Search results for: KaraAgroAI cocoa dataset

Authors: Monika Rani, Claudio Marchesi, Stefania Federici, Laura E. Depero

Abstract:

Degradation of the aquatic environment by plastic litter, especially microplastics (MPs), i.e., any water-insoluble solid plastic particle with the longest dimension in the range 1µm and 1000 µm (=1 mm) size, is an unfortunate indication of the advancement of the Anthropocene age on Earth. Microplastics formed due to natural weathering processes are termed as secondary microplastics, while when these are synthesized in industries, they are called primary microplastics. Their presence from the highest peaks to the deepest points in oceans explored and their resistance to biological and chemical decay has adversely affected the environment, especially marine life. Even though the presence of MPs in the marine environment is well-reported, a legitimate and authentic analytical technique to sample, analyze, and quantify the MPs is still under progress and testing stages. Among the characterization techniques, vibrational spectroscopic techniques are largely adopted in the field of polymers. And the ongoing miniaturization of these methods is on the way to revolutionize the plastic recycling industry. In this scenario, the capability and the feasibility of a miniaturized near-infrared (MicroNIR) spectroscopy combined with chemometrics tools for qualitative and quantitative analysis of urban plastic waste collected from a recycling plant and microplastic mixture fragmented in the lab were investigated. Based on the Resin Identification Code, 250 plastic samples were used for macroplastic analysis and to set up a library of polymers. Subsequently, MicroNIR spectra were analysed through the application of multivariate modelling. Principal Components Analysis (PCA) was used as an unsupervised tool to find trends within the data. After the exploratory PCA analysis, a supervised classification tool was applied in order to distinguish the different plastic classes, and a database containing the NIR spectra of polymers was made. For the microplastic analysis, the three most abundant polymers in the plastic litter, PE, PP, PS, were mechanically fragmented in the laboratory to micron size. The distinctive arrangement of blends of these three microplastics was prepared in line with a designed ternary composition plot. After the PCA exploratory analysis, a quantitative model Partial Least Squares Regression (PLSR) allowed to predict the percentage of microplastics in the mixtures. With a complete dataset of 63 compositions, PLS was calibrated with 42 data-points. The model was used to predict the composition of 21 unknown mixtures of the test set. The advantage of the consolidated NIR Chemometric approach lies in the quick evaluation of whether the sample is macro or micro, contaminated, coloured or not, and with no sample pre-treatment. The technique can be utilized with bigger example volumes and even considers an on-site evaluation and in this manner satisfies the need for a high-throughput strategy.

Keywords: chemometrics, microNIR, microplastics, urban plastic waste

Procedia PDF Downloads 165

Authors: Nokwethaba Makhanya, Babatunde J. Abiodun, Piotr Wolski

Abstract:

Climate change possibly intensifies hydrological droughts and reduces water availability in river basins. Despite this, most research on climate change effects in southern Africa has focused exclusively on meteorological droughts. This thesis projects the potential impact of climate change on the future characteristics of hydrological droughts in the Limpopo River Basin (LRB). The study uses regional climate model (RCM) measurements (from the Coordinated Regional Climate Downscaling Experiment, CORDEX) and a combination of hydrological simulations (using the Soil and Water Assessment Tool Plus model, SWAT+) to predict the impacts at four global warming levels (GWLs: 1.5℃, 2.0℃, 2.5℃, and 3.0℃) under the RCP8.5 future climate scenario. The SWAT+ model was calibrated and validated with a streamflow dataset observed over the basin, and the sensitivity of model parameters was investigated. The performance of the SWAT+LRB model was verified using the Nash-Sutcliffe efficiency (NSE), Percent Bias (PBIAS), Root Mean Square Error (RMSE), and coefficient of determination (R²). The Standardized Precipitation Evapotranspiration Index (SPEI) and the Standardized Precipitation Index (SPI) have been used to detect meteorological droughts. The Soil Water Index (SSI) has been used to define agricultural drought, while the Water Yield Drought Index (WYLDI), the Surface Run-off Index (SRI), and the Streamflow Index (SFI) have been used to characterise hydrological drought. The performance of the SWAT+ model simulations over LRB is sensitive to the parameters CN2 (initial SCS runoff curve number for moisture condition II) and ESCO (soil evaporation compensation factor). The best simulation generally performed better during the calibration period than the validation period. In calibration and validation periods, NSE is ≤ 0.8, while PBIAS is ≥ ﹣80.3%, RMSE ≥ 11.2 m³/s, and R² ≤ 0.9. The simulations project a future increase in temperature and potential evapotranspiration over the basin, but they do not project a significant future trend in precipitation and hydrological variables. However, the spatial distribution of precipitation reveals a projected increase in precipitation in the southern part of the basin and a decline in the northern part of the basin, with the region of reduced precipitation projected to increase with GWLs. A decrease in all hydrological variables is projected over most parts of the basin, especially over the eastern part of the basin. The simulations predict meteorological droughts (i.e., SPEI and SPI), agricultural droughts (i.e., SSI), and hydrological droughts (i.e., WYLDI, SRI) would become more intense and severe across the basin. SPEI-drought has a greater magnitude of increase than SPI-drought, and agricultural and hydrological droughts have a magnitude of increase between the two. As a result, this research suggests that future hydrological droughts over the LRB could be more severe than the SPI-drought projection predicts but less severe than the SPEI-drought projection. This research can be used to mitigate the effects of potential climate change on basin hydrological drought.

Keywords: climate change, CORDEX, drought, hydrological modelling, Limpopo River Basin

Procedia PDF Downloads 128

Authors: Aya Al Masri, Naima Oubenali, Safoin Aktaou, Thibault Julien, Malorie Martin, Fouad Maaloul

Abstract:

Introduction: The increased number of performed 'Computed Tomography (CT)' examinations raise public concerns regarding associated stochastic risk to patients. In its Publication 102, the ‘International Commission on Radiological Protection (ICRP)’ emphasized the importance of managing patient dose, particularly from repeated or multiple examinations. We developed a Dose Archiving and Communication System that gives multiple dose indexes (organ dose, effective dose, and skin-dose mapping) for patients undergoing radiological imaging exams. The aim of this study is to compare the organ dose values given by our software for patients undergoing CT exams with those of another software named "VirtualDose". Materials and methods: Our software uses Monte Carlo simulations to calculate organ doses for patients undergoing computed tomography examinations. The general calculation principle consists to simulate: (1) the scanner machine with all its technical specifications and associated irradiation cases (kVp, field collimation, mAs, pitch ...) (2) detailed geometric and compositional information of dozens of well identified organs of computational hybrid phantoms that contain the necessary anatomical data. The mass as well as the elemental composition of the tissues and organs that constitute our phantoms correspond to the recommendations of the international organizations (namely the ICRP and the ICRU). Their body dimensions correspond to reference data developed in the United States. Simulated data was verified by clinical measurement. To perform the comparison, 270 adult patients and 150 pediatric patients were used, whose data corresponds to exams carried out in France hospital centers. The comparison dataset of adult patients includes adult males and females for three different scanner machines and three different acquisition protocols (Head, Chest, and Chest-Abdomen-Pelvis). The comparison sample of pediatric patients includes the exams of thirty patients for each of the following age groups: new born, 1-2 years, 3-7 years, 8-12 years, and 13-16 years. The comparison for pediatric patients were performed on the “Head” protocol. The percentage of the dose difference were calculated for organs receiving a significant dose according to the acquisition protocol (80% of the maximal dose). Results: Adult patients: for organs that are completely covered by the scan range, the maximum percentage of dose difference between the two software is 27 %. However, there are three organs situated at the edges of the scan range that show a slightly higher dose difference. Pediatric patients: the percentage of dose difference between the two software does not exceed 30%. These dose differences may be due to the use of two different generations of hybrid phantoms by the two software. Conclusion: This study shows that our software provides a reliable dosimetric information for patients undergoing Computed Tomography exams.

Keywords: adult and pediatric patients, computed tomography, organ dose calculation, software comparison

Procedia PDF Downloads 163

Authors: Giada Feletti, Daniela Tedesco, Paolo Trucco

Abstract:

The present study aims to develop a dashboard of Key Performance Indicators (KPI) to enhance information and predictive capabilities in Emergency Medical Services (EMS) systems, supporting both operational and strategic decisions of different actors. The employed research methodology consists of the first phase of revision of the technical-scientific literature concerning the indicators currently used for the performance measurement of EMS systems. From this literature analysis, it emerged that current studies focus on two distinct perspectives: the ambulance service, a fundamental component of pre-hospital health treatment, and the patient care in the Emergency Department (ED). The perspective proposed by this study is to consider an integrated view of the ambulance service process and the ED process, both essential to ensure high quality of care and patient safety. Thus, the proposal focuses on the entire healthcare service process and, as such, allows considering the interconnection between the two EMS processes, the pre-hospital and hospital ones, connected by the assignment of the patient to a specific ED. In this way, it is possible to optimize the entire patient management. Therefore, attention is paid to the dependency of decisions that in current EMS management models tend to be neglected or underestimated. In particular, the integration of the two processes enables the evaluation of the advantage of an ED selection decision having visibility on EDs’ saturation status and therefore considering the distance, the available resources and the expected waiting times. Starting from a critical review of the KPIs proposed in the extant literature, the design of the dashboard was carried out: the high number of analyzed KPIs was reduced by eliminating the ones firstly not in line with the aim of the study and then the ones supporting a similar functionality. The KPIs finally selected were tested on a realistic dataset, which draws us to exclude additional indicators due to the unavailability of data required for their computation. The final dashboard, which was discussed and validated by experts in the field, includes a variety of KPIs able to support operational and planning decisions, early warning, and citizens’ awareness of EDs accessibility in real-time. By associating each KPI to the EMS phase it refers to, it was also possible to design a well-balanced dashboard covering both efficiency and effective performance of the entire EMS process. Indeed, just the initial phases related to the interconnection between ambulance service and patient’s care are covered by traditional KPIs compared to the subsequent phases taking place in the hospital ED. This could be taken into consideration for the potential future development of the dashboard. Moreover, the research could proceed by building a multi-layer dashboard composed of the first level with a minimal set of KPIs to measure the basic performance of the EMS system at an aggregate level and further levels with KPIs that can bring additional and more detailed information.

