Search results for: generalized autoregressive score model

Authors: Pei Yi Lin

Abstract:

Objective: The objective of this study is to use machine learning methods to build an early prediction classifier model for acute delirium to improve the quality of medical care for intensive care patients. Background: Delirium is a common acute and sudden disturbance of consciousness in critically ill patients. After the occurrence, it is easy to prolong the length of hospital stay and increase medical costs and mortality. In 2021, the incidence of delirium in the intensive care unit of internal medicine was as high as 59.78%, which indirectly prolonged the average length of hospital stay by 8.28 days, and the mortality rate is about 2.22% in the past three years. Therefore, it is expected to build a delirium prediction classifier through big data analysis and machine learning methods to detect delirium early. Method: This study is a retrospective study, using the artificial intelligence big data database to extract the characteristic factors related to delirium in intensive care unit patients and let the machine learn. The study included patients aged over 20 years old who were admitted to the intensive care unit between May 1, 2022, and December 31, 2022, excluding GCS assessment <4 points, admission to ICU for less than 24 hours, and CAM-ICU evaluation. The CAMICU delirium assessment results every 8 hours within 30 days of hospitalization are regarded as an event, and the cumulative data from ICU admission to the prediction time point are extracted to predict the possibility of delirium occurring in the next 8 hours, and collect a total of 63,754 research case data, extract 12 feature selections to train the model, including age, sex, average ICU stay hours, visual and auditory abnormalities, RASS assessment score, APACHE-II Score score, number of invasive catheters indwelling, restraint and sedative and hypnotic drugs. Through feature data cleaning, processing and KNN interpolation method supplementation, a total of 54595 research case events were extracted to provide machine learning model analysis, using the research events from May 01 to November 30, 2022, as the model training data, 80% of which is the training set for model training, and 20% for the internal verification of the verification set, and then from December 01 to December 2022 The CU research event on the 31st is an external verification set data, and finally the model inference and performance evaluation are performed, and then the model has trained again by adjusting the model parameters. Results: In this study, XG Boost, Random Forest, Logistic Regression, and Decision Tree were used to analyze and compare four machine learning models. The average accuracy rate of internal verification was highest in Random Forest (AUC=0.86), and the average accuracy rate of external verification was in Random Forest and XG Boost was the highest, AUC was 0.86, and the average accuracy of cross-validation was the highest in Random Forest (ACC=0.77). Conclusion: Clinically, medical staff usually conduct CAM-ICU assessments at the bedside of critically ill patients in clinical practice, but there is a lack of machine learning classification methods to assist ICU patients in real-time assessment, resulting in the inability to provide more objective and continuous monitoring data to assist Clinical staff can more accurately identify and predict the occurrence of delirium in patients. It is hoped that the development and construction of predictive models through machine learning can predict delirium early and immediately, make clinical decisions at the best time, and cooperate with PADIS delirium care measures to provide individualized non-drug interventional care measures to maintain patient safety, and then Improve the quality of care.

Keywords: critically ill patients, machine learning methods, delirium prediction, classifier model

Procedia PDF Downloads 80

Authors: Nida Amin, Fahad Tanveer, Hina Shabbir, Ayesha Saeed, Attiqa Riaz

Abstract:

The gradual progression of corneal disorder keratoconus significantly impairs eyesight and quality of life, increasing the likelihood of depression. Using the Hospital Anxiety and Depression Scale (HADS) at the AL-Ibrahim Eye Hospital in Karachi, this study aimed to evaluate the occurrence of depression and anxiety symptoms in patients with keratoconus and to suggest better treatment. A descriptive-analytical study was conducted at Al-Ibrahim Eye Hospital Karachi from March to April 2022, and patients diagnosed with symptomatic keratoconus were recruited using a non-probability convenient sampling technique. After obtaining written informed consent from patients, keratoconus severity was assessed using visual acuity and corneal topography. Symptoms of anxiety and depression were assessed using the Hospital Anxiety and Depression (HADS) Scale. The data were analyzed using SPSS version 20.0. Spearman correlation coefficient. Of the 108 participants, 60 (56%) were female and 48 (44%) were male. Using the HADS scale, 44 (40.7%) patients were classified as normal with a HADS score of (0-7), 23 (21.3%) as borderline with a HADS score of (8-10) and 41 (38%) patients were diagnosed with anxiety and depression with a HADS score of (11-21). Depression and anxiety are highly prevalent among patients in advanced stages of the disease.

Keywords: cornea, keratoconus, anxiety, depression, corneal topography, mental health

Procedia PDF Downloads 42

Authors: M. A. Ansari, A. Hussain, A. Uddin

Abstract:

A side weir is a flow diversion structure provided in the side wall of a channel to divert water from the main channel to a branch channel. The trapezoidal labyrinth weir is a special type of weir in which crest length of the weir is increased to pass higher discharge. Experimental and numerical studies related to the coefficient of discharge of trapezoidal labyrinth weir in an open channel have been presented in the present study. Group Method of Data Handling (GMDH) with the transfer function of quadratic polynomial has been used to predict the coefficient of discharge for the side trapezoidal labyrinth weir. A new model is developed for coefficient of discharge of labyrinth weir by regression method. Generalized models for predicting the coefficient of discharge for labyrinth weir using Group Method of Data Handling (GMDH) network have also been developed. The prediction based on GMDH model is more satisfactory than those given by traditional regression equations.

