design for remanufacturing are deduced. Two mathematical models describe the feasibility of pursuing a circular economy of lithium ion batteries and evaluate remanufacturing in terms of sustainability and economic efficiency. Taking into consideration not only labor and material cost but also capital costs for equipment and factory facilities to support the remanufacturing process, cost benefit analysis prognosticate that a remanufacturing battery can be produced more cost-efficiently. The ecological benefits were calculated on a broad database from different research projects which focus on the recycling, the second use and the assembly of lithium ion batteries. The results of this calculations show a significant improvement by remanufacturing in all relevant factors especially in the consumption of resources and greenhouse warming potential. Exemplarily suitable design guidelines for future remanufacturing lithium ion batteries, which consider modularity, interfaces and disassembly, are used to illustrate the findings. For one guideline, potential cost improvements were calculated and upcoming challenges are pointed out.

Keywords: circular economy, electric mobility, lithium ion batteries, remanufacturing

Procedia PDF Downloads 358

Authors: Nafiseh Mirabdolhosseini, Jerry Zhou, Vincent Ho

Abstract:

The dramatic global rise in acid reflux has also led to oesophageal adenocarcinoma (OAC) becoming the fastest-growing cancer in developed countries. While gastroscopy with biopsy is used to diagnose OAC patients, this labour-intensive and expensive process is not suitable for population screening. This study aims to design, develop, and implement a minimally invasive system to capture optical data of the distal oesophagus for rapid screening of potential abnormalities. To develop the system and understand user requirements, a user-centric approach was employed by utilising co-design strategies. Target users’ segments were identified, and 38 patients and 14 health providers were interviewed. Next, the technical requirements were developed based on consultations with the industry. A minimally invasive optical system was designed and developed considering patient comfort. This system consists of the sensing catheter, controller unit, and analysis program. Its procedure only takes 10 minutes to perform and does not require cleaning afterward since it has a single-use catheter. A prototype system was evaluated for safety and efficacy for both laboratory and clinical performance. This prototype performed successfully when submerged in simulated gastric fluid without showing evidence of erosion after 24 hours. The system effectively recorded a video of the mid-distal oesophagus of a healthy volunteer (34-year-old male). The recorded images were used to develop an automated program to identify abnormalities in the distal oesophagus. Further data from a larger clinical study will be used to train the automated program. This system allows for quick visual assessment of the lower oesophagus in primary care settings and can serve as a screening tool for oesophageal adenocarcinoma. In addition, this system is able to be coupled with 24hr ambulatory pH monitoring to better correlate oesophageal physiological changes with reflux symptoms. It also can provide additional information on lower oesophageal sphincter functions such as opening times and bolus retention.

Keywords: endoscopy, MedTech, oesophageal adenocarcinoma, optical system, screening tool

Procedia PDF Downloads 88

Authors: Nolitha L. Skenjana, Bongani P. Kubheka, Maxwell A. Poswal

Abstract:

The Eastern Cape Province of South Africa is characterized by subsistence farming, which is mostly distributed in the rural areas of the province. It is estimated that cereal crops such as maize and sorghum, and vegetables such as cabbage are grown in more than 400.000 rural households, with maize being the most dominant crop. However, compared to commercial agriculture, small-scale farmers receive minimal support from research and development, limited technology transfer on the latest production practices and systems and have poor production infrastructure and equipment. Similarly, there is limited farmers' appreciation on best practices in insect pest management and control. The paper presents findings from the primary literature and personal observations on insect pest management practices of small-scale farmers in the province. Inferences from literature and personal experiences in the production areas have led to a number of deductions regarding the level of exposure and extent of vulnerability. Farmers' pest management practices, which included not controlling at all though there is a pest problem, resulted in their crop stands to be more vulnerable to pest attacks. This became more evident with the recent brown locust, African armyworm, and Fall armyworm outbreaks, and with the incidences of opportunistic phytophagous insects previously collected on wild hosts only, found causing serious damages on crops. In most of these occurrences, damage to crops resulted in low or no yield. Improvements on farmers' reaction and response to pest problems were only observed in areas where focused awareness campaigns and trainings on specific pests and their management techniques were done. This then calls for a concerted effort from all role players in the sphere of small-scale crop production, to train and equip farmers with relevant skills, and provide them with information on affordable and climate-smart strategies and technologies in order to create a state of preparedness. This is necessary for the prevention of substantial crop losses that may exacerbate food insecurity in the province.

Keywords: Eastern Cape Province, small-scale farmers, insect pest management, vulnerability

Procedia PDF Downloads 140

Authors: Heewon Jeong, Seongpyo Kim, Joon Ha Kim

Abstract:

Over the last few tens of years, the coastal regions of Korea have experienced red tide algal blooms, which are harmful and toxic to both humans and marine organisms due to their potential threat. It was accelerated owing to eutrophication by human activities, certain oceanic processes, and climate change. Previous studies have tried to monitoring and predicting the algae concentration of the ocean with the bio-optical algorithms applied to color images of the satellite. However, the accurate estimation of algal blooms remains problems to challenges because of the complexity of coastal waters. Therefore, this study suggests a new method to identify the concentration of red tide algal bloom from images of geostationary ocean color imager (GOCI) which are representing the water environment of the sea in Korea. The method employed GOCI images, which took the water leaving radiances centered at 443nm, 490nm and 660nm respectively, as well as observed weather data (i.e., humidity, temperature and atmospheric pressure) for the database to apply optical characteristics of algae and train deep learning algorithm. Convolution neural network (CNN) was used to extract the significant features from the images. And then artificial neural network (ANN) was used to estimate the concentration of algae from the extracted features. For training of the deep learning model, backpropagation learning strategy is developed. The established methods were tested and compared with the performances of GOCI data processing system (GDPS), which is based on standard image processing algorithms and optical algorithms. The model had better performance to estimate algae concentration than the GDPS which is impossible to estimate greater than 5mg/m³. Thus, deep learning model trained successfully to assess algae concentration in spite of the complexity of water environment. Furthermore, the results of this system and methodology can be used to improve the performances of remote sensing. Acknowledgement: This work was supported by the 'Climate Technology Development and Application' research project (#K07731) through a grant provided by GIST in 2017.