Keywords: dashboard, decision support, emergency medical services, key performance indicators

Procedia PDF Downloads 113

Authors: Manaranjan Pradhan, Shailaja Grover, U. Dinesh Kumar

Abstract:

Farming community in India, as well as other parts of the world, is one of the highly stressed communities due to reasons such as increasing input costs (cost of seeds, fertilizers, pesticide), droughts, reduced revenue leading to farmer suicides. Lack of integrated farm advisory system in India adds to the farmers problems. Farmers need right information during the early stages of crop’s lifecycle to prevent damage and loss in revenue. In this paper, we use deep learning techniques to develop an early warning system for detection of crop diseases using images taken by farmers using their smart phone. The research work leads to building a smart assistant using analytics and big data which could help the farmers with early diagnosis of the crop diseases and corrective actions. The classical approach for crop disease management has been to identify diseases at crop level. Recently, ImageNet Classification using the convolutional neural network (CNN) has been successfully used to identify diseases at individual plant level. Our model uses convolution filters, max pooling, dense layers and dropouts (to avoid overfitting). The models are built for binary classification (healthy or not healthy) and multi class classification (identifying which disease). Transfer learning is used to modify the weights of parameters learnt through ImageNet dataset and apply them on crop diseases, which reduces number of epochs to learn. One shot learning is used to learn from very few images, while data augmentation techniques are used to improve accuracy with images taken from farms by using techniques such as rotation, zoom, shift and blurred images. Models built using combination of these techniques are more robust for deploying in the real world. Our model is validated using tomato crop. In India, tomato is affected by 10 different diseases. Our model achieves an accuracy of more than 95% in correctly classifying the diseases. The main contribution of our research is to create a personal assistant for farmers for managing plant disease, although the model was validated using tomato crop, it can be easily extended to other crops. The advancement of technology in computing and availability of large data has made possible the success of deep learning applications in computer vision, natural language processing, image recognition, etc. With these robust models and huge smartphone penetration, feasibility of implementation of these models is high resulting in timely advise to the farmers and thus increasing the farmers' income and reducing the input costs.

Keywords: analytics in agriculture, CNN, crop disease detection, data augmentation, image recognition, one shot learning, transfer learning

Procedia PDF Downloads 119

Authors: Saul Estrin, Ute Stephan, Suncica Vujic

Abstract:

Background and research question: Gender differences in pay are present at all organisational levels, including at the very top. One possible way for women to circumvent organizational norms and discrimination is to engage in entrepreneurship because, as CEOs of their own organizations, entrepreneurs largely determine their own pay. While commercial entrepreneurship plays an important role in job creation and economic growth, social entrepreneurship has come to prominence because of its promise of addressing societal challenges such as poverty, social exclusion, or environmental degradation through market-based rather than state-sponsored activities. This opens the research question whether social entrepreneurship might be a form of entrepreneurship in which the pay of men and women is the same, or at least more similar; that is to say there is little or no gender pay gap. If the gender gap in pay persists also at the top of social enterprises, what are the factors, which might explain these differences? Methodology: The Oaxaca-Blinder Decomposition (OBD) is the standard approach of decomposing the gender pay gap based on the linear regression model. The OBD divides the gender pay gap into the ‘explained’ part due to differences in labour market characteristics (education, work experience, tenure, etc.), and the ‘unexplained’ part due to differences in the returns to those characteristics. The latter part is often interpreted as ‘discrimination’. There are two issues with this approach. (i) In many countries there is a notable convergence in labour market characteristics across genders; hence the OBD method is no longer revealing, since the largest portion of the gap remains ‘unexplained’. (ii) Adding covariates to a base model sequentially either to test a particular coefficient’s ‘robustness’ or to account for the ‘effects’ on this coefficient of adding covariates might be problematic, due to sequence-sensitivity when added covariates are correlated. Gelbach’s decomposition (GD) addresses latter by using the omitted variables bias formula, which constructs a conditional decomposition thus accounting for sequence-sensitivity when added covariates are correlated. We use GD to decompose the differences in gaps of pay (annual and hourly salary), size of the organisation (revenues), effort (weekly hours of work), and sources of finances (fees and sales, grants and donations, microfinance and loans, and investors’ capital) between men and women leading social enterprises. Database: Our empirical work is made possible by our collection of a unique dataset using respondent driven sampling (RDS) methods to address the problem that there is as yet no information on the underlying population of social entrepreneurs. The countries that we focus on are the United Kingdom, Spain, Romania and Hungary. Findings and recommendations: We confirm the existence of a gender pay gap between men and women leading social enterprises. This gap can be explained by differences in the accumulation of human capital, psychological and social factors, as well as cross-country differences. The results of this study contribute to a more rounded perspective, highlighting that although social entrepreneurship may be a highly satisfying occupation, it also perpetuates gender pay inequalities.

Keywords: Gelbach’s decomposition, gender gap, returns to social entrepreneurship, values and preferences

Procedia PDF Downloads 244

Authors: Francesca Nardi, Kashish Aneja, Katherine Ginsbach

Abstract:

The COVID-19 pandemic has demonstrated both a need for evidence-based and rights-based public health policy and how challenging it can be to make effective decisions with limited information, evidence, and data. The O’Neill Institute, in conjunction with several partners, has been working since the beginning of the pandemic to collect, analyze, and distribute critical data on public health policies enacted in response to COVID-19 around the world in the COVID-19 Law Lab. Well-designed laws and policies can help build strong health systems, implement necessary measures to combat viral transmission, enforce actions that promote public health and safety for everyone, and on the individual level have a direct impact on health outcomes. Poorly designed laws and policies, on the other hand, can fail to achieve the intended results and/or obstruct the realization of fundamental human rights, further disease spread, or cause unintended collateral harms. When done properly, laws can provide the foundation that brings clarity to complexity, embrace nuance, and identifies gaps of uncertainty. However, laws can also shape the societal factors that make disease possible. Law is inseparable from the rest of society, and COVID-19 has exposed just how much laws and policies intersects all facets of society. In the COVID-19 context, evidence-based and well-informed law and policy decisions—made at the right time and in the right place—can and have meant the difference between life or death for many. Having a solid evidentiary base of legal information can promote the understanding of what works well and where, and it can drive resources and action to where they are needed most. We know that legal mechanisms can enable nations to reduce inequities and prepare for emerging threats, like novel pathogens that result in deadly disease outbreaks or antibiotic resistance. The collection and analysis of data on these legal mechanisms is a critical step towards ensuring that legal interventions and legal landscapes are effectively incorporated into more traditional kinds of health science data analyses. The COVID-19 Law Labs see a unique opportunity to collect and analyze this kind of non-traditional data to inform policy using laws and policies from across the globe and across diseases. This global view is critical to assessing the efficacy of policies in a wide range of cultural, economic, and demographic circumstances. The COVID-19 Law Lab is not just a collection of legal texts relating to COVID-19; it is a dataset of concise and actionable legal information that can be used by health researchers, social scientists, academics, human rights advocates, law and policymakers, government decision-makers, and others for cross-disciplinary quantitative and qualitative analysis to identify best practices from this outbreak, and previous ones, to be better prepared for potential future public health events.

