Search results for: empathic accuracy

Authors: Abdulaziz Afandi

Abstract:

Minimizing the inventory cost, optimizing the inventory quantities, and increasing system operational availability are the main motivations to enhance forecasting demand of spare parts in a major power utility company in Medina. This paper reports in an effort made to optimize the orders quantities of spare parts by improving the method of forecasting the demand. The study focuses on equipment that has frequent spare parts purchase orders with uncertain demand. The pattern of the demand considers a lumpy pattern which makes conventional forecasting methods less effective. A comparison was made by benchmarking various methods of forecasting based on experts’ criteria to select the most suitable method for the case study. Three actual data sets were used to make the forecast in this case study. Two neural networks (NN) approaches were utilized and compared, namely long short-term memory (LSTM) and multilayer perceptron (MLP). The results as expected, showed that the NN models gave better results than traditional forecasting method (judgmental method). In addition, the LSTM model had a higher predictive accuracy than the MLP model.

Keywords: neural network, LSTM, MLP, forecasting demand, inventory management

Procedia PDF Downloads 127

Authors: Jiahe Liu, HongyangYu, Feng Qian, Miao Luo

Abstract:

The aim of this study was to improve the effects of human body 3D reconstruction. The MvP algorithm was adopted to obtain key point information from multiple perspectives. This algorithm allowed the capture of human posture and joint positions from multiple angles, providing more comprehensive and accurate data. The study also incorporated the SMPL-X model, which has been widely used for human body modeling, to achieve more accurate 3D reconstruction results. The use of the MvP algorithm made it possible to observe the reconstructed object from multiple angles, thus reducing the problems of blind spots and missing information. This algorithm was able to effectively capture key point information, including the position and rotation angle of limbs, providing key data for subsequent 3D reconstruction. Compared with traditional single-view methods, the method of multi-view fusion significantly improved the accuracy and stability of reconstruction. By combining the MvP algorithm with the SMPL-X model, we successfully achieved better human body 3D reconstruction effects. The SMPL-X model is highly scalable and can generate highly realistic 3D human body models, thus providing more detail and shape information.

Keywords: 3D human reconstruction, multi-view, joint point, SMPL-X

Procedia PDF Downloads 70

Authors: Valentina Milanova, Kalina Mihaylova, Iva Lazarova

Abstract:

The mechanisms by which female genital infections are detected are varied and include clinician-collected high vaginal swabs, clinician-collected endocervical swabs, patient-collected vaginal swabs, and first-pass urine samples. Vaginal health screening has chronically low rates of uptake. This highlights the unmet need for a screening tool with comparable diagnostic accuracy which is familiar, convenient and easy to use for people. The Daye™ medical grade tampon offers an alternative to traditional sampling methods with the potential of increasing screening uptake among people previously too embarrassed or busy to attend gynecological appointments. In this white paper, the results of stability studies and a comparative clinical trial are discussed to assess the suitability of the device for the collection of vaginal samples for various clinical assessments. The tampon has demonstrated good sample stability and comparable sample quality compared to a self-collected vaginal swab and a clinician-collected cervical swab.

Keywords: vaginal microbiome, vaginal infections, gynaecological infections, female health, menstrual tampons, in vitro diagnostics

Procedia PDF Downloads 103

Authors: Denys Osipenko, Jonathan Crook

Abstract:

Because of the variety of the card holders’ behaviour types and income sources each consumer account can be transferred to a variety of states. Each consumer account can be inactive, transactor, revolver, delinquent, defaulted and requires an individual model for the income prediction. The estimation of transition probabilities between statuses at the account level helps to avoid the memorylessness of the Markov Chains approach. This paper investigates the transition probabilities estimation approaches to credit cards income prediction at the account level. The key question of empirical research is which approach gives more accurate results: multinomial logistic regression or multistage conditional logistic regression with binary target. Both models have shown moderate predictive power. Prediction accuracy for conditional logistic regression depends on the order of stages for the conditional binary logistic regression. On the other hand, multinomial logistic regression is easier for usage and gives integrate estimations for all states without priorities. Thus further investigations can be concentrated on alternative modeling approaches such as discrete choice models.