Keywords: deep learning, algae concentration, remote sensing, satellite

Procedia PDF Downloads 183

Authors: Xiaobo Liu, Rakesh Mishra, Yun Zhang

Abstract:

Continuous monitoring of forest canopy height with large coverage is essential for obtaining forest carbon stocks and emissions, quantifying biomass estimation, analyzing vegetation coverage, and determining biodiversity. LiDAR can be used to collect accurate woody vegetation structure such as canopy height. However, LiDAR’s coverage is usually limited because of its high cost and limited maneuverability, which constrains its use for dynamic and large area forest canopy monitoring. On the other hand, optical satellite images, like Sentinel-2, have the ability to cover large forest areas with a high repeat rate, but they do not have height information. Hence, exploring the solution of integrating LiDAR data and Sentinel-2 images to enlarge the coverage of forest canopy height prediction and increase the prediction repeat rate has been an active research topic in the environmental remote sensing community. In this study, we explore the potential of training a Random Forest Regression (RFR) model and a Convolutional Neural Network (CNN) model, respectively, to develop two predictive models for predicting and validating the forest canopy height of the Acadia Forest in New Brunswick, Canada, with a 10m ground sampling distance (GSD), for the year 2018 and 2021. Two 10m airborne LiDAR-derived canopy height models, one for 2018 and one for 2021, are used as ground truth to train and validate the RFR and CNN predictive models. To evaluate the prediction performance of the trained RFR and CNN models, two new predicted canopy height maps (CHMs), one for 2018 and one for 2021, are generated using the trained RFR and CNN models and 10m Sentinel-2 images of 2018 and 2021, respectively. The two 10m predicted CHMs from Sentinel-2 images are then compared with the two 10m airborne LiDAR-derived canopy height models for accuracy assessment. The validation results show that the mean absolute error (MAE) for year 2018 of the RFR model is 2.93m, CNN model is 1.71m; while the MAE for year 2021 of the RFR model is 3.35m, and the CNN model is 3.78m. These demonstrate the feasibility of using the RFR and CNN models developed in this research for predicting large-coverage forest canopy height at 10m spatial resolution and a high revisit rate.

Keywords: remote sensing, forest canopy height, LiDAR, Sentinel-2, artificial intelligence, random forest regression, convolutional neural network

Procedia PDF Downloads 92

Authors: H. Jung, N. Kim, B. Kang, J. Choe

Abstract:

History matching is a crucial procedure for predicting reservoir performances and making future decisions. However, it is difficult due to uncertainties of initial reservoir models. Therefore, it is important to have reliable initial models for successful history matching of highly heterogeneous reservoirs such as channel reservoirs. In this paper, we proposed a novel scheme for regenerating geological models using support vector machine (SVM) and principal component analysis (PCA). First, we perform PCA for figuring out main geological characteristics of models. Through the procedure, permeability values of each model are transformed to new parameters by principal components, which have eigenvalues of large magnitude. Secondly, the parameters are projected into two-dimensional plane by multi-dimensional scaling (MDS) based on Euclidean distances. Finally, we train an SVM classifier using 20% models which show the most similar or dissimilar well oil production rates (WOPR) with the true values (10% for each). Then, the other 80% models are classified by trained SVM. We select models on side of low WOPR errors. One hundred channel reservoir models are initially generated by single normal equation simulation. By repeating the classification process, we can select models which have similar geological trend with the true reservoir model. The average field of the selected models is utilized as a probability map for regeneration. Newly generated models can preserve correct channel features and exclude wrong geological properties maintaining suitable uncertainty ranges. History matching with the initial models cannot provide trustworthy results. It fails to find out correct geological features of the true model. However, history matching with the regenerated ensemble offers reliable characterization results by figuring out proper channel trend. Furthermore, it gives dependable prediction of future performances with reduced uncertainties. We propose a novel classification scheme which integrates PCA, MDS, and SVM for regenerating reservoir models. The scheme can easily sort out reliable models which have similar channel trend with the reference in lowered dimension space.

Keywords: history matching, principal component analysis, reservoir modelling, support vector machine

Procedia PDF Downloads 160

Authors: Dávid Kovács, Bálint Csanády, Dániel Boros, Iván Ivkovic, Lóránt Nagy, Dalma Tóth-Lakits, László Márkus, András Lukács

Abstract:

The modeling practice of financial instruments has seen significant change over the last decade due to the recognition of time-dependent and stochastically changing correlations among the market prices or the prices and market characteristics. To represent this phenomenon, the Stochastic Correlation Process (SCP) has come to the fore in the joint modeling of prices, offering a more nuanced description of their interdependence. This approach has allowed for the attainment of realistic tail dependencies, highlighting that prices tend to synchronize more during intense or volatile trading periods, resulting in stronger correlations. Evidence in statistical literature suggests that, similarly to the volatility, the SCP of certain stock prices follows rough paths, which can be described using fractional differential equations. However, estimating parameters for these equations often involves complex and computation-intensive algorithms, creating a necessity for alternative solutions. In this regard, the Fractional Ornstein-Uhlenbeck (fOU) process from the family of fractional processes offers a promising path. We can effectively describe the rough SCP by utilizing certain transformations of the fOU. We employed neural networks to understand the behavior of these processes. We had to develop a fast algorithm to generate a valid and suitably large sample from the appropriate process to train the network. With an extensive training set, the neural network can estimate the process parameters accurately and efficiently. Although the initial focus was the fOU, the resulting model displayed broader applicability, thus paving the way for further investigation of other processes in the realm of financial mathematics. The utility of SCP extends beyond its immediate application. It also serves as a springboard for a deeper exploration of fractional processes and for extending existing models that use ordinary Wiener processes to fractional scenarios. In essence, deploying both SCP and fractional processes in financial models provides new, more accurate ways to depict market dynamics.