Keywords: public health law, surveillance, policy, legal, data

Procedia PDF Downloads 141

Authors: Rasmita Mishra, Chander Shekher

Abstract:

Background: Population ageing is an inevitable outcome of demographic transition. Due to increased life expectancy, the old age population in India and worldwide has increased, and it will continue to grow more alarmingly in the near future. There are redundant austerity that has been bestowed upon the widows, thus, the life of widows is never been easy in India. The loss of spouse along with other disadvantaged socioeconomic intermediaries like illiteracy and poverty often make the life of widows more difficult to live. Methodology: Ethical statement: The study used secondary data available in the public domain for its wider use in social research. Thus, there was no requirement of ethical consent in the present study. Data source: Building a Knowledge Base on Population Aging in India (BKPAI), 2011 dataset is used to fulfill the objectives of this study. It was carried out in seven states – Himachal Pradesh, Kerala, Maharashtra, Odisha, Punjab, Tamil Nadu, and West Bengal – having a higher percentage of the population in the age group 60 years and above compared to the national average. Statistical analysis: Descriptive and inferential statistics were used to understand the level of elderly widows and incidence of abuse against them in India and selected states. Bivariate and Trivariate analysis were carried out to check the pattern of abuse by selected covariates. Chi-Square test is used to verify the significance of the association. Further, Discriminant Analysis (DA) is carried out to understand which factor can separate out group of neglect and non-neglect elderly. Result: With the addition of 27 million from 2001 to 2011, the total elderly population in India is more than 100 million. Elderly females aged 60+ were more widows than their counterpart elderly males. This pattern was observed across selected states and at national level. At national level, more than one tenth (12 percent) of elderly experienced abuse in their lifetime. Incidence of abuse against elderly widows within family was considerably higher than the outside the family. This pattern was observed across the selected place and abuse in the study. In discriminant analysis, the significant difference between neglected and non-neglected elderly on each of the independent variables was examined using group mean and ANOVA. Discussion: The study is the first of its kind to assess the incidence of abuse against elderly widows using large-scale survey data. Another novelty of this study is that it has assessed for those states in India whereby the proportion of elderly is higher than the national average. Place and perpetrators involved in the abuse against elderly widows certainly envisaged the safeness in the present living arrangement of elderly widows. Conclusion: Due to the increasing life expectancy it is expected that the number of elderly will increase much faster than before. As biologically women live longer than men, there will be more women elderly than men. With respect to the living arrangement, after the demise of the spouse, elderly widows are more likely to live with their children who emerged as the main perpetrator of abuse.

Keywords: elderly abuse, emotional abuse physical abuse, material abuse, psychological abuse, quality of life

Procedia PDF Downloads 426

Authors: Simona Maria Ignat

Abstract:

Bullying online has been an increasing studied topic during the last years. Different approaches, psychological, linguistic, or computational, have been applied. To our best knowledge, a definition and a set of characteristics of phenomenon agreed internationally as a common framework are still waiting for answers. Thus, the objectives of this paper are the identification of bullying utterances on Twitter and their algorithms. This research paper is focused on the identification of words or groups of words, categorized as “utterances”, with bullying effect, from Twitter platform, extracted on a set of legislative criteria. This set is the result of analysis followed by synthesis of law documents on bullying(online) from United States of America, European Union, and Ireland. The outcome is a linguistic corpus with approximatively 10,000 entries. The methods applied to the first objective have been the following. The discourse analysis has been applied in identification of keywords with bullying effect in texts from Google search engine, Images link. Transcription and anonymization have been applied on texts grouped in CL1 (Corpus linguistics 1). The keywords search method and the legislative criteria have been used for identifying bullying utterances from Twitter. The texts with at least 30 representations on Twitter have been grouped. They form the second corpus linguistics, Bullying utterances from Twitter (CL2). The entries have been identified by using the legislative criteria on the the BoW method principle. The BoW is a method of extracting words or group of words with same meaning in any context. The methods applied for reaching the second objective is the conversion of parts of speech to alphabetical and numerical symbols and writing the bullying utterances as algorithms. The converted form of parts of speech has been chosen on the criterion of relevance within bullying message. The inductive reasoning approach has been applied in sampling and identifying the algorithms. The results are groups with interchangeable elements. The outcomes convey two aspects of bullying: the form and the content or meaning. The form conveys the intentional intimidation against somebody, expressed at the level of texts by grammatical and lexical marks. This outcome has applicability in the forensic linguistics for establishing the intentionality of an action. Another outcome of form is a complex of graphemic variations essential in detecting harmful texts online. This research enriches the lexicon already known on the topic. The second aspect, the content, revealed the topics like threat, harassment, assault, or suicide. They are subcategories of a broader harmful content which is a constant concern for task forces and legislators at national and international levels. These topic – outcomes of the dataset are a valuable source of detection. The analysis of content revealed algorithms and lexicons which could be applied to other harmful contents. A third outcome of content are the conveyances of Stylistics, which is a rich source of discourse analysis of social media platforms. In conclusion, this corpus linguistics is structured on legislative criteria and could be used in various fields.

Keywords: corpus linguistics, cyberbullying, legislation, natural language processing, twitter

Procedia PDF Downloads 86

Authors: Mehrnaz Mostafavi

Abstract:

The assessment and categorization of incidental lung nodules present a considerable challenge in healthcare, often necessitating resource-intensive multiple computed tomography (CT) scans for growth confirmation. This research addresses this issue by introducing a distinct computational approach leveraging radiomics and deep-learning methods. However, understanding local services is essential before implementing these advancements. With diverse tracking methods in place, there is a need for efficient and accurate identification approaches, especially in the context of managing lung nodules alongside pre-existing cancer scenarios. This study explores the integration of text-based algorithms in medical data curation, indicating their efficacy in conjunction with machine learning and deep-learning models for identifying lung nodules. Combining medical images with text data has demonstrated superior data retrieval compared to using each modality independently. While deep learning and text analysis show potential in detecting previously missed nodules, challenges persist, such as increased false positives. The presented research introduces a Structured-Query-Language (SQL) algorithm designed for identifying pulmonary nodules in a tertiary cancer center, externally validated at another hospital. Leveraging natural language processing (NLP) and machine learning, the algorithm categorizes lung nodule reports based on sentence features, aiming to facilitate research and assess clinical pathways. The hypothesis posits that the algorithm can accurately identify lung nodule CT scans and predict concerning nodule features using machine-learning classifiers. Through a retrospective observational study spanning a decade, CT scan reports were collected, and an algorithm was developed to extract and classify data. Results underscore the complexity of lung nodule cohorts in cancer centers, emphasizing the importance of careful evaluation before assuming a metastatic origin. The SQL and NLP algorithms demonstrated high accuracy in identifying lung nodule sentences, indicating potential for local service evaluation and research dataset creation. Machine-learning models exhibited strong accuracy in predicting concerning changes in lung nodule scan reports. While limitations include variability in disease group attribution, the potential for correlation rather than causality in clinical findings, and the need for further external validation, the algorithm's accuracy and potential to support clinical decision-making and healthcare automation represent a significant stride in lung nodule management and research.

Keywords: lung cancer diagnosis, structured-query-language (SQL), natural language processing (NLP), machine learning, CT scans

Procedia PDF Downloads 101

Authors: Muhammmad Riandhy Anindika Yudhy, Harry Patria, Ramadhani Santoso

Abstract:

Loss of containment is the primary hazard that process safety management is concerned within the oil and gas industry. Escalation to more serious consequences all begins with the loss of containment, starting with oil and gas release from leakage or spillage from primary containment resulting in pool fire, jet fire and even explosion when reacted with various ignition sources in the operations. Therefore, the heart of process safety management is avoiding loss of containment and mitigating its impact through the implementation of safeguards. The most effective safeguard for the case is an early detection system to alert Operations to take action prior to a potential case of loss of containment. The detection system value increases when applied to a long surface pipeline that is naturally difficult to monitor at all times and is exposed to multiple causes of loss of containment, from natural corrosion to illegal tapping. Based on prior researches and studies, detecting loss of containment accurately in the surface pipeline is difficult. The trade-off between cost-effectiveness and high accuracy has been the main issue when selecting the traditional detection method. The current best-performing method, Real-Time Transient Model (RTTM), requires analysis of closely positioned pressure, flow and temperature (PVT) points in the pipeline to be accurate. Having multiple adjacent PVT sensors along the pipeline is expensive, hence generally not a viable alternative from an economic standpoint.A conceptual approach to combine mathematical modeling using computational fluid dynamics and a supervised machine learning model has shown promising results to predict leakage in the pipeline. Mathematical modeling is used to generate simulation data where this data is used to train the leak detection and localization models. Mathematical models and simulation software have also been shown to provide comparable results with experimental data with very high levels of accuracy. While the supervised machine learning model requires a large training dataset for the development of accurate models, mathematical modeling has been shown to be able to generate the required datasets to justify the application of data analytics for the development of model-based leak detection systems for petroleum pipelines. This paper presents a review of key leak detection strategies for oil and gas pipelines, with a specific focus on crude oil applications, and presents the opportunities for the use of data analytics tools and mathematical modeling for the development of robust real-time leak detection and localization system for surface pipelines. A case study is also presented.