Keywords: multinomial regression, conditional logistic regression, credit account state, transition probability

Procedia PDF Downloads 487

Authors: Reza Behboodian

Abstract:

Monitoring the structural health and diagnosing their damage in the early stages has always been one of the topics of concern. Nowadays, research on structural damage detection methods based on vibration analysis is very extensive. Moreover, these methods can be used as methods of permanent and timely inspection of structures and prevent further damage to structures. Non-destructive methods are the low-cost and economical methods for determining the damage of structures. In this research, a non-destructive method for detecting and identifying the failure location in structures based on dynamic responses resulting from time history analysis is proposed. When the structure is damaged due to the reduction of stiffness, and due to the applied loads, the displacements in different parts of the structure were increased. In the proposed method, the damage position is determined based on the calculation of the strain energy difference in each member of the damaged structure and the healthy structure at any time. Defective members of the structure are indicated by the amount of strain energy relative to the healthy state. The results indicated that the proper accuracy and performance of the proposed method for identifying failure in structures.

Keywords: failure, time history analysis, dynamic response, strain energy

Procedia PDF Downloads 133

Authors: Peng Li, Luan Wang, Haifeng Fei, Renhong Xie, Yibin Rui, Shanhong Guo

Abstract:

The applications of satellite-based automatic identification system (S-AIS) pave the road for wide-range maritime traffic monitoring and management. But the coverage of satellite’s view includes multiple AIS self-organizing networks, which leads to the collision of AIS signals from different cells. The contribution of this work is to propose an improved multi-channel blind source separation algorithm based on Artificial Bee Colony (ABC) and advanced stochastic optimization to perform separation of the mixed AIS signals. The proposed approach adopts modified ABC algorithm to get an optimized initial separating matrix, which can expedite the initialization bias correction, and utilizes the Adaptive Moment Estimation (Adam) to update the separating matrix by adjusting the learning rate for each parameter dynamically. Simulation results show that the algorithm can speed up convergence and lead to better performance in separation accuracy.

Keywords: satellite-based automatic identification system, blind source separation, artificial bee colony, adaptive moment estimation

Procedia PDF Downloads 186

Authors: E. Pretorius, A. M. Viljoen, M. van der Bank

Abstract:

Introduction: The phytochemistry of Agathosma crenulata and A. betulina has been studied extensively, while their molecular analysis through DNA barcoding remains virtually unexplored. This technique can confirm the identity of plant species included in a herbal product, thereby ensuring the efficacy of the herbal product and the accuracy of its label. Materials and methods: Authentic Agathosma reference material of A. betulina (n=16) and A. crenulata (n=10) were obtained. Thirteen commercial products were purchased from various health shops around Johannesburg, South Africa, using the search term “Agathosma” or “Buchu.” The plastid regions matK and ycf1 were used to barcode the Buchu products, and BRONX analysis confirmed the taxonomic identity of the samples. UPLC-MS analyses were also performed. Results: Only (30/60) 60% of the traded samples tested from 13 suppliers contained A. betulina in their herbal products. Similar results were also obtained for the UPLC-MS analysis. Conclusion: In this study, we demonstrate the application of DNA barcoding in combination with phytochemical analysis to authenticate herbal products claiming to contain Agathosma plants as an ingredient in their products. This supports manufacturing efforts to ensure that herbal products that are safe for the consumer.

Keywords: Buchu, substitution, barcoding, BRONX algorithm, matK, ycf1, UPLC-MS

Procedia PDF Downloads 129

Authors: Fatemeh Sadat Sharifi

Abstract:

In the present study, the capabilities of WAVEWATCH-III and MIKE21-SW for predicting the characteristics of wind waves in Hormuz Strait are evaluated. The GFS wind data (Global Forecast System) were derived. The bathymetry of gride with 2 arc-minute resolution, also were extracted from the ETOPO1. WAVEWATCH-III findings illustrate more valid prediction of wave features comparing to the MIKE-21 SW in deep water. Apparently, in shallow area, the MIKE-21 provides more uniformities with altimetry measurements. This may be due to the merits of the unstructured grid which are used in MIKE-21, leading to better representations of the coastal area. The findings on the direction of waves generated by wind in the modeling area indicate that in some regions, despite the increase in wind speed, significant wave height stays nearly unchanged. This is fundamental because of swift changes in wind track over the Strait of Hormuz. After discussing wind-induced waves in the region, the impact of instability of the surface layer on wave growth has been considered. For this purpose, the average monthly mean air temperature has been used. The results in cold months, when the surface layer is unstable, indicates an acceptable increase in the accuracy of prediction of the indicator wave height.