Keywords: fractional Ornstein-Uhlenbeck process, fractional stochastic processes, Heston model, neural networks, stochastic correlation, stochastic differential equations, stochastic volatility

Procedia PDF Downloads 118

Authors: G. Calleja-Rodriguez, N. Jimenez-Redondo, J. J. Peralta Escalante

Abstract:

This research work aims at developing a solution to take advantage of the potential energy saving related to occupants behaviour estimated in between 5-30 % according to existing studies. For that purpose, the following methodology has been followed: 1) literature review and gap analysis, 2) define concept and functional requirements, 3) evaluation and feedback by experts. As result, the concept for a tool-box that implements continuous behavior change interventions named as engagement methods and based on increasing energy literacy, increasing energy visibility, using bonus system, etc. has been defined. These engagement methods are deployed through a set of ICT tools: Building Automation and Control System (BACS) add-ons services installed in buildings and Users Apps installed in smartphones, smart-TVs or dashboards. The tool-box called eTEACHER identifies energy conservation measures (ECM) based on energy behavioral change through a what-if analysis that collects information about the building and its users (comfort feedback, behavior, etc.) and carry out cost-effective calculations to provide outputs such us efficient control settings of building systems. This information is processed and showed in an attractive way as tailored advice to the energy end-users. Therefore, eTEACHER goal is to change the behavior of building´s energy users towards energy efficiency, comfort and better health conditions by deploying customized ICT-based interventions taking into account building typology (schools, residential, offices, health care centres, etc.), users profile (occupants, owners, facility managers, employers, etc.) as well as cultural and demographic factors. One of the main findings of this work is the common failure when technological interventions on behavioural change are done to not consult, train and support users regarding technological changes leading to poor performance in practices. As conclusion, a strong need to carry out social studies to identify relevant behavioural issues and to identify effective pro-evironmental behavioral change strategies has been identified.

Keywords: energy saving, behavioral bhange, building users, engagement methods, energy conservation measures

Procedia PDF Downloads 170

Authors: Yonggyu Lee, Byungkyu Jung, Bom Yoon, Jongil Yoon

Abstract:

MR-based simulation devices have been recently used in various fields such as entertainment, medicine, manufacturing, and education. Different simulation devices are also being developed for military equipment training. This is to address the concerns regarding safety accidents as well as cost issues associated with training with expensive equipment. An important aspect of developing simulation devices to replicate military training is that trainees experience the same effect as training with real devices. In this study, the criteria for performance evaluation are established to compare the training effect of an MR-based simulation device to that of an actual device. K-9 Self-propelled artillery (SPA) operators are selected as training subjects. First, MR-based software is developed to simulate the training ground and training scenarios currently used for training SPA operators in South Korea. Hardware that replicates the interior of SPA is designed, and a simulation device that is linked to the software is developed. Second, criteria are established to evaluate the simulation device based on real-life training scenarios. A total of nine performance evaluation criteria were selected based on the actual SPA operation training scenarios. Evaluation items were selected to evaluate whether the simulation device was designed such that trainees would experience the same effect as training in the field with a real SPA. To eval-uate the level of replication by the simulation device of the actual training environments (driving and passing through trenches, pools, protrusions, vertical obstacles, and slopes) and driving conditions (rapid steering, rapid accelerating, and rapid braking) as per the training scenarios, tests were performed under the actual training conditions and in the simulation device, followed by the comparison of the results. In addition, the level of noise felt by operators during training was also selected as an evaluation criterion. Due to the nature of the simulation device, there may be data latency between HW and SW. If the la-tency in data transmission is significant, the VR image information delivered to trainees as they maneuver HW might not be consistent. This latency in data transmission was also selected as an evaluation criterion to improve the effectiveness of the training. Through this study, the key evaluation metrics were selected to achieve the same training effect as training with real equipment in a training ground during the develop-ment of the simulation device for military equipment training.

Keywords: K-9 self-propelled artillery, mixed reality, simulation device, synchronization

Procedia PDF Downloads 66

Authors: Igor A. Bessmertny, Bidru C. Enkomaryam

Abstract:

Active involving patients and communities in health decisions can improve both people’s health and the healthcare system. Adopting artificial intelligence can lead to more accurate and complete patient record management. This review aims to identify the current state of researches conducted using artificial intelligence techniques to improve patient engagement and wellbeing, medical domains used in patient engagement context, and lastly, to assess opportunities and challenges for patient engagement in the wellness process. A search of peer-reviewed publications, reviews, conceptual analyses, white papers, author’s manuscripts and theses was undertaken. English language literature published in 2013– 2022 period and publications, report and guidelines of World Health Organization (WHO) were also assessed. About 281 papers were retrieved. Duplicate papers in the databases were removed. After application of the inclusion and exclusion criteria, 41 papers were included to the analysis. Patient counseling in preventing adverse drug events, in doctor-patient risk communication, surgical, drug development, mental healthcare, hypertension & diabetes, metabolic syndrome and non-communicable chronic diseases are implementation areas in healthcare where patient engagement can be implemented using artificial intelligence, particularly machine learning and deep learning techniques and tools. The five groups of factors that potentially affecting patient engagement in safety are related to: patient, health conditions, health care professionals, tasks and health care setting. Active involvement of patients and families can help accelerate the implementation of healthcare safety initiatives. In sub-Saharan Africa, using digital technologies like artificial intelligence in patient engagement context is low due to poor level of technological development and deployment. The opportunities and challenges available to implement patient engagement strategies vary greatly from country to country and from region to region. Thus, further investigation will be focused on methods and tools using the potential of artificial intelligence to support more simplified care that might be improve communication with patients and train health care professionals.

Keywords: artificial intelligence, patient engagement, machine learning, patient involvement

Procedia PDF Downloads 76

Authors: Anwesha Pathak

Abstract:

With the increasing number of satellites and space debris, space traffic management (STM) becomes crucial. AI can aid in STM by predicting and preventing potential collisions, optimizing satellite trajectories, and managing orbital slots. Governance frameworks need to address the integration of AI algorithms in STM to ensure safe and sustainable satellite activities. AI and governance play significant roles in the context of satellite activities in space. Artificial intelligence (AI) technologies, such as machine learning and computer vision, can be utilized to process vast amounts of data received from satellites. AI algorithms can analyse satellite imagery, detect patterns, and extract valuable information for applications like weather forecasting, urban planning, agriculture, disaster management, and environmental monitoring. AI can assist in automating and optimizing satellite operations. Autonomous decision-making systems can be developed using AI to handle routine tasks like orbit control, collision avoidance, and antenna pointing. These systems can improve efficiency, reduce human error, and enable real-time responsiveness in satellite operations. AI technologies can be leveraged to enhance the security of satellite systems. AI algorithms can analyze satellite telemetry data to detect anomalies, identify potential cyber threats, and mitigate vulnerabilities. Governance frameworks should encompass regulations and standards for securing satellite systems against cyberattacks and ensuring data privacy. AI can optimize resource allocation and utilization in satellite constellations. By analyzing user demands, traffic patterns, and satellite performance data, AI algorithms can dynamically adjust the deployment and routing of satellites to maximize coverage and minimize latency. Governance frameworks need to address fair and efficient resource allocation among satellite operators to avoid monopolistic practices. Satellite activities involve multiple countries and organizations. Governance frameworks should encourage international cooperation, information sharing, and standardization to address common challenges, ensure interoperability, and prevent conflicts. AI can facilitate cross-border collaborations by providing data analytics and decision support tools for shared satellite missions and data sharing initiatives. AI and governance are critical aspects of satellite activities in space. They enable efficient and secure operations, ensure responsible and ethical use of AI technologies, and promote international cooperation for the benefit of all stakeholders involved in the satellite industry.