Keywords: pipeline, leakage, detection, AI

Procedia PDF Downloads 191

Authors: Burak Oksuzoglu, Savas Yildirim, Ferhat Kutlu

Abstract:

Opinion mining (OM) is one of the natural language processing (NLP) problems to determine the polarity of opinions, mostly represented on a positive-neutral-negative axis. The data for OM is usually collected from various social media platforms. In an era where social media has considerable control over companies’ futures, it’s worth understanding social media and taking actions accordingly. OM comes to the fore here as the scale of the discussion about companies increases, and it becomes unfeasible to gauge opinion on individual levels. Thus, the companies opt to automize this process by applying machine learning (ML) approaches to their data. For the last two decades, OM or sentiment analysis (SA) has been mainly performed by applying ML classification algorithms such as support vector machines (SVM) and Naïve Bayes to a bag of n-gram representations of textual data. With the advent of deep learning and its apparent success in NLP, traditional methods have become obsolete. Transfer learning paradigm that has been commonly used in computer vision (CV) problems started to shape NLP approaches and language models (LM) lately. This gave a sudden rise to the usage of the pretrained language model (PTM), which contains language representations that are obtained by training it on the large datasets using self-supervised learning objectives. The PTMs are further fine-tuned by a specialized downstream task dataset to produce efficient models for various NLP tasks such as OM, NER (Named-Entity Recognition), Question Answering (QA), and so forth. In this study, the traditional and modern NLP approaches have been evaluated for OM by using a sizable corpus belonging to a large private company containing about 76,000 comments in Turkish: SVM with a bag of n-grams, and two chosen pre-trained models, multilingual universal sentence encoder (MUSE) and bidirectional encoder representations from transformers (BERT). The MUSE model is a multilingual model that supports 16 languages, including Turkish, and it is based on convolutional neural networks. The BERT is a monolingual model in our case and transformers-based neural networks. It uses a masked language model and next sentence prediction tasks that allow the bidirectional training of the transformers. During the training phase of the architecture, pre-processing operations such as morphological parsing, stemming, and spelling correction was not used since the experiments showed that their contribution to the model performance was found insignificant even though Turkish is a highly agglutinative and inflective language. The results show that usage of deep learning methods with pre-trained models and fine-tuning achieve about 11% improvement over SVM for OM. The BERT model achieved around 94% prediction accuracy while the MUSE model achieved around 88% and SVM did around 83%. The MUSE multilingual model shows better results than SVM, but it still performs worse than the monolingual BERT model.

Keywords: BERT, MUSE, opinion mining, pretrained language model, SVM, Turkish

Procedia PDF Downloads 146

Authors: Lucia Billeci, Gennaro Tartarisco, Maurizio Varanini

Abstract:

Fetal cardiac monitoring by fetal electrocardiogram (fECG) can provide significant clinical information about the healthy condition of the fetus. Despite this potentiality till now the use of fECG in clinical practice has been quite limited due to the difficulties in its measuring. The recovery of fECG from the signals acquired non-invasively by using electrodes placed on the maternal abdomen is a challenging task because abdominal signals are a mixture of several components and the fetal one is very weak. This paper presents an approach for fECG extraction from abdominal maternal recordings, which exploits the characteristics of pseudo-periodicity of fetal ECG. It consists of devising a quality index (fQI) for fECG and of finding the linear combinations of preprocessed abdominal signals, which maximize these fQI (quality index optimization - QIO). It aims at improving the performances of the most commonly adopted methods for fECG extraction, usually based on maternal ECG (mECG) estimating and canceling. The procedure for the fECG extraction and fetal QRS (fQRS) detection is completely unsupervised and based on the following steps: signal pre-processing; maternal ECG (mECG) extraction and maternal QRS detection; mECG component approximation and canceling by weighted principal component analysis; fECG extraction by fQI maximization and fetal QRS detection. The proposed method was compared with our previously developed procedure, which obtained the highest at the Physionet/Computing in Cardiology Challenge 2013. That procedure was based on removing the mECG from abdominal signals estimated by a principal component analysis (PCA) and applying the Independent component Analysis (ICA) on the residual signals. Both methods were developed and tuned using 69, 1 min long, abdominal measurements with fetal QRS annotation of the dataset A provided by PhysioNet/Computing in Cardiology Challenge 2013. The QIO-based and the ICA-based methods were compared in analyzing two databases of abdominal maternal ECG available on the Physionet site. The first is the Abdominal and Direct Fetal Electrocardiogram Database (ADdb) which contains the fetal QRS annotations thus allowing a quantitative performance comparison, the second is the Non-Invasive Fetal Electrocardiogram Database (NIdb), which does not contain the fetal QRS annotations so that the comparison between the two methods can be only qualitative. In particular, the comparison on NIdb was performed defining an index of quality for the fetal RR series. On the annotated database ADdb the QIO method, provided the performance indexes Sens=0.9988, PPA=0.9991, F1=0.9989 overcoming the ICA-based one, which provided Sens=0.9966, PPA=0.9972, F1=0.9969. The comparison on NIdb was performed defining an index of quality for the fetal RR series. The index of quality resulted higher for the QIO-based method compared to the ICA-based one in 35 records out 55 cases of the NIdb. The QIO-based method gave very high performances with both the databases. The results of this study foresees the application of the algorithm in a fully unsupervised way for the implementation in wearable devices for self-monitoring of fetal health.

Keywords: fetal electrocardiography, fetal QRS detection, independent component analysis (ICA), optimization, wearable

Procedia PDF Downloads 280

Authors: Almudena Espin Perez, Tyler Risom, Carl Pelz, Isabel English, Robert M. Angelo, Rosalie Sears, Andrew J. Gentles

Abstract:

Pancreatic ductal adenocarcinoma (PDAC) is one of the most deadly malignancies. The role of the tumor microenvironment (TME) is gaining significant attention in cancer research. Despite ongoing efforts, the nature of the interactions between tumors, immune cells, and stromal cells remains poorly understood. The cell-intrinsic properties that govern cell lineage plasticity in PDAC and extrinsic influences of immune populations require technically challenging approaches due to the inherently heterogeneous nature of PDAC. Understanding the cell lineage plasticity of PDAC will improve the development of novel strategies that could be translated to the clinic. Members of the team have demonstrated that the acquisition of ductal to neuroendocrine lineage plasticity in PDAC confers therapeutic resistance and is a biomarker of poor outcomes in patients. Our approach combines computational methods for deconvolving bulk transcriptomic cancer data using CIBERSORTx and high-throughput single-cell imaging using Multiplexed Ion Beam Imaging (MIBI) to study lineage plasticity in PDAC and its relationship to the infiltrating immune system. The CIBERSORTx algorithm uses signature matrices from immune cells and stroma from sorted and single-cell data in order to 1) infer the fractions of different immune cell types and stromal cells in bulked gene expression data and 2) impute a representative transcriptome profile for each cell type. We studied a unique set of 300 genomically well-characterized primary PDAC samples with rich clinical annotation. We deconvolved the PDAC transcriptome profiles using CIBERSORTx, leveraging publicly available single-cell RNA-seq data from normal pancreatic tissue and PDAC to estimate cell type proportions in PDAC, and digitally reconstruct cell-specific transcriptional profiles from our study dataset. We built signature matrices and optimized by simulations and comparison to ground truth data. We identified cell-type-specific transcriptional programs that contribute to cancer cell lineage plasticity, especially in the ductal compartment. We also studied cell differentiation hierarchies using CytoTRACE and predict cell lineage trajectories for acinar and ductal cells that we believe are pinpointing relevant information on PDAC progression. Collaborators (Angelo lab, Stanford University) has led the development of the Multiplexed Ion Beam Imaging (MIBI) platform for spatial proteomics. We will use in the very near future MIBI from tissue microarray of 40 PDAC samples to understand the spatial relationship between cancer cell lineage plasticity and stromal cells focused on infiltrating immune cells, using the relevant markers of PDAC plasticity identified from the RNA-seq analysis.

Keywords: deconvolution, imaging, microenvironment, PDAC

Procedia PDF Downloads 128

Authors: Eashwar V. Somasundaram, Raoul R. Wadhwa, Jacob G. Scott

Abstract:

The use of radiomics in measuring geometric properties of tumor images such as size, surface area, and volume has been invaluable in assessing cancer diagnosis, treatment, and prognosis. In addition to analyzing geometric properties, radiomics would benefit from measuring topological properties using persistent homology. Intuitively, features uncovered by persistent homology may correlate to tumor structural features. One example is necrotic cavities (corresponding to 2D topological features), which are markers of very aggressive tumors. We develop a data pipeline in R that clusters tumors images based on persistent homology is used to identify meaningful clinical distinctions between tumors and possibly new relationships not captured by established clinical categorizations. A preliminary analysis was performed on 16 Magnetic Resonance Imaging (MRI) breast tissue segments downloaded from the 'Investigation of Serial Studies to Predict Your Therapeutic Response with Imaging and Molecular Analysis' (I-SPY TRIAL or ISPY1) collection in The Cancer Imaging Archive. Each segment represents a patient’s breast tumor prior to treatment. The ISPY1 dataset also provided the estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status data. A persistent homology matrix up to 2-dimensional features was calculated for each of the MRI segmentation. Wasserstein distances were then calculated between all pairwise tumor image persistent homology matrices to create a distance matrix for each feature dimension. Since Wasserstein distances were calculated for 0, 1, and 2-dimensional features, three hierarchal clusters were constructed. The adjusted Rand Index was used to see how well the clusters corresponded to the ER/PR/HER2 status of the tumors. Triple-negative cancers (negative status for all three receptors) significantly clustered together in the 2-dimensional features dendrogram (Adjusted Rand Index of .35, p = .031). It is known that having a triple-negative breast tumor is associated with aggressive tumor growth and poor prognosis when compared to non-triple negative breast tumors. The aggressive tumor growth associated with triple-negative tumors may have a unique structure in an MRI segmentation, which persistent homology is able to identify. This preliminary analysis shows promising results in the use of persistent homology on tumor imaging to assess the severity of breast tumors. The next step is to apply this pipeline to other tumor segment images from The Cancer Imaging Archive at different sites such as the lung, kidney, and brain. In addition, whether other clinical parameters, such as overall survival, tumor stage, and tumor genotype data are captured well in persistent homology clusters will be assessed. If analyzing tumor MRI segments using persistent homology consistently identifies clinical relationships, this could enable clinicians to use persistent homology data as a noninvasive way to inform clinical decision making in oncology.