Keywords: discharge coefficient, group method of data handling, open channel, side labyrinth weir

Procedia PDF Downloads 163

Authors: Oktavian Prasetia Wardana, Martono Tri Utomo, Risa Etika, Kartika Darma Handayani, Dina Angelika, Wurry Ayuningtyas

Abstract:

Background: Critically ill neonates are newborn babies with high-risk factors that potentially cause disability and/or death. Scoring systems for determining the severity of the disease have been widely developed as well as some designs for use in neonates. The SNAPPE-II method, which has been used as a mortality predictor scoring system in several referral centers, was found to be slow in assessing the outcome of critically ill neonates in the Neonatal Intensive Care Unit (NICU). Objective: To analyze the predictive value of MSNS on the outcome of critically ill neonates at the time of arrival up to 24 hours after being admitted to the NICU. Methods: A longitudinal observational analytic study based on medical record data was conducted from January to August 2022. Each sample was recorded from medical record data, including data on gestational age, mode of delivery, APGAR score at birth, resuscitation measures at birth, duration of resuscitation, post-resuscitation ventilation, physical examination at birth (including vital signs and any congenital abnormalities), the results of routine laboratory examinations, as well as the neonatal outcomes. Results: This study involved 105 critically ill neonates who were admitted to the NICU. The outcome of critically ill neonates was 50 (47.6%) neonates died, and 55 (52.4%) neonates lived. There were more males than females (61% vs. 39%). The mean gestational age of the subjects in this study was 33.8 ± 4.28 weeks, with the mean birth weight of the subjects being 1820.31 ± 33.18 g. The mean MSNS score of neonates with a deadly outcome was lower than that of the lived outcome. ROC curve with a cut point MSNS score <10.5 obtained an AUC of 93.5% (95% CI: 88.3-98.6) with a sensitivity value of 84% (95% CI: 80.5-94.9), specificity 80 % (CI 95%: 88.3-98.6), Positive Predictive Value (PPV) 79.2%, Negative Predictive Value (NPV) 84.6%, Risk Ratio (RR) 5.14 with Hosmer & Lemeshow test results p>0.05. Conclusion: The MSNS score has a good predictive value and good calibration of the outcomes of critically ill neonates admitted to the NICU.

Keywords: critically ill neonate, outcome, MSNS, NICU, predictive value

Procedia PDF Downloads 73

Authors: Mohammad Maruf Hasan

Abstract:

Foreign Direct Investment (FDI) has been increased at a remarkable position around the globe in which emerging economies are getting more FDI compared to industrialized economies. This study aims to examine the determinants of inward FDI flows in Bangladesh. To estimate the long and short-run impact of the FDI determinants for 1996-2020, we employed the Autoregressive-Distributed Lag (ARDL) model. Results show that: (1) macroeconomic determinants, such as economic growth, infrastructure, and market size, have a significant and strong positive effect.(2) Inflation exchange rate shows insignificant effects, while trade openness has mixed (short-run negative, long-run positive) effects on FDI inflows in both the long and short run. (3) Current institutional determinants rule of law has a positive effect on FDI inflows but is statistically insignificant, political stability has a negative, and the rule of law has a considerable beneficial impact on inflows of FDI. (4) The macroeconomic factors have been determined to impact Bangladesh's FDI inflows. Finally, a stable macroeconomic climate is more effective at luring FDI, as this study confirms. From a policy perspective, this study will help the government and policymakers to make a new investment policy.

Keywords: determinants, FDI, ARDL, Bangladesh

Procedia PDF Downloads 76

Authors: Veronica Sri Lestari, Sitti Nurani Sirajuddin, Kasmiyati Kasim

Abstract:

The purpose of this research was to know the attitude of beef cattle farmers toward bio security practices. This research was conducted in Barru regency, South Sulawesi province, Indonesia, in 2014. Thirty beef cattle farmers were selected through random sampling. Primary and secondary data were collected through report, observation and deep interview by using questionnaire. Bio security practices consisted of 35 questions. Every answer of the question was scored based on three categories: score 1 (not important), score 2 (important) and 3 (very important). The results of this research showed that the attitude of beef cattle farmers toward bio security practices was categorized as important.

Keywords: attitude, beef cattle, biosecurity, farmers

Procedia PDF Downloads 299

Authors: Fatma Abdel Moneim Al Tawil

Abstract:

This study was undertaken to evaluate an educational program regarding scientific writing and presentation skills among university students. This interventional study used a one-group, pretest/posttest design and was conducted in Al Jouf University among four colleges in Saudi Arabia. Baseline students’ assessment was conducted for developing educational program. Interventional, one group, pretest/posttest study was designed to evaluate the effectiveness of the educational program. Three parts evaluation sheet with total scores of 30 was used for 113 students for the development of the program and 52 students for test pretest phase. Wilcoxon signed ranks showed statistically significant improvement in the combined overall program skills score from a median of 56.7 pre to a median of 86.7 post, (z = 6.231, p < 0.001). When compared to preprogram intervention, post interventions 51.9 % of students achieve excellent performance. While pre intervention no students (0.0 %) achieve this score. Regarding to scientific writing skills, Wilcoxon signed ranks showed statistically significant improvement in the score from a median of 60 pre to a median of 90 post, (z = 6.122, p < 0.001). None of students had excellent performance changed to 73.1%. Regarding to oral presentation skills, Wilcoxon signed ranks showed statistically significant improvement in the score from a median of 50 pre to a median of 80 post, (z = 6.153, p < 0.001). None of students had excellent performance changed to 48.1%. Such educational program needs to be incorporated into classroom delivery of the students’ curriculum. Scientific writing skills book needed to be developed to be recommended as a basic educational strategy for all university faculties.