Keywords: satellite, space debris, traffic, threats, cyber security.

Procedia PDF Downloads 76

Authors: M. Roberts Masillamani

Abstract:

There is a phenomenal difference, as regard to the teaching methodology adopted at the urban and the rural area colleges. However, bright and talented student may be from rural back ground even. But there is huge dearth of the digitization in the rural areas and lesser developed countries. Today’s students need new skills to compete and successful in the future. Education should be combination of practical, intellectual, and social skills. What does this mean for rural classrooms and how can it be achieved. Rural colleges are not able to hire the best resources, since the best teacher’s aim is to move towards the city. If city is provided everywhere, then there will be no rural area. This is possible by forming university clusters (UC). The University cluster is a group of renowned and accredited universities coming together to bridge this dearth. The UC will deliver the live lectures and allow the students’ from remote areas to actively participate in the classroom. This paper tries to present a plan of action of providing a better live classroom teaching and learning system from the city to the rural and the lesser developed countries. This paper titled “University Clusters using ICT for teaching and learning” provides a true concept of opening live digital classroom windows for rural colleges, where resources are not available, thus reducing the digital divide. This is different from pod casting a lecture or distance learning and eLearning. The live lecture can be streamed through digital equipment to another classroom. The rural students can collaborate with their peers and critiques, be assessed, collect information, acquire different techniques in assessment and learning process. This system will benefit rural students and teachers and develop socio economic status. This will also will increase the degree of confidence of the Rural students and teachers. Thus bringing about the concept of ‘Train the Trainee’ in reality. An educational university cloud for each cluster will be built remote infrastructure facilities (RIF) for the above program. The users may be informed, about the available lecture schedules, through the RIF service. RIF with an educational cloud can be set by the universities under one cluster. This paper talks a little more about University clusters and the methodology to be adopted as well as some extended features like, tutorial classes, library grids, remote laboratory login, research and development.

Keywords: lesser developed countries, digital divide, digital learning, education, e-learning, ICT, library grids, live classroom windows, RIF, rural, university clusters and urban

Procedia PDF Downloads 471

Authors: Besma Khalfoun

Abstract:

In many current Human Activity Recognition (HAR) applications, users' data is frequently shared and centrally stored by third parties, posing a significant privacy risk. This practice makes these entities attractive targets for extracting sensitive information about users, including their identity, health status, and location, thereby directly violating users' privacy. To tackle the issue of centralized data storage, a relatively recent paradigm known as federated learning has emerged. In this approach, users' raw data remains on their smartphones, where they train the HAR model locally. However, users still share updates of their local models originating from raw data. These updates are vulnerable to several attacks designed to extract sensitive information, such as determining whether a data sample is used in the training process, recovering the training data with inversion attacks, or inferring a specific attribute or property from the training data. In this paper, we first introduce PUR-Attack, a parameter-based user re-identification attack developed for HAR applications within a federated learning setting. It involves associating anonymous model updates (i.e., local models' weights or parameters) with the originating user's identity using background knowledge. PUR-Attack relies on a simple yet effective machine learning classifier and produces promising results. Specifically, we have found that by considering the weights of a given layer in a HAR model, we can uniquely re-identify users with an attack success rate of almost 100%. This result holds when considering a small attack training set and various data splitting strategies in the HAR model training. Thus, it is crucial to investigate protection methods to mitigate this privacy threat. Along this path, we propose SAFER, a privacy-preserving mechanism based on adaptive local differential privacy. Before sharing the model updates with the FL server, SAFER adds the optimal noise based on the re-identification risk assessment. Our approach can achieve a promising tradeoff between privacy, in terms of reducing re-identification risk, and utility, in terms of maintaining acceptable accuracy for the HAR model.

Keywords: federated learning, privacy risk assessment, re-identification risk, privacy preserving mechanisms, local differential privacy, human activity recognition

Procedia PDF Downloads 11

Authors: Kayne Uriel K. Rodrigo, Jerriane Hillary Heart S. Marcial, Samuel C. Brillo

Abstract:

Rice leaf diseases significantly impact yield production in rice-dependent countries, affecting their agricultural sectors. As part of precision agriculture, early and accurate detection of these diseases is crucial for effective mitigation practices and minimizing crop losses. Hence, this study proposes an enhancement to the Convolutional Neural Network (CNN), a widely-used method for Rice Leaf Disease Image Classification, by incorporating MobileViTV2—a recently advanced architecture that combines CNN and Vision Transformer models while maintaining fewer parameters, making it suitable for broader deployment on edge devices. Our methodology utilizes a publicly available rice disease image dataset from Kaggle, which was validated by a university structural biologist following the guidelines provided by the Philippine Rice Institute (PhilRice). Modifications to the dataset include renaming certain disease categories and augmenting the rice leaf image data through rotation, scaling, and flipping. The enhanced dataset was then used to train the MobileViTV2 model using the Timm library. The results of our approach are as follows: the model achieved notable performance, with 98% accuracy in both training and validation, 6% training and validation loss, and a Receiver Operating Characteristic (ROC) curve ranging from 95% to 100% for each label. Additionally, the F1 score was 97%. These metrics demonstrate a significant improvement compared to a conventional CNN-based approach, which, in a previous 2022 study, achieved only 78% accuracy after using 5 convolutional layers and 2 dense layers. Thus, it can be concluded that MobileViTV2, with its fewer parameters, outperforms traditional CNN models, particularly when applied to Rice Leaf Disease Image Identification. For future work, we recommend extending this model to include datasets validated by international rice experts and broadening the scope to accommodate biotic factors such as rice pest classification, as well as abiotic stressors such as climate, soil quality, and geographic information, which could improve the accuracy of disease prediction.