Keywords: numerical modeling, WAVEWATCH-III, Strait of Hormuz, MIKE21-SW

Procedia PDF Downloads 207

Authors: Tawfik Thelaidjia, Salah Chenikher

Abstract:

Bearing fault diagnosis has attracted significant attention over the past few decades. It consists of two major parts: vibration signal feature extraction and condition classification for the extracted features. In this paper, feature extraction from faulty bearing vibration signals is performed by a combination of the signal’s Kurtosis and features obtained through the preprocessing of the vibration signal samples using Db2 discrete wavelet transform at the fifth level of decomposition. In this way, a 7-dimensional vector of the vibration signal feature is obtained. After feature extraction from vibration signal, the support vector machine (SVM) was applied to automate the fault diagnosis procedure. To improve the classification accuracy for bearing fault prediction, particle swarm optimization (PSO) is employed to simultaneously optimize the SVM kernel function parameter and the penalty parameter. The results have shown feasibility and effectiveness of the proposed approach

Keywords: condition monitoring, discrete wavelet transform, fault diagnosis, kurtosis, machine learning, particle swarm optimization, roller bearing, rotating machines, support vector machine, vibration measurement

Procedia PDF Downloads 437

Authors: Wei-Jong Yang, Yu-Siang Su, Pau-Choo Chung, Jar-Ferr Yang

Abstract:

Moving object detection (MOD) is an important issue in advanced driver assistance systems (ADAS). There are two important moving objects, pedestrians and scooters in ADAS. In real-world systems, there exist two important challenges for MOD, including the computational complexity and the detection accuracy. The histogram of oriented gradient (HOG) features can easily detect the edge of object without invariance to changes in illumination and shadowing. However, to reduce the execution time for real-time systems, the image size should be down sampled which would lead the outlier influence to increase. For this reason, we propose the histogram of uniformly-oriented gradient (HUG) features to get better accurate description of the contour of human body. In the testing phase, the support vector machine (SVM) with linear kernel function is involved. Experimental results show the correctness and effectiveness of the proposed method. With SVM classifiers, the real testing results show the proposed HUG features achieve better than classification performance than the HOG ones.

Keywords: moving object detection, histogram of oriented gradient, histogram of uniformly-oriented gradient, linear support vector machine

Procedia PDF Downloads 594

Authors: Mohammad Hadi Khorashadi Zadeh, Amin Karkon, Hamid Golnari

Abstract:

This study aimed to investigate the impact of information technology on the quality of accounting information was made in 2014. A survey of 425 executives of listed companies in Tehran Stock Exchange, using the Cochran formula simple random sampling method, 84 managers of these companies as the sample size was considered. Methods of data collection based on questionnaire information technology some of the questions of the impact of information technology was standardized questionnaires and the questions were designed according to existing components. After the distribution and collection of questionnaires, data analysis and hypothesis testing using structural equation modeling Smart PLS2 and software measurement model and the structure was conducted in two parts. In the first part of the questionnaire technical characteristics including reliability, validity, convergent and divergent validity for PLS has been checked and in the second part, application no significant coefficients were used to examine the research hypotheses. The results showed that IT and its dimensions (timeliness, relevance, accuracy, adequacy, and the actual transfer rate) affect the quality of accounting information of listed companies in Tehran Stock Exchange influence.

Keywords: information technology, information quality, accounting, transfer speed

Procedia PDF Downloads 277

Authors: K. Shiba, T. Kobayashi, T. Kaburagi, Y. Kurihara

Abstract:

Elderly population in the world is increasing, and consequently, their nursing burden is also increasing. In such situations, monitoring and evaluating their daily action facilitates efficient nursing care. Especially, we focus on an unconscious activity during sleep, i.e. turnover. Monitoring turnover during sleep is essential to evaluate various conditions related to sleep. Bedsores are considered as one of the monitoring conditions. Changing patient’s posture every two hours is required for caregivers to prevent bedsore. Herein, we attempt to develop an unconstrained nocturnal monitoring system using a sensing device based on piezoelectric ceramics that can detect the vibrations owing to human body movement on the bed. In the proposed method, in order to construct a multi-points sensing, we placed two sensing devices under the right and left legs at the head-side of an ordinary bed. Using this equipment, when a subject lies on the bed, feature is calculated from the output voltages of the sensing devices. In order to evaluate our proposed method, we conducted an experiment with six healthy male subjects. Consequently, the period during which turnover occurs can be correctly classified as the turnover period with 100% accuracy.