Keywords: scientific writing, presentation skills, university students, educational program

Procedia PDF Downloads 454

Authors: Baoubadi Atozou, Koffi Akakpo

Abstract:

Since the decade 2000, the use of foodstuffs such as corn, wheat, and soybeans in biofuel production has been growing sharply in the United States, Canada, and Europe. Thus, prices for these agricultural products are rising in the world market. These cereals are the most important source of calorific energy for West African Economic and Monetary Union (WAEMU) countries members’ population. These countries are highly dependent on imports of most of these products. Thereby, rising prices can have an important impact on import levels and consequently on food security in these countries. This study aims to analyze the interrelationship between the prices of these commodities and their volatilities, and their effects on imports of these agricultural products by each WAEMU ’country member. The Autoregressive Distributed Lag (ARDL) model, the GARCH Multivariate model, and the Granger Causality Test are used in this investigation. The results show that import levels are highly and significantly sensitive to price changes as well as their volatility. In the short term as well as in the long term, there is a significant relationship between the prices of these products. There is a positive relationship in general between price volatility. And these volatilities have negative effects on the level of imports. The market characteristics affect food security in these countries, especially access to food for vulnerable and low-income populations. The policies makers must adopt viable strategies to increase agricultural production and limit their dependence on imports.

Keywords: price volatility, import of agricultural products, food safety, WAEMU

Procedia PDF Downloads 197

Authors: Michalis Linardakis, Vasilis Grammatikopoulos, Athanasios Gregoriadis, Kalliopi Trouli

Abstract:

Discrete choice models belong to the family of Conjoint analysis that are applied on the preferences of the respondents towards a set of scenarios that describe alternative choices. The scenarios have been pre-designed to cover all the attributes of the alternatives that may affect the choices. In this study, we examine how preschool educators integrate physical activities into their everyday teaching practices through the use of discrete choice models. One of the advantages of discrete choice models compared to other more traditional data collection methods (e.g. questionnaires and interviews that use ratings) is that the respondent is called to select among competitive and realistic alternatives, rather than objectively rate each attribute that the alternatives may have. We present the effort to construct and choose representative attributes that would cover all possible choices of the respondents, and the scenarios that have arisen. For the purposes of the study, we used a sample of 50 preschool educators in Greece that responded to 4 scenarios (from the total of 16 scenarios that the orthogonal design resulted), with each scenario having three alternative teaching practices. Seven attributes of the alternatives were used in the scenarios. For the analysis of the data, we used multinomial logit model with random effects, multinomial probit model and generalized mixed logit model. The conclusions drawn from the estimated parameters of the models are discussed.

Keywords: conjoint analysis, discrete choice models, educational data, multivariate statistical analysis

Procedia PDF Downloads 469

Authors: Sophia Shi

Abstract:

Acute kidney injury (AKI) is the sudden onset of kidney damage in which the kidneys cannot filter waste from the blood, requiring emergency hospitalization. AKI patient mortality rate is high in the ICU and is virtually impossible for doctors to predict because it is so unexpected. Currently, there is no hybrid model predicting AKI that takes advantage of two types of data. De-identified patient data from the MIMIC-III database and de-identified kidney images and corresponding patient records from the Beijing Hospital of the Ministry of Health were collected. Using data features including serum creatinine among others, two numeric models using MIMIC and Beijing Hospital data were built, and with the hospital ultrasounds, an image-only model was built. Convolutional neural networks (CNN) were used, VGG and Resnet for numeric data and Resnet for image data, and they were combined into a hybrid model by concatenating feature maps of both types of models to create a new input. This input enters another CNN block and then two fully connected layers, ending in a binary output after running through Softmax and additional code. The hybrid model successfully predicted AKI and the highest AUROC of the model was 0.953, achieving an accuracy of 90% and F1-score of 0.91. This model can be implemented into urgent clinical settings such as the ICU and aid doctors by assessing the risk of AKI shortly after the patient’s admission to the ICU, so that doctors can take preventative measures and diminish mortality risks and severe kidney damage.

Keywords: Acute kidney injury, Convolutional neural network, Hybrid deep learning, Patient record data, ResNet, Ultrasound kidney images, VGG

Procedia PDF Downloads 135

Authors: Qing Wu, Xuchun Ye, Qiuchen Wang, Kirsten Corazzini

Abstract:

Objective: It’s increasing acknowledged that patient engagement in healthcare and health literacy both have positive impact on patient outcome. Health literacy emphasizes the ability of individuals to understand and apply health information and manage health. Patients' health literacy affected their willingness to participate in decision-making, but its impact on the behavior and willingness of patient engagement in healthcare is not clear, especially in China. Therefore, this study aimed to explore the correlation between the behavior and willingness of patient engagement and health literacy. Methods: A cross-sectional survey was employed using the behavior and willingness of patient engagement in healthcare questionnaire, Chinese version All Aspects of Health Literacy Scale (AAHLS). A convenient sample of 443 patients was recruited from 8 general hospitals in Shanghai, Jiangsu Province and Zhejiang Province, from September 2016 to January 2017. Results: The mean score for the willingness was (4.41±0.45), and the mean score for the patient engagement behavior was (4.17±0.49); the mean score for the patient's health literacy was (2.36±0.29)，the average score of its three dimensions- the functional literacy, the Communicative/interactive literacy and the Critical literacy, was (2.26±0.38), (2.28±0.42), and (2.61±0.43), respectively. Patients' health literacy was positively correlated with their willingness of engagement (r = 0.367, P < 0.01), and positively correlated with patient engagement behavior (r = 0.357, P < 0.01). All dimensions of health literacy were positively correlated with the behavior and willingness of patient engagement in healthcare; the dimension of Communicative/interactive literacy (r = 0.312, P < 0.01; r = 0.357, P < 0.01) and the Critical literacy (r = 0.357, P < 0.01; r = 0.357, P < 0.01) are more relevant to the behavior and willingness than the dimension of basic/functional literacy (r=0.150, P < 0.01; r = 0.150, P < 0.01). Conclusions: The behavior and willingness of patient engagement in healthcare are positively correlated with health literacy and its dimensions. In clinical work, medical staff should pay attention to patients’ health literacy, especially the situation that low literacy leads to low participation and provide health information to patients through health education or communication to improve their health literacy as well as guide them to actively and rationally participate in their own health care.