Keywords: convolutional neural network, MobileViTV2, rice leaf disease, precision agriculture, image classification, vision transformer

Procedia PDF Downloads 22

Authors: Bheki R. Mngomezulu

Abstract:

The present paper cogently and systematically traces the history of Umkhonto Wesizwe (MK) and identifies its important role in shaping South Africa’s post-apartheid foreign policy and international relations under black leadership. It provides the political and historical contexts within which we can interpret and better understand South Africa’s controversial ‘Quiet Diplomacy’ approach to Zimbabwe’s endemic political and economic crises, which have dragged for too long. On 16 December 1961, the African National Congress (ANC) officially launched the MK as its military wing. The main aim was to train liberation fighters outside South Africa who would return into the country to topple the apartheid regime. Subsequently, the ANC established links with various countries across Africa and the globe in order to solicit arms, financial resources and military training for its recruits into the MK. Drawing from archival research and empirical data obtained through oral interviews that were conducted with some of the former MK cadres, this paper demonstrates how the ANC forged relations with a number of countries that were like-minded in order to ensure that its dream of removing the apartheid government became a reality. The findings reveal that South Africa’s foreign policy posture and international relations after the demise of apartheid in 1994 built on these relations. As such, even former and current socialist countries that were frowned upon by the Western world became post-apartheid South Africa’s international partners. These include countries such as Cuba and China, among others. Even countries that were not recognized by the Western world as independent states received good reception in post-apartheid South Africa’s foreign policy agenda. One of these countries is Palestine. Within Africa, countries with questionable human rights records such as Nigeria and Zimbabwe were accommodated in South Africa’s foreign policy agenda after 1994. Drawing from this history, the paper concludes that it would be difficult to fully understand and appreciate South Africa’s foreign policy direction and international relations after 1994 without bringing the history and the politics of the MK into the equation. Therefore, the paper proposes that the utilitarian role of history should never be undermined in the analysis of a country’s foreign policy direction and international relations. Umkhonto Wesizwe and South Africa are used as examples to demonstrate how such a link could be drawn through archival and empirical evidence.

Keywords: African National Congress, apartheid, foreign policy, international relations

Procedia PDF Downloads 185

Authors: Shreya Goyal

Abstract:

The study aims to build user-friendly software to understand patient arrival patterns and compute the number of potential patients who will visit a particular health facility for a given period by using a machine learning algorithm. The underlying machine learning algorithm used in this study is the Support Vector Machine (SVM). Accurate prediction of patient arrival allows hospitals to operate more effectively, providing timely and efficient care while optimizing resources and improving patient experience. It allows for better allocation of staff, equipment, and other resources. If there's a projected surge in patients, additional staff or resources can be allocated to handle the influx, preventing bottlenecks or delays in care. Understanding patient arrival patterns can also help streamline processes to minimize waiting times for patients and ensure timely access to care for patients in need. Another big advantage of using this software is adhering to strict data protection regulations such as the Health Insurance Portability and Accountability Act (HIPAA) in the United States as the hospital will not have to share the data with any third party or upload it to the cloud because the software can read data locally from the machine. The data needs to be arranged in. a particular format and the software will be able to read the data and provide meaningful output. Using software that operates locally can facilitate compliance with these regulations by minimizing data exposure. Keeping patient data within the hospital's local systems reduces the risk of unauthorized access or breaches associated with transmitting data over networks or storing it in external servers. This can help maintain the confidentiality and integrity of sensitive patient information. Historical patient data is used in this study. The input variables used to train the model include patient age, time of day, day of the week, seasonal variations, and local events. The algorithm uses a Supervised learning method to optimize the objective function and find the global minima. The algorithm stores the values of the local minima after each iteration and at the end compares all the local minima to find the global minima. The strength of this study is the transfer function used to calculate the number of patients. The model has an output accuracy of >95%. The method proposed in this study could be used for better management planning of personnel and medical resources.

Keywords: machine learning, SVM, HIPAA, data

Procedia PDF Downloads 65

Authors: Shamie Kumar

Abstract:

Background: Radiologists and radiographers (RR) need to educate themselves and their colleagues to ensure that AI is integrated safely, useful, and in a meaningful way with the direction it always benefits the patients. AI education and training are fundamental to the way RR work and interact with it, such that they feel confident using it as part of their clinical practice in a way they understand it. Methodology: This exploratory research will outline the current educational and training gaps for radiographers and radiologists in AI radiology diagnostics. It will review the status, skills, challenges of educating and teaching. Understanding the use of artificial intelligence within daily clinical practice, why it is fundamental, and justification on why learning about AI is essential for wider adoption. Results: The current knowledge among RR is very sparse, country dependent, and with radiologists being the majority of the end-users for AI, their targeted training and learning AI opportunities surpass the ones available to radiographers. There are many papers that suggest there is a lack of knowledge, understanding, and training of AI in radiology amongst RR, and because of this, they are unable to comprehend exactly how AI works, integrates, benefits of using it, and its limitations. There is an indication they wish to receive specific training; however, both professions need to actively engage in learning about it and develop the skills that enable them to effectively use it. There is expected variability amongst the profession on their degree of commitment to AI as most don’t understand its value; this only adds to the need to train and educate RR. Currently, there is little AI teaching in either undergraduate or postgraduate study programs, and it is not readily available. In addition to this, there are other training programs, courses, workshops, and seminars available; most of these are short and one session rather than a continuation of learning which cover a basic understanding of AI and peripheral topics such as ethics, legal, and potential of AI. There appears to be an obvious gap between the content of what the training program offers and what the RR needs and wants to learn. Due to this, there is a risk of ineffective learning outcomes and attendees feeling a lack of clarity and depth of understanding of the practicality of using AI in a clinical environment. Conclusion: Education, training, and courses need to have defined learning outcomes with relevant concepts, ensuring theory and practice are taught as a continuation of the learning process based on use cases specific to a clinical working environment. Undergraduate and postgraduate courses should be developed robustly, ensuring the delivery of it is with expertise within that field; in addition, training and other programs should be delivered as a way of continued professional development and aligned with accredited institutions for a degree of quality assurance.