Keywords: cancer biology, oncology, persistent homology, radiomics, topological data analysis, tumor imaging

Procedia PDF Downloads 135

Authors: Linyi Fan, C.J. Schumaker

Abstract:

Background: Biomedical efforts to treat Alzheimer’s disease (AD) have typically produced mixed to poor results, while more lifestyle-focused treatments such as exercise may fare better than existing biomedical treatments. A few promising studies have indicated that activities of daily life (ADL) may be a useful way of predicting AD. However, the existing cross-sectional studies fail to show how functional-related issues such as ADL in early years predict AD and how social factors influence health either in addition to or in interaction with individual risk factors. This study would helpbetterscreening and early treatments for the elderly population and healthcare practice. The findings have significance academically and practically in terms of creating positive social change. Methodology: The purpose of this quantitative historical, correlational study was to examine the relationship between early years’ ADL and the development of AD in later years. The studyincluded 4,526participantsderived fromRAND HRS dataset. The Health and Retirement Study (HRS) is a longitudinal household survey data set that is available forresearchof retirement and health among the elderly in the United States. The sample was selected by the completion of survey questionnaire about AD and dementia. The variablethat indicates whether the participant has been diagnosed with AD was the dependent variable. The ADL indices and changes in ADL were the independent variables. A four-step multilevel regression model approach was utilized to address the research questions. Results: Amongst 4,526 patients who completed the AD and dementia questionnaire, 144 (3.1%) were diagnosed with AD. Of the 4,526 participants, 3,465 (76.6%) have high school and upper education degrees,4,074 (90.0%) were above poverty threshold. The model evaluatedthe effect of ADL and change in ADL on onset of AD in late years while allowing the intercept of the model to vary by level of education. The results suggested that the only significant predictor of the onset of AD was changes in early years’ ADL (b = 20.253, z = 2.761, p < .05). However, the result of the sensitivity analysis (b = 7.562, z = 1.900, p =.058), which included more control variables and increased the observation period of ADL, are not supported this finding. The model also estimated whether the variances of random effect vary by Level-2 variables. The results suggested that the variances associated with random slopes were approximately zero, suggesting that the relationship between early years’ ADL were not influenced bysociodemographic factors. Conclusion: The finding indicated that an increase in changes in ADL leads to an increase in the probability of onset AD in the future. However, this finding is not support in a broad observation period model. The study also failed to reject the hypothesis that the sociodemographic factors explained significant amounts of variance in random effect. Recommendations were then made for future research and practice based on these limitations and the significance of the findings.

Keywords: alzheimer’s disease, epidemiology, moderation, multilevel modeling

Procedia PDF Downloads 135

Authors: Krešimir Begović, Miloš Rydval, Jan Tumajer, Kristyna Svobodová, Thomas Langbehn, Yumei Jiang, Vojtech Čada, Vaclav Treml, Ryszard Kaczka, Miroslav Svoboda

Abstract:

Environmental changes have increased the frequency and intensity of climatic extremes, particularly hotter droughts, leading to altered tree growth patterns and multi-year lags in tree recovery. The effects of shifting climatic conditions on tree growth are inhomogeneous across species’ natural distribution ranges, with large spatial heterogeneity and inter-population variability, but generally have significant consequences for contemporary forest dynamics and future ecosystem functioning. Despite numerous studies on the impacts of regional drought effects, large uncertainties remain regarding the mechanistic basis of drought legacy effects on wood formation and the ability of individual species to cope with increasingly drier growing conditions and rising year-to-year climatic variability. To unravel the complexity of climate-growth interactions and assess species-specific responses to severe droughts, we combined forward modeling of tree growth (VS-lite model) with correlation analyses against climate (temperature, precipitation, and the SPEI-3 moisture index) and growth responses to extreme drought events from multiple tree-ring parameters (tree-width and blue intensity parameters). We used an extensive dataset with over 1000 tree-ring samples from 23 nature forest reserves across an altitudinal range in Czechia and Slovakia. Our results revealed substantial spatiotemporal variability in growth responses to summer season temperature and moisture availability across species and tree-ring parameters. However, a general trend of increasing spring moisture-growth sensitivity in recent decades was observed in the Scots pine mountain forests and lowland forests of both species. The VS-lite model effectively captured nonstationary climate-growth relationships and accurately estimated high-frequency growth variability, indicating a significant incidence of regional drought events and growth reductions. Notably, growth reductions during extreme drought years and discrete legacy effects identified in individual wood components were most pronounced in the lowland forests. Together with the observed growth declines in recent decades, these findings suggest an increasing vulnerability of Norway spruce and Scots pine in dry lowlands under intensifying climatic constraints.

Keywords: dendroclimatology, Vaganova–Shashkin lite, conifers, central Europe, drought, blue intensity

Procedia PDF Downloads 59

Authors: Daniel R. dos Santos, Mateus S. Maldonado, Estevão J. R. Batista

Abstract:

The military commanders increasingly dependent on spatial awareness, as knowing where enemy are, understanding how war battle scenarios change over time, and visualizing these trends in ways that offer insights for decision-making. Thanks to advancements in geospatial technologies and artificial intelligence algorithms, the commanders are now able to modernize military operations on a universal scale. Thus, geovisualisation has become an essential asset in the defense sector. It has become indispensable for better decisionmaking in dynamic/temporal scenarios, operation planning and management for the war field, situational awareness, effective planning, monitoring, and others. For example, a 3D visualization of war field data contributes to intelligence analysis, evaluation of postmission outcomes, and creation of predictive models to enhance decision-making and strategic planning capabilities. However, old-school visualization methods are slow, expensive, and unscalable. Despite modern technologies in generating 3D point clouds, such as LIDAR and stereo sensors, monocular depth values based on deep learning can offer a faster and more detailed view of the environment, transforming single images into visual information for valuable insights. We propose a dedicated monocular depth reconstruction approach via deep learning techniques for 3D geovisualisation of satellite images. It introduces scalability in terrain reconstruction and data visualization. First, a dataset with more than 7,000 satellite images and associated digital elevation model (DEM) is created. It is based on high resolution optical and radar imageries collected from Planet and Copernicus, on which we fuse highresolution topographic data obtained using technologies such as LiDAR and the associated geographic coordinates. Second, we developed an imagery-DEM fusion strategy that combine feature maps from two encoder-decoder networks. One network is trained with radar and optical bands, while the other is trained with DEM features to compute dense 3D depth. Finally, we constructed a benchmark with sparse depth annotations to facilitate future research. To demonstrate the proposed method's versatility, we evaluated its performance on no annotated satellite images and implemented an enclosed environment useful for Geovisualisation applications. The algorithms were developed in Python 3.0, employing open-source computing libraries, i.e., Open3D, TensorFlow, and Pythorch3D. The proposed method provides fast and accurate decision-making with GIS for localization of troops, position of the enemy, terrain and climate conditions. This analysis enhances situational consciousness, enabling commanders to fine-tune the strategies and distribute the resources proficiently.

Keywords: depth, deep learning, geovisualisation, satellite images

Procedia PDF Downloads 10

Authors: Violetta Cataldo, Renata Savy, Simona Sbranna

Abstract:

Background: Foreign Language Learning (FLL) is characterised by prosodic transfer phenomena regarding pitch accents placement, intonation patterns, and pitch range excursion from the learners’ mother tongue to their Foreign Language (FL) which suggests that the gradual development of general linguistic competence in FL does not imply an equally correspondent improvement of the prosodic competence. Topic: The present study aims to monitor the development of prosodic competence of learners of Italian and German throughout the FLL process. The primary object of this study is to investigate the intonational features and the f₀ range excursion of Italian and German from a cross-linguistic perspective; analyses of native speakers’ productions point out the differences between this pair of languages and provide models for the Target Language (TL). A following crosscheck compares the L2 productions in Italian and German by non-native speakers to the Target Language models, in order to verify the occurrence of prosodic interference phenomena, i.e., type, degree, and modalities. Methodology: The subjects of the research are university students belonging to two groups: Italian native speakers learning German as FL and German native speakers learning Italian as FL. Both of them have been divided into three subgroups according to the FL proficiency level (beginners, intermediate, advanced). The dataset consists of wh-questions placed in situational contexts uttered in both speakers’ L1 and FL. Using a phonetic approach, analyses have considered three domains of intonational contours (Initial Profile, Nuclear Accent, and Terminal Contour) and two dimensions of the f₀ range parameter (span and level), which provide a basis for comparison between L1 and L2 productions. Findings: Results highlight a strong presence of prosodic transfer phenomena affecting L2 productions in the majority of both Italian and German learners, irrespective of their FL proficiency level; the transfer concerns all the three domains of the contour taken into account, although with different modalities and characteristics. Currently, L2 productions of German learners show a pitch span compression on the domain of the Terminal Contour compared to their L1 towards the TL; furthermore, German learners tend to use lower pitch range values in deviation from their L1 when improving their general linguistic competence in Italian FL proficiency level. Results regarding pitch range span and level in L2 productions by Italian learners are still in progress. At present, they show a similar tendency to expand the pitch span and to raise the pitch level, which also reveals a deviation from the L1 possibly in the direction of German TL. Conclusion: Intonational features seem to be 'resistant' parameters to which learners appear not to be particularly sensitive. By contrast, they show a certain sensitiveness to FL pitch range dimensions. Making clear which the most resistant and the most sensitive parameters are when learning FL prosody could lay groundwork for the development of prosodic trainings thanks to which learners could finally acquire a clear and natural pronunciation and intonation.