Keywords: patient engagement, health literacy, healthcare, correlation

Procedia PDF Downloads 169

Authors: Raman Sharma, S. K. Jain

Abstract:

Aims and Objectives: Fixation of the mesh during laparoscopic total extraperitoneal (TEP) repair of inguinal hernia is thought to be necessary to prevent recurrence. However, mesh fixation may increase surgical complications and postoperative pain. Our objective was to compare the outcomes of nonfixation with fixation of polypropylene mesh by metal tacks during TEP repair of inguinal hernia. Methods: Forty patients aged 18 to72 years with inguinal hernia were included who underwent laparoscopic TEP repair of inguinal hernia with (n=20) or without (n=20) fixation of the mesh. The outcomes were operative duration, postoperative pain score, cost, in-hospital stay, time to return to normal activity, and complications. Results: Patients in whom the mesh was not fixed had shorter mean operating time (p < 0.05). We found no difference between groups in the postoperative pain score, incidence of recurrence, in-hospital stay, time to return to normal activity and complications (P > 0.05). Moreover, a net cost savings was realized for each hernia repair performed without stapled mesh. Conclusions: TEP repair without mesh fixation resulted in the shorter operating time and lower operative cost with no difference between groups in the postoperative pain score, incidence of recurrence, in-hospital stay, time to return to normal activity and complications. All this contribute to make TEP repair without mesh fixation a better choice for repair of uncomplicated inguinal hernia, especially in developing nations with scarce resources.

Keywords: postoperative pain score, inguinal hernia, nonfixation of mesh, total extra peritoneal (TEP)

Procedia PDF Downloads 347

Authors: Yishu Gong, Liangliang Yang, Jianyu Zhang, Zhengyu Chen, Sihong He, Xusheng Zhang, Wei Zhang

Abstract:

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that affects millions of people worldwide and is characterized by cognitive decline and behavioral changes. People living with Alzheimer’s disease often find it hard to complete routine tasks. However, there are limited objective assessments that aim to quantify the difficulty of certain tasks for AD patients compared to non-AD people. In this study, we propose to use speech emotion recognition (SER), especially the frustration level, as a potential biomarker for quantifying the difficulty patients experience when describing a picture. We build an SER model using data from the IEMOCAP dataset and apply the model to the DementiaBank data to detect the AD/non-AD group difference and perform longitudinal analysis to track the AD disease progression. Our results show that the frustration level detected from the SER model can possibly be used as a cost-effective tool for objective tracking of AD progression in addition to the Mini-Mental State Examination (MMSE) score.

Keywords: Alzheimer’s disease, speech emotion recognition, longitudinal biomarker, machine learning

Procedia PDF Downloads 117

Authors: Deepika Christopher, Garima Anand

Abstract:

To determine the best model for churn prediction in the telecom industry, this paper compares 11 machine learning algorithms, namely Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, XGBoost, LightGBM, Cat Boost, AdaBoost, Extra Trees, Deep Neural Network, and Hybrid Model (MLPClassifier). It also aims to pinpoint the top three factors that lead to customer churn and conducts customer segmentation to identify vulnerable groups. According to the data, the Logistic Regression model performs the best, with an F1 score of 0.6215, 81.76% accuracy, 68.95% precision, and 56.57% recall. The top three attributes that cause churn are found to be tenure, Internet Service Fiber optic, and Internet Service DSL; conversely, the top three models in this article that perform the best are Logistic Regression, Deep Neural Network, and AdaBoost. The K means algorithm is applied to establish and analyze four different customer clusters. This study has effectively identified customers that are at risk of churn and may be utilized to develop and execute strategies that lower customer attrition.

Keywords: attrition, retention, predictive modeling, customer segmentation, telecommunications

Procedia PDF Downloads 62

Authors: Nopadol Burananat, Kedsara Tripaichayonsak

Abstract:

This research aims to study the level of expectations and participation of personnel in implementing educational quality assurance of programs in Graduate School, Rajabhat Suan Sunandha University. The sample used in this study is 60 participants. The tool used for data collection is a questionnaire constructed by the researcher. The analysis is done by frequency, percentage, mean and standard deviation. It was found that the level of expectations personnel in Graduate School, Suan Sunandha Rajabhat University in implementing educational quality assurance is at high level. The category which received the most score is Action, followed by Check, Do and Plan, respectively. For the level of participation of personnel at program level of Graduate School, Suan Sunandha Rajabhat University in implementing educational quality assurance, the overall score is at high level. The category which received the most score is Action, followed by Do, Check and Plan, respectively.

Keywords: participation, implementation of educational quality assurance, educational quality assurance, expectations and participation

Procedia PDF Downloads 388

Authors: Harsh Thakkar, Ganesh Iyer

Abstract:

In this paper, we survey various user-centered or context-based biomedical health information retrieval systems. We present and discuss the performance of systems submitted in CLEF eHealth 2014 Task 3 for this purpose. We classify and focus on comparing the two most prevalent retrieval models in biomedical information retrieval namely: Language Model (LM) and Vector Space Model (VSM). We also report on the effectiveness of using external medical resources and ontologies like MeSH, Metamap, UMLS, etc. We observed that the LM based retrieval systems outperform VSM based systems on various fronts. From the results we conclude that the state-of-art system scores for MAP was 0.4146, P@10 was 0.7560 and NDCG@10 was 0.7445, respectively. All of these score were reported by systems built on language modeling approaches.