Keywords: artificial intelligence, training, radiology, education, learning

Procedia PDF Downloads 85

Authors: Marlene Rosa, Susana Lopes

Abstract:

Board games have been used for different purposes in geriatric care, demonstrating good results for health in general. However, there is not a conceptual framework that can help professionals and researchers in this area to design intervention programs or to think about future studies in this area. The aim of this study was to provide a pilot collection of board games’ serious purposes in geriatric care, using a WHO framework for health and disability. Study cases were developed in seven geriatric residential institutions from the center region in Portugal that are included in AGILAB program. The AGILAB program is a serious game-based method to train and spread out the implementation of board games in geriatric care. Each institution provides 2-hours/week of experiences using TATI Hand Game for serious purposes and then fulfill questions about a study-case (player characteristics; explain changes in players health according to this game experience). Two independent researchers read the information and classified it according to the International Classification for Functioning and Disability (ICF) categories. Any discrepancy was solved in a consensus meeting. Results indicate an important variability in body functions and structures: specific mental functions (e.g., b140 Attention functions, b144 Memory functions), b156 Perceptual functions, b2 sensory functions and pain (e.g., b230 Hearing functions; b265 Touch function; b280 Sensation of pain), b7 neuromusculoskeletal and movement-related functions (e.g., b730 Muscle power functions; b760 Control of voluntary movement functions; b710 Mobility of joint functions). Less variability was found in activities and participation domains, such as purposeful sensory experiences (d110-d129) (e.g., d115 Listening), communication (d3), d710 basic interpersonal interactions, d920 recreation and leisure (d9200 Play; d9205 Socializing). Concluding, this framework designed from a brief gamed-based experience includes mental, perceptual, sensory, neuromusculoskeletal, and movement-related functions and participation in sensory, communication, and leisure domains. More studies, including different experiences and a high number of users, should be developed to provide a more comprehensive ICF framework for game-based experiences in geriatric care.

Keywords: board game, aging, framework, experience

Procedia PDF Downloads 126

Authors: Habtamu Ayenew Asegie

Abstract:

Wealth, as opposed to income or consumption, implies a more stable and permanent status. Due to natural and human-made difficulties, households' economies will be diminished, and their well-being will fall into trouble. Hence, governments and humanitarian agencies offer considerable resources for poverty and malnutrition reduction efforts. One key factor in the effectiveness of such efforts is the accuracy with which low-income or poor populations can be identified. As a result, this study aims to predict a household’s wealth status using ensemble Machine learning (ML) algorithms. In this study, design science research methodology (DSRM) is employed, and four ML algorithms, Random Forest (RF), Adaptive Boosting (AdaBoost), Light Gradient Boosted Machine (LightGBM), and Extreme Gradient Boosting (XGBoost), have been used to train models. The Ethiopian Demographic and Health Survey (EDHS) dataset is accessed for this purpose from the Central Statistical Agency (CSA)'s database. Various data pre-processing techniques were employed, and the model training has been conducted using the scikit learn Python library functions. Model evaluation is executed using various metrics like Accuracy, Precision, Recall, F1-score, area under curve-the receiver operating characteristics (AUC-ROC), and subjective evaluations of domain experts. An optimal subset of hyper-parameters for the algorithms was selected through the grid search function for the best prediction. The RF model has performed better than the rest of the algorithms by achieving an accuracy of 96.06% and is better suited as a solution model for our purpose. Following RF, LightGBM, XGBoost, and AdaBoost algorithms have an accuracy of 91.53%, 88.44%, and 58.55%, respectively. The findings suggest that some of the features like ‘Age of household head’, ‘Total children ever born’ in a family, ‘Main roof material’ of their house, ‘Region’ they lived in, whether a household uses ‘Electricity’ or not, and ‘Type of toilet facility’ of a household are determinant factors to be a focal point for economic policymakers. The determinant risk factors, extracted rules, and designed artifact achieved 82.28% of the domain expert’s evaluation. Overall, the study shows ML techniques are effective in predicting the wealth status of households.

Keywords: ensemble machine learning, households wealth status, predictive model, wealth status prediction

Procedia PDF Downloads 38

Authors: Matt Stichter

Abstract:

Moral development requires learning from one’s failures, but that turnsout to be especially challenging when dealing with moral failures. The distress prompted by moral failure can cause responses ofdefensiveness or disengagement rather than attempts to make amends and work on self-change. The most potentially distressing response to moral failure is a shame. However, there appears to be two different senses of “shame” that are conflated in the literature, depending on whether the failure is appraised as the result of a global and unalterable self-defect, or a local and alterable self-defect. One of these forms of shame does prompt self-improvement in response to moral failure. This occurs if one views the failure as indicating only a specific (local) defect in one’s identity, where that’s something repairable, rather than asanoverall(orglobal)defectinyouridentity that can’t be fixed. So, if the whole of one’s identity as a morally good person isn’t being called into question, but only a part, then that is something one could work on to improve. Shame, in this sense, provides motivation for self-improvement to fix this part oftheselfinthe long run, and this would be important for moral development. One factor that looks to affect these different self-attributions in the wake of moral failure can be found in mindset theory, as reactions to moral failure in these two forms of shame are similar to how those with a fixed or growth mindset of their own abilities, such as intelligence, react to failure. People fall along a continuum with respect to how they view abilities – it is more of a fixed entity that you cannot do much to change, or it is malleable such that you can train to improve it. These two mindsets, ‘fixed’ versus ‘growth’, have different consequences for how we react to failure – a fixed mindset leads to maladaptive responses because of feelings of helplessness to do better; whereas a growth mindset leads to adaptive responses where a person puts forth effort to learn how to act better the next time. Here we can see the parallels between a fixed mindset of one’s own (im)morality, as the way people respond to shame when viewed as indicating a global and unalterable self-defect parallels the reactions people have to failure when they have a fixed mindset. In addition, it looks like there may be a similar structure to pride. Pride is, like shame, a self-conscious emotion that arises from internal attributions about the self as being the cause of some event. There are also paradoxical results from research on pride, where pride was found to motivate pro-social behavior in some cases but aggression in other cases. Research suggests that there may be two forms of pride, authentic and hubristic, that are also connected to different self-attributions, depending on whether one is feeling proud about a particular (local) aspect of the self versus feeling proud about the whole of oneself (global).