Keywords: turnover, piezoelectric ceramics, multi-points sensing, unconstrained monitoring system

Procedia PDF Downloads 194

Authors: Hao Mi, Ming Yang, Tian-yue Yang

Abstract:

Ultrasonic infrared nondestructive testing is a kind of testing method with high speed, accuracy and localization. However, there are still some problems, such as the detection requires manual real-time field judgment, the methods of result storage and viewing are still primitive. An intelligent non-destructive detection system based on embedded linux is put forward in this paper. The hardware part of the detection system is based on the ARM (Advanced Reduced Instruction Set Computer Machine) core and an embedded linux system is built to realize image processing and defect detection of thermal images. The CLAHE algorithm and the Butterworth filter are used to process the thermal image, and then the boa server and CGI (Common Gateway Interface) technology are used to transmit the test results to the display terminal through the network for real-time monitoring and remote monitoring. The system also liberates labor and eliminates the obstacle of manual judgment. According to the experiment result, the system provides a convenient and quick solution for industrial non-destructive testing.

Keywords: remote monitoring, non-destructive testing, embedded Linux system, image processing

Procedia PDF Downloads 224

Authors: Yin Yuanling

Abstract:

A Deep Feedback Recurrent Neural Network (DFRNN) and Bidirectional Long Short-Term Memory (BiLSTM) are designed to address the problem of low accuracy of traditional relationship extraction models. This method combines a deep feedback-based recurrent neural network (DFRNN) with a bi-directional long short-term memory (BiLSTM) approach. The method combines DFRNN, which extracts local features of text based on deep feedback recurrent mechanism, BiLSTM, which better extracts global features of text, and Self-Attention, which extracts semantic information. Experiments show that the method achieves an F1 value of 76.69% on the CEC dataset, which is 0.0652 better than the BiLSTM+Self-ATT model, thus optimizing the performance of the deep learning method in the event relationship extraction task.

Keywords: event relations, deep learning, DFRNN models, bi-directional long and short-term memory networks

Procedia PDF Downloads 144

Authors: Chien-Kuo Chiu

Abstract:

In the displacement-based seismic design and evaluation, equivalent linearization method is one of the approximation methods to estimate the maximum inelastic displacement response of a system. In this study, the accuracy of two equivalent linearization methods are investigated. The investigation consists of three soil condition in Taiwan (Taipei Basin 1, 2, and 3) and five different heights of building (H_r= 10, 20, 30, 40, and 50 m). The first method is the Taiwan equivalent linearization method (TELM) which was proposed based on Japanese equivalent linear method considering the modification factor, α_T= 0.85. On the basis of Lin and Miranda study, the second method is proposed with some modification considering Taiwan soil conditions. From this study, it is shown that Taiwanese equivalent linearization method gives better estimation compared to the modified Lin and Miranda method (MLM). The error index for the Taiwanese equivalent linearization method are 16%, 13%, and 12% for Taipei Basin 1, 2, and 3, respectively. Furthermore, a ductility demand spectrum of single-degree-of-freedom (SDOF) system is presented in this study as a guide for engineers to estimate the ductility demand of a structure.

Keywords: displacement-based design, ductility demand spectrum, equivalent linearization method, RC buildings, single-degree-of-freedom

Procedia PDF Downloads 162

Authors: S. M. Mahbobur Rahman

Abstract:

Whole building energy simulation models are widely used for predicting future energy consumption, performance diagnosis and optimum control.  Black box building energy modeling approach has been heavily studied in the past decade. The thermal response of a building can also be modeled using a network of interconnected resistors (R) and capacitors (C) at each node called R-C network. In this study, a model building, Case 600, as described in the “Standard Method of Test for the Evaluation of Building Energy Analysis Computer Program”, ASHRAE standard 140, is studied along with a 3R2C thermal network model and the ASHRAE clear sky solar radiation model. Although building an energy model involves two important parts of building component i.e., the envelope and internal mass, the effect of building internal mass is not considered in this study. All the characteristic parameters of the building envelope are evaluated as on Case 600. Finally, monthly building energy consumption from the thermal network model is compared with a simple-box energy model within reasonable accuracy. From the results, 0.6-9.4% variation of monthly energy consumption is observed because of the south-facing windows.

Keywords: ASHRAE case study, clear sky solar radiation model, energy modeling, thermal network model

Procedia PDF Downloads 146

Authors: Chandu Rathnayake, Isuri Anuradha

Abstract:

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

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

Procedia PDF Downloads 73

Authors: Jianhua Zhou, Yuwen Zhang

Abstract:

A two-step multigrid approach is proposed to solve the inverse heat conduction problem in a 3-D object under laser irradiation. In the first step, the location of the laser center is estimated using a coarse and uniform grid system. In the second step, the front-surface temperature is recovered in good accuracy using a multiple grid system in which fine mesh is used at laser spot center to capture the drastic temperature rise in this region but coarse mesh is employed in the peripheral region to reduce the total number of sensors required. The effectiveness of the two-step approach and the multiple grid system are demonstrated by the illustrative inverse solutions. If the measurement data for the temperature and heat flux on the back surface do not contain random error, the proposed multigrid approach can yield more accurate inverse solutions. When the back-surface measurement data contain random noise, accurate inverse solutions cannot be obtained if both temperature and heat flux are measured on the back surface.