Keywords: clinical document retrieval, concept-based information retrieval, query expansion, language models, vector space models

Procedia PDF Downloads 323

Authors: Heba Esaily, Rawhia Eledl, Daila Aboelela, Rasha Noreldin

Abstract:

Objective: Assess serum level of Transforming growth factor beta 2 (TGF-β2) and Matrilin-3 (MATN3) SNP6 polymorphism in osteoarthritic patients Background: Osteoarthritis (OA) is a musculoskeletal disease characterized by pain and joint stiffness. TGF-β 2 is involved in chondrogenesis and osteogenesis, It has found that MATN3 gene and protein expression was correlated with the extent of tissue damage in OA. Findings suggest that regulation of MATN3 expression is essential for maintenance of the cartilage extracellular matrix microenvironment Subjects and Methods: 72 cases of primary OA (56 with knee OA and 16 with generalized OA were compared with that of 18 healthy controls. Radiographs were scored with the Kellgren-Lawrence scale. Serum TGF-β2 was measured by using (ELISA), levels of marker were correlated to radiographic grading of disease and MATN3 SNP6 polymorphism was determined by (PCR-RFLP). Results: MATN3 SNP6 polymorphism and serum level of TGF-β2 were higher in OA compared with controls. Genotype, NN and N allele frequency were higher in patients with OA compared with controls. NN genotype and N allele frequency were higher in knee osteoarthritis than generalized OA. Significant positive correlation between level of TGFβ2 and radiographic grading in group with knee OA, but no correlation between serum level of TGFβ2 and radiographic grading in generalized OA. Conclusion: MATN3 SNP6 polymorphism and TGF-β2 implicated in the pathogenesis of osteoarthritis. Association of N/N genotype with primary osteoarthritis emphasizes on the need for prospective study include larger sample size to confirm the results of the present study.

Keywords: Matrilin-3, transforming growth factor beta 2, primary osteoarthritis, knee osteoarthritis

Procedia PDF Downloads 271

Authors: Zhidong Zhang, Zhi-Chao Zhang, Georgianna Duarte

Abstract:

The social problems of early adolescents in Northeast China were examined with the instrument of Achenbach System of Empirically Based Assessment (ASEBA). In this study, the data consisted of 2532 early adolescents. The relevant variables such as sports activities, hobbies, chores and the number of close friends, as independent variables have been included in this study. The stratified sampling method was used to collect data from 2532 participants. The analysis results indicated that sports activities, hobbies, chores and the number of close friends, as predictors can be used in a predictive model, which significantly predict the social problem T-score.

Keywords: social problems, ASEBA, early adolescents, predictive Model

Procedia PDF Downloads 351

Authors: Nurul Mazni Abdullah, Saw Kian Cheah, Raha Abdul Rahman, Qurratu 'Aini Musthafa

Abstract:

Purpose: Candidemia was associated with high mortality in critically ill patients. Early candidemia prediction is imperative for preemptive antifungal treatment. This study aimed to externally validate the candidemia risk prediction scores by Jameran et al. (2021) by identifying risk factors of acute kidney injury, renal replacement therapy, parenteral nutrition, and multifocal candida colonization. Methods: This single-center, retrospective observational study included all critically ill patients admitted to the intensive care unit (ICU) in a tertiary referral center from January 2018 to December 2023. The study evaluated the candidemia risk prediction score performance by analyzing the occurrence of candidemia within the study period. Patients’ demographic characteristics, comorbidities, SOFA scores, and ICU outcomes were analyzed. Patients who were diagnosed with candidemia before ICU admission were excluded. Results: A total of 500 patients were analyzed with 2 dropouts due to incomplete data. Validation analysis showed that the candidemia risk prediction score has a sensitivity of 75.00% (95% CI: 59.66-86.81), specificity of 65.35% (95% CI: 60.78-69.72), positive predictive value of 17.28, and negative predictive value of 96.44. The incidence of candidemia was 8.86% with no significant differences in the demographic and comorbidities except higher SOFA scoring in the candidemia group. The candidemia group showed significantly longer ICU and hospital LOS and higher ICU and in-hospital mortality. Conclusion: This study concluded the candidemia risk prediction score by Jameran et al (2021) had good sensitivity and a high negative prediction value.

Keywords: candidemia, intensive care, clinical prediction rule, incidence

Procedia PDF Downloads 26

Authors: Semachew M. Kassa, Africa M Geremew, Tezera F. Azmatch, Nandyala Darga Kumar

Abstract:

Frequency ratio (FR) and analytical hierarchy process (AHP) methods are developed based on past landslide failure points to identify the landslide susceptibility mapping because landslides can seriously harm both the environment and society. However, it is still difficult to select the most efficient method and correctly identify the main driving factors for particular regions. In this study, we used fourteen landslide conditioning factors (LCFs) and five soft computing algorithms, including Random Forest (RF), Support Vector Machine (SVM), Logistic Regression (LR), Artificial Neural Network (ANN), and Naïve Bayes (NB), to predict the landslide susceptibility at 12.5 m spatial scale. The performance of the RF (F1-score: 0.88, AUC: 0.94), ANN (F1-score: 0.85, AUC: 0.92), and SVM (F1-score: 0.82, AUC: 0.86) methods was significantly better than the LR (F1-score: 0.75, AUC: 0.76) and NB (F1-score: 0.73, AUC: 0.75) method, according to the classification results based on inventory landslide points. The findings also showed that around 35% of the study region was made up of places with high and very high landslide risk (susceptibility greater than 0.5). The very high-risk locations were primarily found in the western and southeastern regions, and all five models showed good agreement and similar geographic distribution patterns in landslide susceptibility. The towns with the highest landslide risk include Amhara Saint Town's western part, the Northern part, and St. Gebreal Church villages, with mean susceptibility values greater than 0.5. However, rainfall, distance to road, and slope were typically among the top leading factors for most villages. The primary contributing factors to landslide vulnerability were slightly varied for the five models. Decision-makers and policy planners can use the information from our study to make informed decisions and establish policies. It also suggests that various places should take different safeguards to reduce or prevent serious damage from landslide events.