Keywords: emotion, mindset, moral development, moral psychology, pride, shame, self-regulation

Procedia PDF Downloads 107

Authors: Timothy Quinn, Ronald Croce, Aliaksandr Leuchanka, Justin Walker

Abstract:

Many endurance athletes train at or just below an intensity associated with their lactate threshold (LT) and often the heart rate (HR) that these athletes use for their LT are above their true LT-HR measured in a laboratory. Training above their true LT-HR may lead to overtraining and injury. Few athletes have the capability of measuring their LT in a laboratory and rely on perception to guide them, as accurate field tests to determine LT are limited. Therefore, the purpose of this study was to determine if an 8-minute field test could accurately define the HR associated with LT as measured in the laboratory. On Day 1, fifteen male runners (mean±SD; age, 27.8±4.1 years; height, 177.9±7.1 cm; body mass, 72.3±6.2 kg; body fat, 8.3±3.1%) performed a discontinuous treadmill LT/maximal oxygen consumption (LT/VO2max) test using a portable metabolic gas analyzer (Cosmed K4b2) and a lactate analyzer (Analox GL5). The LT (and associated HR) was determined using the 1/+1 method, where blood lactate increased by 1 mMol•L-1 over baseline followed by an additional 1 mMol•L-1 increase. Days 2 and 3 were randomized, and the athletes performed either an 8-minute run on the treadmill (TM) or on a 160-m indoor track (TR) in an effort to cover as much distance as possible while maintaining a high intensity throughout the entire 8 minutes. VO2, HR, ventilation (VE), and respiratory exchange ratio (RER) were measured using the Cosmed system, and rating of perceived exertion (RPE; 6-20 scale) was recorded every minute. All variables were averaged over the 8 minutes. The total distance covered over the 8 minutes was measured in both conditions. At the completion of the 8-minute runs, blood lactate was measured. Paired sample t-tests and pairwise Pearson correlations were computed to determine the relationship between variables measured in the field tests versus those obtained in the laboratory at LT. An alpha level of <0.05 was required for statistical significance. The HR (mean +SD) during the TM (167+9 bpm) and TR (172+9 bpm) tests were strongly correlated to the HR measured during the laboratory LT (169+11 bpm) test (r=0.68; p<0.03 and r=0.88; p<0.001, respectively). Blood lactate values during the TM and TR tests were not different from each other but were strongly correlated with the laboratory LT (r=0.73; p<0.04 and r=0.66; p<0.05, respectively). VE (Lmin-1) was significantly greater during the TR (134.8+11.4 Lmin-1) as compared to the TM (123.3+16.2 Lmin-1) with moderately strong correlations to the laboratory threshold values (r=0.38; p=0.27 and r=0.58; p=0.06, respectively). VO2 was higher during TR (51.4 mlkg-1min-1) compared to TM (47.4 mlkg-1min-1) with correlations of 0.33 (p=0.35) and 0.48 (p=0.13), respectively to threshold values. Total distance run was significantly greater during the TR (2331.6+180.9 m) as compared to the TM (2177.0+232.6 m), but they were strongly correlated with each other (r=0.82; p<0.002). These results suggest that an 8-minute running field test can accurately predict the HR associated with the LT and may be a simple test that athletes and coaches could implement to aid in training techniques.

Keywords: blood lactate, heart rate, running, training

Procedia PDF Downloads 252

Authors: Aziz Ikhlef

Abstract:

Most of multi-slices CT scanners are built with detectors composed of scintillator - photodiodes arrays. The photodiodes arrays are mainly based on front-illuminated technology for detectors under 64 slices and on back-illuminated photodiode for systems of 64 slices or more. The designs based on back-illuminated photodiodes were being investigated for CT machines to overcome the challenge of the higher number of runs and connection required in front-illuminated diodes. In backlit diodes, the electronic noise has already been improved because of the reduction of the load capacitance due to the routing reduction. This translated by a better image quality in low signal application, improving low dose imaging in large patient population. With the fast development of multi-detector-rows CT (MDCT) scanners and the increasing number of examinations, the clinical community has raised significant concerns on radiation dose received by the patient in both medical and regulatory community. In order to reduce individual exposure and in response to the recommendations of the International Commission on Radiological Protection (ICRP) which suggests that all exposures should be kept as low as reasonably achievable (ALARA), every manufacturer is trying to implement strategies and solutions to optimize dose efficiency and image quality based on x-ray emission and scanning parameters. The added demands on the CT detector performance also comes from the increased utilization of spectral CT or dual-energy CT in which projection data of two different tube potentials are collected. One of the approaches utilizes a technology called fast-kVp switching in which the tube voltage is switched between 80kVp and 140kVp in fraction of a millisecond. To reduce the cross-contamination of signals, the scintillator based detector temporal response has to be extremely fast to minimize the residual signal from previous samples. In addition, this paper will present an overview of detector technologies and image chain improvement which have been investigated in the last few years to improve the signal-noise ratio and the dose efficiency CT scanners in regular examinations and in energy discrimination techniques. Several parameters of the image chain in general and in the detector technology contribute in the optimization of the final image quality. We will go through the properties of the post-patient collimation to improve the scatter-to-primary ratio, the scintillator material properties such as light output, afterglow, primary speed, crosstalk to improve the spectral imaging, the photodiode design characteristics and the data acquisition system (DAS) to optimize for crosstalk, noise and temporal/spatial resolution.

Keywords: computed tomography, X-ray detector, medical imaging, image quality, artifacts

Procedia PDF Downloads 271

Authors: Stig H. Jørgensen, Trond Nordfjærn, Øyvind Teige Hedenstrøm, Torbjørn Rundmo

Abstract:

The aim of the paper is to estimate and forecast road traffic injuries in the coming 10-15 years given new targets in urban transport policy and shifts of mode of transport, including injury cross-effects of mode changes. The paper discusses possibilities and limitations in measuring and quantifying possible injury reductions. Injury data (killed and seriously injured road users) from six urban areas in Norway from 1998-2012 (N= 4709 casualties) form the basis for estimates of changing injury patterns. For the coming period calculation of number of injuries and injury rates by type of road user (categories of motorized versus non-motorized) by sex, age and type of road are made. A prognosticated population increase (25 %) in total population within 2025 in the six urban areas will curb the proceeded fall in injury figures. However, policy strategies and measures geared towards a stronger modal shift from use of private vehicles to safer public transport (bus, train) will modify this effect. On the other side will door to door transport (pedestrians on their way to/from public transport nodes) imply a higher exposure for pedestrians (bikers) converting from private vehicle use (including fall accidents not registered as traffic accidents). The overall effect is the sum of these modal shifts in the increasing urban population and in addition diminishing return to the majority of road safety countermeasures has also to be taken into account. The paper demonstrates how uncertainties in the various estimates (prediction factors) on increasing injuries as well as decreasing injury figures may partly offset each other. The paper discusses road safety policy and welfare consequences of transport mode shift, including reduced use of private vehicles, and further environmental impacts. In this regard, safety and environmental issues will as a rule concur. However pursuing environmental goals (e.g. improved air quality, reduced co2 emissions) encouraging more biking may generate more biking injuries. The study was given financial grants from the Norwegian Research Council’s Transport Safety Program.