Keywords: foreign language learning, German, Italian, L2 prosody, pitch range, transfer

Procedia PDF Downloads 286

Authors: Elham Koohikerade, Silvio Jose Gumiere

Abstract:

In this research, irrigation is highlighted as crucial for improving both the yield and quality of ‎potatoes due to their high sensitivity to soil moisture changes. The study presents a hybrid Long ‎Short-Term Memory (LSTM) model aimed at optimizing irrigation scheduling in potato fields in ‎Quebec City, Canada. This model integrates model-based and satellite-derived datasets to simulate ‎soil moisture content, addressing the limitations of field data. Developed under the guidance of the ‎Food and Agriculture Organization (FAO), the simulation approach compensates for the lack of direct ‎soil sensor data, enhancing the LSTM model's predictions. The model was calibrated using indices ‎like Surface Soil Moisture (SSM), Normalized Vegetation Difference Index (NDVI), Enhanced ‎Vegetation Index (EVI), and Normalized Multi-band Drought Index (NMDI) to effectively forecast ‎soil moisture reductions. Understanding soil moisture and plant development is crucial for assessing ‎drought conditions and determining irrigation needs. This study validated the spectral characteristics ‎of vegetation and soil using ECMWF Reanalysis v5 (ERA5) and Moderate Resolution Imaging ‎Spectrometer (MODIS) data from 2019 to 2023, collected from agricultural areas in Dolbeau and ‎Peribonka, Quebec. Parameters such as surface volumetric soil moisture (0-7 cm), NDVI, EVI, and ‎NMDI were extracted from these images. A regional four-year dataset of soil and vegetation moisture ‎was developed using a machine learning approach combining model-based and satellite-based ‎datasets. The LSTM model predicts soil moisture dynamics hourly across different locations and ‎times, with its accuracy verified through cross-validation and comparison with existing soil moisture ‎datasets. The model effectively captures temporal dynamics, making it valuable for applications ‎requiring soil moisture monitoring over time, such as anomaly detection and memory analysis. By ‎identifying typical peak soil moisture values and observing distribution shapes, irrigation can be ‎scheduled to maintain soil moisture within Volumetric Soil Moisture (VSM) values of 0.25 to 0.30 ‎m²/m², avoiding under and over-watering. The strong correlations between parcels suggest that a ‎uniform irrigation strategy might be effective across multiple parcels, with adjustments based on ‎specific parcel characteristics and historical data trends. The application of the LSTM model to ‎predict soil moisture and vegetation indices yielded mixed results. While the model effectively ‎captures the central tendency and temporal dynamics of soil moisture, it struggles with accurately ‎predicting EVI, NDVI, and NMDI.‎

Keywords: irrigation scheduling, LSTM neural network, remotely sensed indices, soil and vegetation ‎monitoring

Procedia PDF Downloads 41

Authors: Haya Salah, Srinivas Sharan

Abstract:

Healthcare facilities use appointment systems to schedule their appointments and to manage access to their medical services. With the growing demand for outpatient care, it is now imperative to manage physician's time effectively. However, high variation in consultation duration affects the clinical scheduler's ability to estimate the appointment duration and allocate provider time appropriately. Underestimating consultation times can lead to physician's burnout, misdiagnosis, and patient dissatisfaction. On the other hand, appointment durations that are longer than required lead to doctor idle time and fewer patient visits. Therefore, a good estimation of consultation duration has the potential to improve timely access to care, resource utilization, quality of care, and patient satisfaction. Although the literature on factors influencing consultation length abound, little work has done to predict it using based data-driven approaches. Therefore, this study aims to predict consultation duration using supervised machine learning algorithms (ML), which predicts an outcome variable (e.g., consultation) based on potential features that influence the outcome. In particular, ML algorithms learn from a historical dataset without explicitly being programmed and uncover the relationship between the features and outcome variable. A subset of the data used in this study has been obtained from the electronic medical records (EMR) of four different outpatient clinics located in central Pennsylvania, USA. Also, publicly available information on doctor's characteristics such as gender and experience has been extracted from online sources. This research develops three popular ML algorithms (deep learning, random forest, gradient boosting machine) to predict the treatment time required for a patient and conducts a comparative analysis of these algorithms with respect to predictive performance. The findings of this study indicate that ML algorithms have the potential to predict the provider service time with superior accuracy. While the current approach of experience-based appointment duration estimation adopted by the clinic resulted in a mean absolute percentage error of 25.8%, the Deep learning algorithm developed in this study yielded the best performance with a MAPE of 12.24%, followed by gradient boosting machine (13.26%) and random forests (14.71%). Besides, this research also identified the critical variables affecting consultation duration to be patient type (new vs. established), doctor's experience, zip code, appointment day, and doctor's specialty. Moreover, several practical insights are obtained based on the comparative analysis of the ML algorithms. The machine learning approach presented in this study can serve as a decision support tool and could be integrated into the appointment system for effectively managing patient scheduling.

Keywords: clinical decision support system, machine learning algorithms, patient scheduling, prediction models, provider service time

Procedia PDF Downloads 121

Authors: Basiru Amuneni

Abstract:

Astronomy is one domain with a rise in data. Traditional tools for data management have been employed in the quest for knowledge discovery. However, these traditional tools become limited in the face of big. One means of maximizing knowledge discovery for big data is the use of scientific visualisation. The aim of the work is to explore the possibilities offered by emerging creative technologies of Virtual Reality (VR) systems and game engines to visualize multiwavelength datasets. Game Engines are primarily used for developing video games, however their advanced graphics could be exploited for scientific visualization which provides a means to graphically illustrate scientific data to ease human comprehension. Modern astronomy is now in the era of multiwavelength data where a single galaxy for example, is captured by the telescope several times and at different electromagnetic wavelength to have a more comprehensive picture of the physical characteristics of the galaxy. Visualising this in an immersive environment would be more intuitive and natural for an observer. This work presents a standalone VR application that accesses galaxy FITS files. The application was built using the Unity Game Engine for the graphics underpinning and the OpenXR API for the VR infrastructure. The work used a methodology known as Design Science Research (DSR) which entails the act of ‘using design as a research method or technique’. The key stages of the galaxy modelling pipeline are FITS data preparation, Galaxy Modelling, Unity 3D Visualisation and VR Display. The FITS data format cannot be read by the Unity Game Engine directly. A DLL (CSHARPFITS) which provides a native support for reading and writing FITS files was used. The Galaxy modeller uses an approach that integrates cleaned FITS image pixels into the graphics pipeline of the Unity3d game Engine. The cleaned FITS images are then input to the galaxy modeller pipeline phase, which has a pre-processing script that extracts, pixel, galaxy world position, and colour maps the FITS image pixels. The user can visualise image galaxies in different light bands, control the blend of the image with similar images from different sources or fuse images for a holistic view. The framework will allow users to build tools to realise complex workflows for public outreach and possibly scientific work with increased scalability, near real time interactivity with ease of access. The application is presented in an immersive environment and can use all commercially available headset built on the OpenXR API. The user can select galaxies in the scene, teleport to the galaxy, pan, zoom in/out, and change colour gradients of the galaxy. The findings and design lessons learnt in the implementation of different use cases will contribute to the development and design of game-based visualisation tools in immersive environment by enabling informed decisions to be made.

Keywords: astronomy, visualisation, multiwavelenght dataset, virtual reality

Procedia PDF Downloads 92

Authors: Kostas Kouvaris, Anastasia Eleftheriou, Georgios A. Sarantitis, Apostolos Chondronasios

Abstract:

To achieve the objective of almost zero carbon energy solutions by 2050, the EU needs to accelerate the development of integrated, highly efficient and environmentally friendly solutions. In this direction, district heating and cooling (DHC) emerges as a viable and more efficient alternative to conventional, decentralized heating and cooling systems, enabling a combination of more efficient renewable and competitive energy supplies. In this paper, we develop a forecasting tool for near real-time local weather and thermal energy demand predictions for an entire DHC network. In this fashion, we are able to extend the functionality and to improve the energy efficiency of the DHC network by predicting and adjusting the heat load that is distributed from the heat generation plant to the connected buildings by the heat pipe network. Two case-studies are considered; one for Vransko, Slovenia and one for Montpellier, France. The data consists of i) local weather data, such as humidity, temperature, and precipitation, ii) weather forecast data, such as the outdoor temperature and iii) DHC operational parameters, such as the mass flow rate, supply and return temperature. The external temperature is found to be the most important energy-related variable for space conditioning, and thus it is used as an external parameter for the energy demand models. For the development of the forecasting tool, we use state-of-the-art deep neural networks and more specifically, recurrent networks with long-short-term memory cells, which are able to capture complex non-linear relations among temporal variables. Firstly, we develop models to forecast outdoor temperatures for the next 24 hours using local weather data for each case-study. Subsequently, we develop models to forecast thermal demand for the same period, taking under consideration past energy demand values as well as the predicted temperature values from the weather forecasting models. The contributions to the scientific and industrial community are three-fold, and the empirical results are highly encouraging. First, we are able to predict future thermal demand levels for the two locations under consideration with minimal errors. Second, we examine the impact of the outdoor temperature on the predictive ability of the models and how the accuracy of the energy demand forecasts decreases with the forecast horizon. Third, we extend the relevant literature with a new dataset of thermal demand and examine the performance and applicability of machine learning techniques to solve real-world problems. Overall, the solution proposed in this paper is in accordance with EU targets, providing an automated smart energy management system, decreasing human errors and reducing excessive energy production.