Keywords: artificial neural network, logistic regression, landslide susceptibility, naïve Bayes, random forest, support vector machine

Procedia PDF Downloads 87

Authors: Anjali Chouksey, Riya Nimje, Jahanvi Saraf

Abstract:

Online transactions have increased dramatically in this new ‘social-distancing’ era. With online transactions, Fraud in online payments has also increased significantly. Frauds are a significant problem in various industries like insurance companies, baking, etc. These frauds include leaking sensitive information related to the credit card, which can be easily misused. Due to the government also pushing online transactions, E-commerce is on a boom. But due to increasing frauds in online payments, these E-commerce industries are suffering a great loss of trust from their customers. These companies are finding credit card fraud to be a big problem. People have started using online payment options and thus are becoming easy targets of credit card fraud. In this research paper, we will be discussing machine learning algorithms. We have used a decision tree, XGBOOST, k-nearest neighbour, logistic-regression, random forest, and SVM on a dataset in which there are transactions done online mode using credit cards. We will test all these algorithms for detecting fraud cases using the confusion matrix, F1 score, and calculating the accuracy score for each model to identify which algorithm can be used in detecting frauds.

Keywords: machine learning, fraud detection, artificial intelligence, decision tree, k nearest neighbour, random forest, XGBOOST, logistic regression, support vector machine

Procedia PDF Downloads 153

Authors: Yih-Wenn Laih

Abstract:

In recent times, the dependence and co-movement among international financial markets have become stronger than in the past, as evidenced by commentaries in the news media and the financial sections of newspapers. Studying the co-movement between returns in financial markets is an important issue for portfolio management and risk management. The realization of co-movement helps investors to identify the opportunities for international portfolio management in terms of asset allocation and pricing. Since the election of the new Prime Minister, Shinzo Abe, in November 2012, the yen has weakened against the US dollar from the 80 to the 120 level. The policies, known as “Abenomics,” are to encourage private investment through a more aggressive mix of monetary and fiscal policy. Given the close economic relations and competitions among Asia markets, it is interesting to discover the co-movement relations, affected by the depreciation of yen, between stock market of Japan and 5 major Asia stock markets, including China, Hong Kong, Korea, Singapore, and Taiwan. Specifically, we devote ourselves to measure the co-movement of stock markets between Japan and each one of the 5 Asia stock markets in terms of rank correlation coefficients. To compute the coefficients, return series of each stock market is first fitted by a skewed-t GARCH (generalized autoregressive conditional heteroscedasticity) model. Secondly, to measure the dependence structure between matched stock markets, we employ the symmetrized Joe-Clayton (SJC) copula to calculate the probability density function of paired skewed-t distributions. The joint probability density function is then utilized as the scoring scheme to optimize the sequence alignment by dynamic programming method. Finally, we compute the rank correlation coefficients (Kendall's  and Spearman's ) between matched stock markets based on their aligned sequences. We collect empirical data of 6 stock indexes from Taiwan Economic Journal. The data is sampled at a daily frequency covering the period from January 1, 2013 to July 31, 2015. The empirical distributions of returns indicate fatter tails than the normal distribution. Therefore, the skewed-t distribution and SJC copula are appropriate for characterizing the data. According to the computed Kendall’s τ, Korea has the strongest co-movement relation with Japan, followed by Taiwan, China, and Singapore; the weakest is Hong Kong. On the other hand, the Spearman’s ρ reveals that the strength of co-movement between markets with Japan in decreasing order are Korea, China, Taiwan, Singapore, and Hong Kong. We explore the effects of “Abenomics” on Asia stock markets by measuring the co-movement relation between Japan and five major Asia stock markets in terms of rank correlation coefficients. The matched markets are aligned by a hybrid method consisting of GARCH, copula and sequence alignment. Empirical experiments indicate that Korea has the strongest co-movement relation with Japan. The strength of China and Taiwan are better than Singapore. The Hong Kong market has the weakest co-movement relation with Japan.

Keywords: co-movement, depreciation of Yen, rank correlation, stock market

Procedia PDF Downloads 232

Authors: Danielle Shackley, Yetunde Folajimi

Abstract:

As more people turn to the internet seeking health-related information, there is more risk of finding false, inaccurate, or dangerous information. Sentiment analysis is a natural language processing technique that assigns polarity scores to text, ranging from positive, neutral, and negative. In this research, we evaluate the weight of a sentiment analysis feature added to fake health news classification models. The dataset consists of existing reliably labeled health article headlines that were supplemented with health information collected about COVID-19 from social media sources. We started with data preprocessing and tested out various vectorization methods such as Count and TFIDF vectorization. We implemented 3 Naive Bayes classifier models, including Bernoulli, Multinomial, and Complement. To test the weight of the sentiment analysis feature on the dataset, we created benchmark Naive Bayes classification models without sentiment analysis, and those same models were reproduced, and the feature was added. We evaluated using the precision and accuracy scores. The Bernoulli initial model performed with 90% precision and 75.2% accuracy, while the model supplemented with sentiment labels performed with 90.4% precision and stayed constant at 75.2% accuracy. Our results show that the addition of sentiment analysis did not improve model precision by a wide margin; while there was no evidence of improvement in accuracy, we had a 1.9% improvement margin of the precision score with the Complement model. Future expansion of this work could include replicating the experiment process and substituting the Naive Bayes for a deep learning neural network model.