Keywords: conduction, inverse problems, conjugated gradient method, laser

Procedia PDF Downloads 369

Authors: Sidra Naeem, Ayesha Naeem, Sahar Rahim, Nadia Nawaz Qadri

Abstract:

Citrus greening is a bacterial disease that causes considerable damage to citrus fruits worldwide. Efficient method for this disease detection must be carried out to minimize the production loss. This paper presents a pattern recognition system that comprises three stages for the detection of citrus greening from Orange leaves: segmentation, feature extraction and classification. Image segmentation is accomplished by adaptive thresholding. The feature extraction stage comprises of three visual descriptors i.e. shape, color and texture. From shape feature we have used asymmetry index, from color feature we have used histogram of Cb component from YCbCr domain and from texture feature we have used local binary pattern. Classification was done using support vector machines and k nearest neighbors. The best performances of the system is Accuracy = 88.02% and AUROC = 90.1% was achieved by automatic segmented images. Our experiments validate that: (1). Segmentation is an imperative preprocessing step for computer assisted diagnosis of citrus greening, and (2). The combination of shape, color and texture features form a complementary set towards the identification of citrus greening disease.

Keywords: citrus greening, pattern recognition, feature extraction, classification

Procedia PDF Downloads 184

Authors: Nishank Raisinghani

Abstract:

Efficiently predicting drug response remains a challenge in the realm of drug discovery. To address this issue, we propose four model architectures that combine graphical representation with varying positions of multiheaded self-attention mechanisms. By leveraging two types of multi-omics data, transcriptomics and genomics, we create a comprehensive representation of target cells and enable drug response prediction in precision medicine. A majority of our architectures utilize multiple transformer models, one with a graph attention mechanism and the other with a multiheaded self-attention mechanism, to generate latent representations of both drug and omics data, respectively. Our model architectures apply an attention mechanism to both drug and multiomics data, with the goal of procuring more comprehensive latent representations. The latent representations are then concatenated and input into a fully connected network to predict the IC-50 score, a measure of cell drug response. We experiment with all four of these architectures and extract results from all of them. Our study greatly contributes to the future of drug discovery and precision medicine by looking to optimize the time and accuracy of drug response prediction.

Keywords: drug discovery, transformers, graph neural networks, multiomics

Procedia PDF Downloads 153

Authors: Juan Sebastián Hernández

Abstract:

The Nevado del Ruíz volcano, located between the limits of the Departments of Caldas and Tolima in Colombia, presented an unstable behaviour in the course of the year 2020, this volcanic activity led to secondary effects on the crust, which is why the prediction of deformations becomes the task of geoscientists. In the course of this article, the use of tropospheric variables such as evapotranspiration, UV aerosol index, carbon monoxide, nitrogen dioxide, methane, surface temperature, among others, is used to train a set of neural networks that can predict the behaviour of the resulting phase of an unrolled interferogram with the DInSAR technique, whose main objective is to identify and characterise the behaviour of the crust based on the environmental conditions. For this purpose, variables were collected, a generalised linear model was created, and a set of neural networks was created. After the training of the network, validation was carried out with the test data, giving an MSE of 0.17598 and an associated r-squared of approximately 0.88454. The resulting model provided a dataset with good thematic accuracy, reflecting the behaviour of the volcano in 2020, given a set of environmental characteristics.

Keywords: crustal deformation, Tropomi, neural networks (ANN), volcanic activity, DInSAR

Procedia PDF Downloads 103

Authors: H. J. T. Kevin Mozes, Dyah Purnamasari

Abstract:

Background: Lung cancer accounted for the fourth most common cancer in Indonesia. 85% of all lung cancer cases are the Non-Small Cell Lung Cancer (NSCLC). The indistinct signs and symptoms of NSCLC sometimes lead to misdiagnosis. The gold standard assessment for the diagnosis of NSCLC is the histopathological biopsy, which is invasive. Cyfra 21-1 is a tumor marker, which can be found in the intermediate protein structure in the epitel. The accuracy of Cyfra 21-1 in diagnosing NSCLC is not yet known, so this report is made to seek the answer for the question above. Methods: Literature searching is done using online databases. Proquest and Pubmed are online databases being used in this report. Then, literature selection is done by excluding and including based on inclusion criterias and exclusion criterias. The selected literature is then being appraised using the criteria of validity, importance, and validity. Results: From six journals appraised, five of them are valid. Sensitivity value acquired from all five literature is ranging from 50-84.5 %, meanwhile the specificity is 87.8 %-94.4 %. Likelihood the ratio of all appraised literature is ranging from 5.09 -10.54, which categorized to Intermediate High. Conclusion: Serum Cyfra 21-1 is a sensitive and very specific tumor marker for diagnosis of non-small cell lung cancer (NSCLC).