Keywords: machine learning, LSTMs, district heating and cooling system, thermal demand

Procedia PDF Downloads 142

Authors: Richard Abel

Abstract:

In the United States, increased incarceration rates have caused many adolescents to feel the strain of parental absence. This absence is then manifest through adolescent feelings of parental rejection. Additionally, upon reentry maternal incarceration may be related to adolescents experienced perceived excessive disciple. It is possible parents engage in this manner of discipline attempting to prevent the child from taking the same path to incarceration as the parent. According to General Strain Theory, adolescents encountering strain are likely to experience negative emotions. The emotion that is most likely to lead to delinquency is anger through reduced inhibitions and motivation to act. Additionally, males are more likely to engage in delinquent behavior, regardless of experiencing strain. This is not the case for every male who experiences maternal incarceration, parental rejection, excessive discipline, or anger. There are protective factors that enable agency within individuals. One such protective factor is mastery, or the perception that one is in control of his or her own future. The model proposed in this research suggests maternal incarceration is associated with increased parental rejection and excessive discipline in males. Males experiencing parental rejection and excessive discipline are likely to experience increased anger, which is then associated with increases in delinquent behavior. This model explores whether agency, in the form of mastery, mediates the relationship between strains and negative emotions, or between negative emotions and delinquent behavior. The Kaplan Longitudinal and Multigenerational Study (KLAMS) dataset is uniquely situated to analyze this model providing longitudinal data collected from both parents and their offspring. Maternal incarceration is constructed using parental responses such that the mother was incarcerated after the child’s birth, and any incarceration that happened prior to birth is excluded. The remaining variables of the study are all constructed from varying waves of the adolescent survey. Parental rejection, along with control variables for age, race, parental socioeconomic status, neighborhood effects, delinquent peers, and prior delinquent behavior are all constructed using Wave I data. To increase causal inference, the negative emotion of anger and the mediating variable of mastery are measured during Wave II. Lastly, delinquent behavior is measured at Wave III. Results of the analysis show expected relationships such that adolescent males encountering maternal incarceration show increased perception of parental rejection and excessive discipline. Additionally, there is a positive relationship between parental rejection and excessive discipline at Wave I and feelings of anger at Wave II for males. For males experiencing either of these strains in Wave I, feelings of anger in Wave II are found to be associated with increased delinquent behavior in Wave III. Mastery was found to mediate the relationship between both parental rejection and excessive discipline and anger, but no such mediation occurs in the relationship between anger and delinquency, regardless of the strain being experienced. These findings suggest adolescent males who feel they are in control of their own lives are less likely to experience negative emotions produced by the occurrence of strain, thereby decreasing male engagement in delinquent behavior later in life.

Keywords: delinquency, mastery, maternal incarceration, strain

Procedia PDF Downloads 133

Authors: Sara Rejeb, Catherine Duveau, Tabea Rebafka

Abstract:

To monitor the production process of turbofan aircraft engines, multiple measurements of various geometrical parameters are systematically recorded on manufactured parts. Engine parts are subject to extremely high standards as they can impact the performance of the engine. Therefore, it is essential to analyze these databases to better understand the influence of the different parameters on the engine's performance. Self-organizing maps are unsupervised neural networks which achieve two tasks simultaneously: they visualize high-dimensional data by projection onto a 2-dimensional map and provide clustering of the data. This technique has become very popular for data exploration since it provides easily interpretable results and a meaningful global view of the data. As such, self-organizing maps are usually applied to aircraft engine condition monitoring. As databases in this field are huge and complex, they naturally contain multiple missing entries for various reasons. The classical Kohonen algorithm to compute self-organizing maps is conceived for complete data only. A naive approach to deal with partially observed data consists in deleting items or variables with missing entries. However, this requires a sufficient number of complete individuals to be fairly representative of the population; otherwise, deletion leads to a considerable loss of information. Moreover, deletion can also induce bias in the analysis results. Alternatively, one can first apply a common imputation method to create a complete dataset and then apply the Kohonen algorithm. However, the choice of the imputation method may have a strong impact on the resulting self-organizing map. Our approach is to address simultaneously the two problems of computing a self-organizing map and imputing missing values, as these tasks are not independent. In this work, we propose an extension of self-organizing maps for partially observed data, referred to as missSOM. First, we introduce a criterion to be optimized, that aims at defining simultaneously the best self-organizing map and the best imputations for the missing entries. As such, missSOM is also an imputation method for missing values. To minimize the criterion, we propose an iterative algorithm that alternates the learning of a self-organizing map and the imputation of missing values. Moreover, we develop an accelerated version of the algorithm by entwining the iterations of the Kohonen algorithm with the updates of the imputed values. This method is efficiently implemented in R and will soon be released on CRAN. Compared to the standard Kohonen algorithm, it does not come with any additional cost in terms of computing time. Numerical experiments illustrate that missSOM performs well in terms of both clustering and imputation compared to the state of the art. In particular, it turns out that missSOM is robust to the missingness mechanism, which is in contrast to many imputation methods that are appropriate for only a single mechanism. This is an important property of missSOM as, in practice, the missingness mechanism is often unknown. An application to measurements on one type of part is also provided and shows the practical interest of missSOM.

Keywords: imputation method of missing data, partially observed data, robustness to missingness mechanism, self-organizing maps

Procedia PDF Downloads 151

Authors: Farnaz Alborzi, Angelika Schmitz, Rainer Stamminger

Abstract:

Reducing water temperatures in the wash phase of a washing programme and increasing the overall cycle durations are the latest trends in decreasing energy consumption of washing programmes. Since the implementation of the new energy efficiency classes in 2010, manufacturers seem to apply the aforementioned washing strategy with lower temperatures combined with longer programme durations extensively to realise energy-savings needed to meet the requirements of the highest energy efficiency class possible. A semi-representative on-line survey in eleven European countries (Czech Republic, Finland, France, Germany, Hungary, Italy, Poland, Romania, Spain, Sweden and the United Kingdom) was conducted by Bonn University in 2015 to shed light on consumer opinion and behaviour regarding the effects of the lower washing temperature and longer cycle duration in laundry washing on consumers’ acceptance of the programme. The risk of the long wash cycle is that consumers might not use the energy efficient Standard programmes and will think of this option as inconvenient and therefore switch to shorter, but more energy consuming programmes. Furthermore, washing in a lower temperature may lead to the problem of cross-contamination. Washing behaviour of over 5,000 households was studied in this survey to provide support and guidance for manufacturers and policy designers. Qualified households were chosen following a predefined quota: -Involvement in laundry washing: substantial, -Distribution of gender: more than 50 % female , -Selected age groups: -20–39 years, -40–59 years, -60–74 years, -Household size: 1, 2, 3, 4 and more than 4 people. Furthermore, Eurostat data for each country were used to calculate the population distribution in the respective age class and household size as quotas for the consumer survey distribution in each country. Before starting the analyses, the validity of each dataset was controlled with the aid of control questions. After excluding the outlier data, the number of the panel diminished from 5,100 to 4,843. The primary outcome of the study is European consumers are willing to save water and energy in a laundry washing but reluctant to use long programme cycles since they don’t believe that the long cycles could be energy-saving. However, the results of our survey don’t confirm that there is a relation between frequency of using Standard cotton (Eco) or Energy-saving programmes and the duration of the programmes. It might be explained by the fact that the majority of washing programmes used by consumers do not take so long, perhaps consumers just choose some additional time reduction option when selecting those programmes and this finding might be changed if the Energy-saving programmes take longer. Therefore, it may be assumed that introducing the programme duration as a new measure on a revised energy label would strongly influence the consumer at the point of sale. Furthermore, results of the survey confirm that consumers are more willing to use lower temperature programmes in order to save energy than accepting longer programme cycles and majority of them accept deviation from the nominal temperature of the programme as long as the results are good.