Keywords: sentiment analysis, Naive Bayes model, natural language processing, topic analysis, fake health news classification model

Procedia PDF Downloads 101

Authors: Lina Wu, Ye Li

Abstract:

An equivalent relation between a harmonic form and a closed co-closed form is established on a complete non-compact manifold. This equivalence has been generalized for a differential k-form ω from Lq spaces to non-Lq spaces when q=2 in the context of p-balanced growth where p=2. Especially for a simple differential k-form on a complete non-compact manifold, the equivalent relation has been verified with the extended scope of q for from finite q-energy in Lq spaces to infinite q-energy in non-Lq spaces when with 2-balanced growth. Generalized Hadamard Theorem, Cauchy-Schwarz Inequality, and Calculus skills including Integration by Parts as well as Convergent Series have been applied as estimation techniques to evaluate growth rates for a differential form. In particular, energy growth rates as indicated by an appropriate power range in a selected test function lead to a balance between a harmonic differential form and a closed co-closed differential form. Research ideas and computational methods in this paper could provide an innovative way in the study of broadening Lq spaces to non-Lq spaces with a wide variety of infinite energy growth for a differential form.

Keywords: closed forms, co-closed forms, harmonic forms, L^q spaces, p-balanced growth, simple differential k-forms

Procedia PDF Downloads 453

Authors: A. D. Zikiryahodjaev, L. V. Bolotina, A. S. Sukhotko

Abstract:

Purpose. To evaluate the expediency and timeliness of performance of surgical treatment as a component of multi-therapy treatment of patients with stage IV breast cancers. Materials and Methods. This investigation comparatively analyzed the results of complex treatment with or without surgery in patients with metastatic breast cancer. We analyzed retrospectively treatment experience of 196 patients with generalized breast cancer in the department of oncology and breast reconstructive surgery of P.A. Herzen Moscow Cancer Research Institute from 2000 to 2012. The average age was (58±1,1) years. Invasive ductul carcinoma was verified in128 patients (65,3%), invasive lobular carcinoma-33 (16,8%), complex form - 19 (9,7%). Complex palliative care involving drug and radiation therapies was performed in two patient groups. The first group includes 124 patients who underwent surgical intervention as complex treatment, the second group includes 72 patients with only medical therapy. Standard systemic therapy was given to all patients. Results. Overall, 3-and 5-year survival in fist group was 43,8 and 21%, in second - 15,1 and 9,3% respectively [p=0,00002 log-rank]. Median survival in patients with surgical treatment composed 32 months, in patients with only systemic therapy-21. The factors having influencing an influence on the prognosis and the quality of life outcomes for of patients with generalized breast cancer were are also studied: hormone-dependent tumor, Her2/neu hyper-expression, reproductive function status (age, menopause existence). Conclusion.Removing primary breast tumor in patients with generalized breast cancer improve long-term outcomes. Three- and five-year survival increased by 28,7 and 16,3% respectively, and median survival–for 11 months. These patients may benefit from resection of the breast tumor. One explanation for the effect of this resection is that reducing the tumor load influences metastatic growth.

Keywords: breast cancer, combination therapy, factors of prognosis, primary tumor

Procedia PDF Downloads 424

Authors: Godwin Chigozie Okpara

Abstract:

This paper using standard GARCH, EGARCH, and TARCH models on day of the week return series (of 246 days) from the Nigerian Stock market estimated the model variants’ VaR. An asymmetric return distribution and fat-tail phenomenon in financial time series were considered by estimating the models with normal, student t and generalized error distributions. The analysis based on Akaike Information Criterion suggests that the EGARCH model with student t innovation distribution can furnish more accurate estimate of VaR. In the light of this, we apply the likelihood ratio tests of proportional failure rates to VaR derived from EGARCH model in order to determine the short and long positions VaR performances. The result shows that as alpha ranges from 0.05 to 0.005 for short positions, the failure rate significantly exceeds the prescribed quintiles while it however shows no significant difference between the failure rate and the prescribed quantiles for long positions. This suggests that investors and portfolio managers in the Nigeria stock market have long trading position or can buy assets with concern on when the asset prices will fall. Precisely, the VaR estimates for the long position range from -4.7% for 95 percent confidence level to -10.3% for 99.5 percent confidence level.

Keywords: downside risk, value-at-risk, failure rate, kupiec LR tests, GARCH models

Procedia PDF Downloads 447

Authors: Pranav Ajmera, Amit Kharat, Tanveer Gupte, Richa Pant, Viraj Kulkarni, Vinay Duddalwar, Purnachandra Lamghare

Abstract:

The cardiothoracic ratio (CTR) is the ratio of the diameter of the heart to the diameter of the thorax. An abnormal CTR, that is, a value greater than 0.55, is often an indicator of an underlying pathological condition. The accurate prediction of an abnormal CTR from chest X-rays (CXRs) aids in the early diagnosis of clinical conditions. We propose a deep learning-based model for automatic CTR calculation that can assist the radiologist with the diagnosis of cardiomegaly and optimize the radiology flow. The study population included 1012 posteroanterior (PA) CXRs from a single institution. The Attention U-Net deep learning (DL) architecture was used for the automatic calculation of CTR. A CTR of 0.55 was used as a cut-off to categorize the condition as cardiomegaly present or absent. An observer performance test was conducted to assess the radiologist's performance in diagnosing cardiomegaly with and without artificial intelligence (AI) assistance. The Attention U-Net model was highly specific in calculating the CTR. The model exhibited a sensitivity of 0.80 [95% CI: 0.75, 0.85], precision of 0.99 [95% CI: 0.98, 1], and a F1 score of 0.88 [95% CI: 0.85, 0.91]. During the analysis, we observed that 51 out of 1012 samples were misclassified by the model when compared to annotations made by the expert radiologist. We further observed that the sensitivity of the reviewing radiologist in identifying cardiomegaly increased from 40.50% to 88.4% when aided by the AI-generated CTR. Our segmentation-based AI model demonstrated high specificity and sensitivity for CTR calculation. The performance of the radiologist on the observer performance test improved significantly with AI assistance. A DL-based segmentation model for rapid quantification of CTR can therefore have significant potential to be used in clinical workflows.