Keywords: road injuries, forecasting, reduced private care use, urban, Norway

Procedia PDF Downloads 237

Authors: Ju Peng Poh, Jun Yu Charles Lee, Jonathan Chew Hoe Khoo

Abstract:

The emergence of digital twin technology, a digital replica of physical world, has improved the real-time access to data from sensors about the performance of buildings. This digital transformation has opened up many opportunities to improve the management of the building by using the data collected to help monitor consumption patterns and energy leakages. One example is the integration of predictive models for anomaly detection. In this paper, we use the GAM (Generalised Additive Model) for the anomaly detection of Air Handling Units (AHU) power consumption pattern. There is ample research work on the use of GAM for the prediction of power consumption at the office building and nation-wide level. However, there is limited illustration of its anomaly detection capabilities, prescriptive analytics case study, and its integration with the latest development of digital twin technology. In this paper, we applied the general GAM modelling framework on the historical data of the AHU power consumption and cooling load of the building between Jan 2018 to Aug 2019 from an education campus in Singapore to train prediction models that, in turn, yield predicted values and ranges. The historical data are seamlessly extracted from the digital twin for modelling purposes. We enhanced the utility of the GAM model by using it to power a real-time anomaly detection system based on the forward predicted ranges. The magnitude of deviation from the upper and lower bounds of the uncertainty intervals is used to inform and identify anomalous data points, all based on historical data, without explicit intervention from domain experts. Notwithstanding, the domain expert fits in through an optional feedback loop through which iterative data cleansing is performed. After an anomalously high or low level of power consumption detected, a set of rule-based conditions are evaluated in real-time to help determine the next course of action for the facilities manager. The performance of GAM is then compared with other approaches to evaluate its effectiveness. Lastly, we discuss the successfully deployment of this approach for the detection of anomalous power consumption pattern and illustrated with real-world use cases.

Keywords: anomaly detection, digital twin, generalised additive model, GAM, power consumption, supervised learning

Procedia PDF Downloads 154

Authors: Jo Angouri, Charlotte Kennedy, Shawnea Ting, David Rawlinson, Matthew Booker, Nigel Rees

Abstract:

Ambulance services need to balance the large volume of emergency (999 in the UK) calls they receive (e.g., West Midlands Ambulance Service reports per day about 4,000 999 calls; about 679,000 calls per year are received in Wales), with dispatching limited resource for on-site intervention to the most critical cases. The process by which Emergency Medical Dispatch (EMD) decisions are made is related to risk assessment and involves the caller and call-taker as well as clinical teams negotiating risk levels on a case-by-case basis. Medical Priority Dispatch System (MPDS – also referred to as Advanced Medical Priority Dispatch System AMPDS) are used in the UK by NHS Trusts (e.,g WAST) to process and prioritise 999 calls. MPDS / AMPDS provide structured protocols for call prioritisation and call management. Protocols/policy frameworks have not been examined before in the way we propose in our project. In more detail, the risk factors that play a role in the EMD negotiation between the caller and call-taker have been analysed in both medical and social science research. Research has focused on the structural, morphological and phonological aspects that could improve, and train, human-to-human interaction or automate risk detection, as well as the medical factors that need to be captured from the caller to inform the dispatch decision. There are two significant gaps in our knowledge that we address in our work: 1. the role of backstage clinical teams in translating the caller/call-taker interaction in their internal risk negotiation and, 2. the role of policy frameworks, protocols and regulations in the framing of institutional priorities and resource allocation. We take a multi method approach and combine the analysis of 999 calls with the analysis of policy documents. We draw on interaction analysis, corpus methodologies and thematic analysis. In this paper, we report on our preliminary findings and focus in particular on the risk factors we have identified and the relationship with the regulations that create the frame within which teams operate. We close the paper with implications of our study for providing evidence-based policy intervention and recommendations for further research.

Keywords: emergency (999) call, interaction analysis, discourse analysis, ambulance dispatch, medical discourse

Procedia PDF Downloads 103

Authors: Aziz Ikhlef

Abstract:

Most of multi-slices Computed Tomography (CT) scanners are built with detectors composed of scintillator - photodiodes arrays. The photodiodes arrays are mainly based on front-illuminated technology for detectors under 64 slices and on back-illuminated photodiode for systems of 64 slices or more. The designs based on back-illuminated photodiodes were being investigated for CT machines to overcome the challenge of the higher number of runs and connection required in front-illuminated diodes. In backlit diodes, the electronic noise has already been improved because of the reduction of the load capacitance due to the routing reduction. This is translated by a better image quality in low signal application, improving low dose imaging in large patient population. With the fast development of multi-detector-rows CT (MDCT) scanners and the increasing number of examinations, the clinical community has raised significant concerns on radiation dose received by the patient in both medical and regulatory community. In order to reduce individual exposure and in response to the recommendations of the International Commission on Radiological Protection (ICRP) which suggests that all exposures should be kept as low as reasonably achievable (ALARA), every manufacturer is trying to implement strategies and solutions to optimize dose efficiency and image quality based on x-ray emission and scanning parameters. The added demands on the CT detector performance also comes from the increased utilization of spectral CT or dual-energy CT in which projection data of two different tube potentials are collected. One of the approaches utilizes a technology called fast-kVp switching in which the tube voltage is switched between 80 kVp and 140 kVp in fraction of a millisecond. To reduce the cross-contamination of signals, the scintillator based detector temporal response has to be extremely fast to minimize the residual signal from previous samples. In addition, this paper will present an overview of detector technologies and image chain improvement which have been investigated in the last few years to improve the signal-noise ratio and the dose efficiency CT scanners in regular examinations and in energy discrimination techniques. Several parameters of the image chain in general and in the detector technology contribute in the optimization of the final image quality. We will go through the properties of the post-patient collimation to improve the scatter-to-primary ratio, the scintillator material properties such as light output, afterglow, primary speed, crosstalk to improve the spectral imaging, the photodiode design characteristics and the data acquisition system (DAS) to optimize for crosstalk, noise and temporal/spatial resolution.

Keywords: computed tomography, X-ray detector, medical imaging, image quality, artifacts

Procedia PDF Downloads 194

Authors: Jonathan Gong

Abstract:

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

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

Procedia PDF Downloads 130