Keywords: duration, energy-saving, standard programmes, washing temperature

Procedia PDF Downloads 221

Authors: Noura Al Akkari, Aurélie Foucquier, Sylvain Lespinats

Abstract:

Load monitoring is a management process for energy consumption towards energy savings and energy efficiency. Non Intrusive Load Monitoring (NILM) is one method of load monitoring used for disaggregation purposes. NILM is a technique for identifying individual appliances based on the analysis of the whole residence data retrieved from the main power meter of the house. Our NILM framework starts with data acquisition, followed by data preprocessing, then event detection, feature extraction, then general appliance modeling and identification at the final stage. The event detection stage is a core component of NILM process since event detection techniques lead to the extraction of appliance features. Appliance features are required for the accurate identification of the household devices. In this research work, we aim at developing a new event detection methodology with accurate load disaggregation to extract appliance features. Time-domain features extracted are used for tuning general appliance models for appliance identification and classification steps. We use unsupervised algorithms such as Dynamic Time Warping (DTW). The proposed method relies on detecting areas of operation of each residential appliance based on the power demand. Then, detecting the time at which each selected appliance changes its states. In order to fit with practical existing smart meters capabilities, we work on low sampling data with a frequency of (1/60) Hz. The data is simulated on Load Profile Generator software (LPG), which was not previously taken into consideration for NILM purposes in the literature. LPG is a numerical software that uses behaviour simulation of people inside the house to generate residential energy consumption data. The proposed event detection method targets low consumption loads that are difficult to detect. Also, it facilitates the extraction of specific features used for general appliance modeling. In addition to this, the identification process includes unsupervised techniques such as DTW. To our best knowledge, there exist few unsupervised techniques employed with low sampling data in comparison to the many supervised techniques used for such cases. We extract a power interval at which falls the operation of the selected appliance along with a time vector for the values delimiting the state transitions of the appliance. After this, appliance signatures are formed from extracted power, geometrical and statistical features. Afterwards, those formed signatures are used to tune general model types for appliances identification using unsupervised algorithms. This method is evaluated using both simulated data on LPG and real-time Reference Energy Disaggregation Dataset (REDD). For that, we compute performance metrics using confusion matrix based metrics, considering accuracy, precision, recall and error-rate. The performance analysis of our methodology is then compared with other detection techniques previously used in the literature review, such as detection techniques based on statistical variations and abrupt changes (Variance Sliding Window and Cumulative Sum).

Keywords: electrical disaggregation, DTW, general appliance modeling, event detection

Procedia PDF Downloads 78

Authors: Muhammad Ahnaf Zahin, Yaw Adu-Gyamfi

Abstract:

Bus transit is a crucial component of transportation networks, especially in urban areas. Any intelligent transportation system must have accurate real-time information on bus travel times since it minimizes waiting times for passengers at different stations along a route, improves service reliability, and significantly optimizes travel patterns. Bus agencies must enhance the quality of their information service to serve their passengers better and draw in more travelers since people waiting at bus stops are frequently anxious about when the bus will arrive at their starting point and when it will reach their destination. For solving this issue, different models have been developed for predicting bus travel times recently, but most of them are focused on smaller road networks due to their relatively subpar performance in high-density urban areas on a vast network. This paper develops a deep learning-based architecture using a single-step multi-station forecasting approach to predict average bus travel times for numerous routes, stops, and trips on a large-scale network using heterogeneous bus transit data collected from the GTFS database. Over one week, data was gathered from multiple bus routes in Saint Louis, Missouri. In this study, Gated Recurrent Unit (GRU) neural network was followed to predict the mean vehicle travel times for different hours of the day for multiple stations along multiple routes. Historical time steps and prediction horizon were set up to 5 and 1, respectively, which means that five hours of historical average travel time data were used to predict average travel time for the following hour. The spatial and temporal information and the historical average travel times were captured from the dataset for model input parameters. As adjacency matrices for the spatial input parameters, the station distances and sequence numbers were used, and the time of day (hour) was considered for the temporal inputs. Other inputs, including volatility information such as standard deviation and variance of journey durations, were also included in the model to make it more robust. The model's performance was evaluated based on a metric called mean absolute percentage error (MAPE). The observed prediction errors for various routes, trips, and stations remained consistent throughout the day. The results showed that the developed model could predict travel times more accurately during peak traffic hours, having a MAPE of around 14%, and performed less accurately during the latter part of the day. In the context of a complicated transportation network in high-density urban areas, the model showed its applicability for real-time travel time prediction of public transportation and ensured the high quality of the predictions generated by the model.

Keywords: gated recurrent unit, mean absolute percentage error, single-step forecasting, travel time prediction.

Procedia PDF Downloads 72

Authors: R. Tamborrino, F. Rinaudo

Abstract:

Digital technologies offer possibilities to effectively convert historical archives into instruments of knowledge able to provide a guide for the interpretation of historical phenomena. Digital conversion and management of those documents allow the possibility to add other sources in a unique and coherent model that permits the intersection of different data able to open new interpretations and understandings. Urban history uses, among other sources, the inventories that register human activities in a specific space (e.g. cadastres, censuses, etc.). The geographic localisation of that information inside cartographic supports allows for the comprehension and visualisation of specific relationships between different historical realities registering both the urban space and the peoples living there. These links that merge the different nature of data and documentation through a new organisation of the information can suggest a new interpretation of other related events. In all these kinds of analysis, the use of GIS platforms today represents the most appropriate answer. The design of the related databases is the key to realise the ad-hoc instrument to facilitate the analysis and the intersection of data of different origins. Moreover, GIS has become the digital platform where it is possible to add other kinds of data visualisation. This research deals with the industrial development of Turin at the beginning of the 20th century. A census of factories realized just prior to WWI provides the opportunity to test the potentialities of GIS platforms for the analysis of urban landscape modifications during the first industrial development of the town. The inventory includes data about location, activities, and people. GIS is shaped in a creative way linking different sources and digital systems aiming to create a new type of platform conceived as an interface integrating different kinds of data visualisation. The data processing allows linking this information to an urban space, and also visualising the growth of the city at that time. The sources, related to the urban landscape development in that period, are of a different nature. The emerging necessity to build, enlarge, modify and join different buildings to boost the industrial activities, according to their fast development, is recorded by different official permissions delivered by the municipality and now stored in the Historical Archive of the Municipality of Turin. Those documents, which are reports and drawings, contain numerous data on the buildings themselves, including the block where the plot is located, the district, and the people involved such as the owner, the investor, and the engineer or architect designing the industrial building. All these collected data offer the possibility to firstly re-build the process of change of the urban landscape by using GIS and 3D modelling technologies thanks to the access to the drawings (2D plans, sections and elevations) that show the previous and the planned situation. Furthermore, they access information for different queries of the linked dataset that could be useful for different research and targets such as economics, biographical, architectural, or demographical. By superimposing a layer of the present city, the past meets to the present-industrial heritage, and people meet urban history.

Keywords: digital urban history, census, digitalisation, GIS, modelling, digital humanities

Procedia PDF Downloads 191

Authors: Lee Tasker, Ali Karrech, Jeffrey Shragge, Matthew Josh

Abstract:

Monitoring for the deterioration of concrete infrastructure is an important assessment tool for an engineer and difficulties can be experienced with monitoring for deterioration within an infrastructure. If a failure crack, or fluid seepage through such a crack, is observed from the surface often the source location of the deterioration is not known. Geophysical methods are used to assist engineers with assessing the subsurface conditions of materials. Techniques such as Ground Penetrating Radar (GPR) provide information on the location of buried infrastructure such as pipes and conduits, positions of reinforcements within concrete blocks, and regions of voids/cavities behind tunnel lining. This experiment underlines the application of GPR as an infrastructure-monitoring tool to highlight and monitor regions of possible deterioration within a concrete test wall due to an increase in the generation of fractures; in particular, during a time period of applied load to a concrete wall up to and including structural failure. A three-point load was applied to a concrete test wall of dimensions 1700 x 600 x 300 mm³ in increments of 10 kN, until the wall structurally failed at 107.6 kN. At each increment of applied load, the load was kept constant and the wall was scanned using GPR along profile lines across the wall surface. The measured radar amplitude responses of the GPR profiles, at each applied load interval, were reconstructed into depth-slice grids and presented at fixed depth-slice intervals. The corresponding depth-slices were subtracted from each data set to compare the radar amplitude response between datasets and monitor for changes in the radar amplitude response. At lower values of applied load (i.e., 0-60 kN), few changes were observed in the difference of radar amplitude responses between data sets. At higher values of applied load (i.e., 100 kN), closer to structural failure, larger differences in radar amplitude response between data sets were highlighted in the GPR data; up to 300% increase in radar amplitude response at some locations between the 0 kN and 100 kN radar datasets. Distinct regions were observed in the 100 kN difference dataset (i.e., 100 kN-0 kN) close to the location of the final failure crack. The key regions observed were a conical feature located between approximately 3.0-12.0 cm depth from surface and a vertical linear feature located approximately 12.1-21.0 cm depth from surface. These key regions have been interpreted as locations exhibiting an increased change in pore-space due to increased mechanical loading, or locations displaying an increase in volume of micro-cracks, or locations showing the development of a larger macro-crack. The experiment showed that GPR is a useful geophysical monitoring tool to assist engineers with highlighting and monitoring regions of large changes of radar amplitude response that may be associated with locations of significant internal structural change (e.g. crack development). GPR is a non-destructive technique that is fast to deploy in a production setting. GPR can assist with reducing risk and costs in future infrastructure maintenance programs by highlighting and monitoring locations within the structure exhibiting large changes in radar amplitude over calendar-time.

Keywords: 4D GPR, engineering geophysics, ground penetrating radar, infrastructure monitoring

Procedia PDF Downloads 179