Keywords: cyfra 21-1, diagnosis, nonsmall cell lung cancer, NSCLC, tumor marker

Procedia PDF Downloads 232

Authors: Ling Qi, Matloob Khushi, Josiah Poon

Abstract:

This paper introduces a pre-trained Masked Candlestick Model (MCM) for trading time-series data. The pre-trained model is based on three core designs. First, we convert trading price data at each data point as a set of normalized elements and produce embeddings of each element. Second, we generate a masked sequence of such embedded elements as inputs for self-supervised learning. Third, we use the encoder mechanism from the transformer to train the inputs. The masked model learns the contextual relations among the sequence of embedded elements, which can aid downstream classification tasks. To evaluate the performance of the pre-trained model, we fine-tune MCM for three different downstream classification tasks to predict future price trends. The fine-tuned models achieved better accuracy rates for all three tasks than the baseline models. To better analyze the effectiveness of MCM, we test the same architecture for three currency pairs, namely EUR/GBP, AUD/USD, and EUR/JPY. The experimentation results demonstrate MCM’s effectiveness on all three currency pairs and indicate the MCM’s capability for signal extraction from trading data.

Keywords: masked language model, transformer, time series prediction, trading prediction, embedding, transfer learning, self-supervised learning

Procedia PDF Downloads 128

Authors: Prem K. Mony, Bharadwaj Amrutur, Prashanth Thankachan, Swarnarekha Bhat, Suman Rao, Maryann Washington, Annamma Thomas, N. Sheela, Hiteshwar Rao, Sumi Antony

Abstract:

The disease burden among mothers and newborns is caused mostly by a handful of avoidable conditions occurring around the time of childbirth and within the first month following delivery. Real-time monitoring of vital parameters of mothers and neonates offers a potential opportunity to impact access as well as the quality of care in vulnerable populations. We describe the design, development and testing of an innovative wearable device for remote biomonitoring (RBM) of body temperatures in mothers and neonates in a hospital in southern India. The architecture consists of: [1] a low-cost, wearable sensor tag; [2] a gateway device for ‘real-time’ communication link; [3] piggy-backing on a commercial GSM communication network; and [4] an algorithm-based data analytics system. Requirements for the device were: long battery-life upto 28 days (with sampling frequency 5/hr); robustness; IP 68 hermetic sealing; and human-centric design. We undertook pre-clinical laboratory testing followed by clinical trial phases I & IIa for evaluation of safety and efficacy in the following sequence: seven healthy adult volunteers; 18 healthy mothers; and three sets of babies – 3 healthy babies; 10 stable babies in the Neonatal Intensive Care Unit (NICU) and 1 baby with hypoxic ischaemic encephalopathy (HIE). The 3-coin thickness, pebble-design sensor weighing about 8 gms was secured onto the abdomen for the baby and over the upper arm for adults. In the laboratory setting, the response-time of the sensor device to attain thermal equilibrium with the surroundings was 4 minutes vis-a-vis 3 minutes observed with a precision-grade digital thermometer used as a reference standard. The accuracy was ±0.1°C of the reference standard within the temperature range of 25-40°C. The adult volunteers, aged 20 to 45 years, contributed a total of 345 hours of readings over a 7-day period and the postnatal mothers provided a total of 403 paired readings. The mean skin temperatures measured in the adults by the sensor were about 2°C lower than the axillary temperature readings (sensor =34.1 vs digital = 36.1); this difference was statistically significant (t-test=13.8; p<0.001). The healthy neonates provided a total of 39 paired readings; the mean difference in temperature was 0.13°C (sensor =36.9 vs digital = 36.7; p=0.2). The neonates in the NICU provided a total of 130 paired readings. Their mean skin temperature measured by the sensor was 0.6°C lower than that measured by the radiant warmer probe (sensor =35.9 vs warmer probe = 36.5; p < 0.001). The neonate with HIE provided a total of 25 paired readings with the mean sensor reading being not different from the radian warmer probe reading (sensor =33.5 vs warmer probe = 33.5; p=0.8). No major adverse events were noted in both the adults and neonates; four adult volunteers reported mild sweating under the device/arm band and one volunteer developed mild skin allergy. This proof-of-concept study shows that real-time monitoring of temperatures is technically feasible and that this innovation appears to be promising in terms of both safety and accuracy (with appropriate calibration) for improved maternal and neonatal health.