Keywords: cardiomegaly, deep learning, chest radiograph, artificial intelligence, cardiothoracic ratio

Procedia PDF Downloads 103

Authors: Adeola Adenikinju, Olusanya E. Olubusoye, Lateef O. Akinpelu, Dilinna L. Nwobi

Abstract:

Since the enactment of the Fiscal Responsibility Act (2007), the Federal Government of Nigeria (FGN) has made public its fiscal budget and the subsequent implementation report. A critical review of these documents shows significant variations in the five macroeconomic variables which are inputs in each Presidential budget; oil Production target (mbpd), oil price ($), Foreign exchange rate(N/$), and Gross Domestic Product growth rate (%) and inflation rate (%). This results in underperformance of the Federal budget expected output in terms of non-oil and oil revenue aggregates. This paper evaluates first the existing variance between budgeted and actuals, then the relationship and causality between the determinants of Federal fiscal budget assumptions, and finally the determinants of FGN’s Gross Oil Revenue. The paper employed the use of descriptive statistics, the Autoregressive distributed lag (ARDL) model, and a Profit oil probabilistic model to achieve these objectives. This model permits for both the static and dynamic effect(s) of the independent variable(s) on the dependent variable, unlike a static model that accounts for static or fixed effect(s) only. It offers a technique for checking the existence of a long-run relationship between variables, unlike other tests of cointegration, such as the Engle-Granger and Johansen tests, which consider only non-stationary series that are integrated of the same order. Finally, even with small sample size, the ARDL model is known to generate a valid result, for it is the dependent variable and is the explanatory variable. The results showed that there is a long-run relationship between oil revenue as a proxy for budget performance and its determinants; oil price, produced oil quantity, and foreign exchange rate. There is a short-run relationship between oil revenue and its determinants; oil price, produced oil quantity, and foreign exchange rate. There is a long-run relationship between non-oil revenue and its determinants; inflation rate, GDP growth rate, and foreign exchange rate. The grangers’ causality test results show that there is a mono-directional causality between oil revenue and its determinants. The Federal budget assumptions only explain 68% of oil revenue and 62% of non-oil revenue. There is a mono-directional causality between non-oil revenue and its determinants. The Profit oil Model describes production sharing contracts, joint ventures, and modified carrying arrangements as the greatest contributors to FGN’s gross oil revenue. This provides empirical justification for the selected macroeconomic variables used in the Federal budget design and performance evaluation. The research recommends other variables, debt and money supply, be included in the Federal budget design to explain the Federal budget revenue performance further.

Keywords: ARDL, budget performance, oil price, oil quantity, oil revenue

Procedia PDF Downloads 179

Authors: V. Mantey, N. Findlay, I. Maddox

Abstract:

The quality and quantity of global satellite data have been increasing exponentially in recent years as spaceborne systems become more affordable and the sensors themselves become more sophisticated. This is a valuable resource for many applications, including disaster management and relief. However, while more information can be valuable, the volume of data available is impossible to manually examine. Therefore, the question becomes how to extract as much information as possible from the data with limited manpower. Buildings are a key feature of interest in satellite imagery with applications including telecommunications, population models, and disaster relief. Machine learning tools are fast becoming one of the key resources to solve this problem, and models have been developed to detect buildings in optical satellite imagery. However, by and large, most models focus on affluent regions where buildings are generally larger and constructed further apart. This work is focused on the more difficult problem of detection in populated regions. The primary challenge with detecting small buildings in densely populated regions is both the spatial and spectral resolution of the optical sensor. Densely packed buildings with similar construction materials will be difficult to separate due to a similarity in color and because the physical separation between structures is either non-existent or smaller than the spatial resolution. This study finds that training models until they are overfitting the input sample can perform better in these areas than a more robust, generalized model. An overfitted model takes less time to fine-tune from a generalized pre-trained model and requires fewer input data. The model developed for this study has also been fine-tuned using existing, open-source, building vector datasets. This is particularly valuable in the context of disaster relief, where information is required in a very short time span. Leveraging existing datasets means that little to no manpower or time is required to collect data in the region of interest. The training period itself is also shorter for smaller datasets. Requiring less data means that only a few quality areas are necessary, and so any weaknesses or underpopulated regions in the data can be skipped over in favor of areas with higher quality vectors. In this study, a landcover classification model was developed in conjunction with the building detection tool to provide a secondary source to quality check the detected buildings. This has greatly reduced the false positive rate. The proposed methodologies have been implemented and integrated into a configurable production environment and have been employed for a number of large-scale commercial projects, including continent-wide DEM production, where the extracted building footprints are being used to enhance digital elevation models. Overfitted machine learning models are often considered too specific to have any predictive capacity. However, this study demonstrates that, in cases where input data is scarce, overfitted models can be judiciously applied to solve time-sensitive problems.

Keywords: building detection, disaster relief, mask-RCNN, satellite mapping

Procedia PDF Downloads 174

Authors: Charline David, Alexandre Blondin Massé, Arnaud Zinflou

Abstract:

In 2016, Clements, Hurn, and Li proposed a multiple equation time series approach for the short-term load forecasting, reporting an average mean absolute percentage error (MAPE) of 1.36% on an 11-years dataset for the Queensland region in Australia. We present an adaptation of their model to the electrical power load consumption for the whole Quebec province in Canada. More precisely, we take into account two additional meteorological variables — cloudiness and wind speed — on top of temperature, as well as the use of multiple meteorological measurements taken at different locations on the territory. We also consider other minor improvements. Our final model shows an average MAPE score of 1:79% over an 8-years dataset.

Keywords: short-term load forecasting, special days, time series, multiple equations, parallelization, clustering

Procedia PDF Downloads 107