Keywords: public health, remote biomonitoring, temperature surveillance, wearable sensors, mothers and newborns

Procedia PDF Downloads 208

Authors: Lina, Arlends Chris, Bagus Mulyawan, Agus B. Dharmawan

Abstract:

Blood cell analysis plays a significant role in the diagnosis of human health. As an alternative to the traditional technique conducted by laboratory technicians, this paper presents an automatic white blood cell (leukocyte) detection system using Image Stitching and Color Overlapping Windows. The advantage of this method is to present a detection technique of white blood cells that are robust to imperfect shapes of blood cells with various image qualities. The input for this application is images from a microscope-slide translation video. The preprocessing stage is performed by stitching the input images. First, the overlapping parts of the images are determined, then stitching and blending processes of two input images are performed. Next, the Color Overlapping Windows is performed for white blood cell detection which consists of color filtering, window candidate checking, window marking, finds window overlaps, and window cropping processes. Experimental results show that this method could achieve an average of 82.12% detection accuracy of the leukocyte images.

Keywords: color overlapping windows, image stitching, leukocyte detection, white blood cell detection

Procedia PDF Downloads 310

Authors: Muhammad Waqar Ashraf

Abstract:

Selected heavy metals, namely Cu, Zn, Fe, Mn, Cd, Pb, and As, in seven popular varieties of edible vegetable oils collected from Saudi Arabia, were determined by graphite furnace atomic absorption spectrometry (GF-AAS) using microwave digestion. The accuracy of procedure was confirmed by certified reference materials (NIST 1577b). The concentrations for copper, zinc, iron, manganese, lead and arsenic were observed in the range of 0.035 - 0.286, 0.955 - 3.10, 17.3 - 57.8, 0.178 - 0.586, 0.011 - 0.017 and 0.011 - 0.018 µg/g, respectively. Cadmium was found to be in the range of 2.36 - 6.34 ng/g. The results are compared internationally and with standards laid down by world health agencies. A risk assessment study has been carried out to assess exposure to these metals via consumption of vegetable oils. A comparison has been made with safety intake levels for these heavy metals recommended by Institute of Medicine of the National Academies (IOM), US Environmental Protection Agency (US EPA) and Joint FAO/WHO Expert Committee on Food Additives (JECFA). The results indicated that the dietary intakes of the selected heavy metals from daily consumption of 25 g of edible vegetable oils for a 70 kg individual should pose no significant health risk to local population.

Keywords: vegetable oils, heavy metals, contamination, health risk assessment

Procedia PDF Downloads 451

Authors: Omer Cahana, Maya Herman, Ofer Levi

Abstract:

Magnetic resonance imaging (MRI) is a long medical scan that stems from a long acquisition time. This length is mainly due to the traditional sampling theorem, which defines a lower boundary for sampling. However, it is still possible to accelerate the scan by using a different approach, such as compress sensing (CS) or parallel imaging (PI). These two complementary methods can be combined to achieve a faster scan with high-fidelity imaging. In order to achieve that, two properties have to exist: i) the signal must be sparse under a known transform domain, ii) the sampling method must be incoherent. In addition, a nonlinear reconstruction algorithm needs to be applied to recover the signal. While the rapid advance in the deep learning (DL) field, which has demonstrated tremendous successes in various computer vision task’s, the field of MRI reconstruction is still in an early stage. In this paper, we present an extension of the state-of-the-art model in MRI reconstruction -VarNet. We utilize VarNet by using dilated convolution in different scales, which extends the receptive field to capture more contextual information. Moreover, we simplified the sensitivity map estimation (SME), for it holds many unnecessary layers for this task. Those improvements have shown significant decreases in computation costs as well as higher accuracy.

Keywords: MRI, deep learning, variational network, computer vision, compress sensing

Procedia PDF Downloads 162

Authors: Utkarsh A. Mishra, Ankit Bansal

Abstract:

At high temperature, radiative heat transfer is the dominant mode of heat transfer. It is governed by various phenomena such as photon emission, absorption, and scattering. The solution of the governing integrodifferential equation of radiative transfer is a complex process, more when the effect of participating medium and wavelength properties are taken into consideration. Although a generic formulation of such radiative transport problem can be modeled for a wide variety of problems with non-gray, non-diffusive surfaces, there is always a trade-off between simplicity and accuracy of the problem. Recently, solutions of complicated mathematical problems with statistical methods based on randomization of naturally occurring phenomena have gained significant importance. Photon bundles with discrete energy can be replicated with random numbers describing the emission, absorption, and scattering processes. Photon Monte Carlo (PMC) is a simple, yet powerful technique, to solve radiative transfer problems in complicated geometries with arbitrary participating medium. The method, on the one hand, increases the accuracy of estimation, and on the other hand, increases the computational cost. The participating media -generally a gas, such as CO₂, CO, and H₂O- present complex emission and absorption spectra. To model the emission/absorption accurately with random numbers requires a weighted sampling as different sections of the spectrum carries different importance. Importance sampling (IS) was implemented to sample random photon of arbitrary wavelength, and the sampled data provided unbiased training of MC estimators for better results. A better replacement to uniform random numbers is using deterministic, quasi-random sequences. Halton, Sobol, and Faure Low-Discrepancy Sequences are used in this study. They possess better space-filling performance than the uniform random number generator and gives rise to a low variance, stable Quasi-Monte Carlo (QMC) estimators with faster convergence. An optimal supervised learning scheme was further considered to reduce the computation costs of the PMC simulation. A one-dimensional plane-parallel slab problem with participating media was formulated. The history of some randomly sampled photon bundles is recorded to train an Artificial Neural Network (ANN), back-propagation model. The flux was calculated using the standard quasi PMC and was considered to be the training target. Results obtained with the proposed model for the one-dimensional problem are compared with the exact analytical and PMC model with the Line by Line (LBL) spectral model. The approximate variance obtained was around 3.14%. Results were analyzed with respect to time and the total flux in both cases. A significant reduction in variance as well a faster rate of convergence was observed in the case of the QMC method over the standard PMC method. However, the results obtained with the ANN method resulted in greater variance (around 25-28%) as compared to the other cases. There is a great scope of machine learning models to help in further reduction of computation cost once trained successfully. Multiple ways of selecting the input data as well as various architectures will be tried such that the concerned environment can be fully addressed to the ANN model. Better results can be achieved in this unexplored domain.

Keywords: radiative heat transfer, Monte Carlo Method, pseudo-random numbers, low discrepancy sequences, artificial neural networks

Procedia PDF Downloads 223

Authors: Xiao Zhou, Jianlin Cheng

Abstract:

A single amino acid mutation can have a significant impact on the stability of protein structure. Thus, the prediction of protein stability change induced by single site mutations is critical and useful for studying protein function and structure. Here, we presented a deep learning network with the dropout technique for predicting protein stability changes upon single amino acid substitution. While using only protein sequence as input, the overall prediction accuracy of the method on a standard benchmark is >85%, which is higher than existing sequence-based methods and is comparable to the methods that use not only protein sequence but also tertiary structure, pH value and temperature. The results demonstrate that deep learning is a promising technique for protein stability prediction. The good performance of this sequence-based method makes it a valuable tool for predicting the impact of mutations on most proteins whose experimental structures are not available. Both the downloadable software package and the user-friendly web server (DNpro) that implement the method for predicting protein stability changes induced by amino acid mutations are freely available for the community to use.

Keywords: bioinformatics, deep learning, protein stability prediction, biological data mining

Procedia PDF Downloads 468

Authors: Mostafa Gamal, Hoda K. Mohamed, Islam El-Maddah, Ali Hamdi

Abstract:

In recent years, many digital documents on the web have been created due to the rapid growth of ’social applications’ communities or ’Data-intensive applications’. The evolution of online-based multimedia data poses new challenges in storing and querying large amounts of data for online recommender systems. Graph data models have been shown to be more efficient than relational data models for processing complex data. This paper will explain the key differences between graph and relational databases, their strengths and weaknesses, and why using graph databases is the best technology for building a realtime recommendation system. Also, The paper will discuss several similarity metrics algorithms that can be used to compute a similarity score of pairs of nodes based on their neighbourhoods or their properties. Finally, the paper will discover how NLP strategies offer the premise to improve the accuracy and coverage of realtime recommendations by extracting the information from the stored unstructured knowledge, which makes up the bulk of the world’s data to enrich the graph database with this information. As the size and number of data items are increasing rapidly, the proposed system should meet current and future needs.

Keywords: graph databases, NLP, recommendation systems, similarity metrics

Procedia PDF Downloads 104