Search results for: forecasting accuracy

Authors: Natthawut Arin

Abstract:

In forensic contexts, competency to stand trial assessments are the most common referrals. The defendants may attempt to endorse psychopathology symptoms and feign incompetent. Coaching, which can be teaching them test-taking strategies to avoid detection of psychopathological symptoms feigning. Recently, the Symptom Validity Testings (SVTs) were created to detect feigning. Moreover, the works of the literature showed that the effects of coaching on SVTs may be more robust to the effects of coaching. Thai Symptom Validity Test (SVT-Th) was designed as SVTs which demonstrated adequate psychometric properties and ability to classify between feigners and honest responders. Thus, the current study to examine the utility as the robustness of SVT-Th in the detection of feigned psychopathology. Participants consisted of 120 were recruited from undergraduate courses in psychology, randomly assigned to one of three groups. The SVT-Th was administered to those three scenario-experimental groups: (a) Uncoached group were asked to respond honestly (n=40), (b) Symptom-coached without warning group were asked to feign psychiatric symptoms to gain incompetency to stand trial (n=40), while (c) Test-coached with warning group were asked to feign psychiatric symptoms to avoid test detection but being incompetency to stand trial (n=40). Group differences were analyzed using one-way ANOVAs. The result revealed an uncoached group (M = 4.23, SD.= 5.20) had significantly lower SVT-Th mean scores than those both coached groups (M =185.00, SD.= 72.88 and M = 132.10, SD.= 54.06, respectively). Classification rates were calculated to determine the classification accuracy. Result indicated that SVT-Th had overall classification accuracy rates of 96.67% with acceptable of 95% sensitivity and 100% specificity rates. Overall, the results of the present study indicate that the SVT-Th yielded high adequate indices of accuracy and these findings suggest that the SVT-Th is robustness against coaching.

Keywords: incompetency to stand trial, coaching, robustness, classification accuracy

Procedia PDF Downloads 115

Authors: Jiun-Horng Tsai, Shi-Jie, Nieh

Abstract:

Fine particulate matter (PM₂.₅) has become an important issue all over the world over the last decade. Annual mean PM₂.₅ concentration has been over the ambient air quality standard of PM₂.₅ (annual average concentration as 15μg/m³) which adapted by Taiwan Environmental Protection Administration (TEPA). TEPA, therefore, has developed a number of air pollution control measures to improve the ambient concentration by reducing the emissions of primary fine particulate matter and the precursors of secondary PM₂.₅. This study investigated the potential improvement of ambient PM₂.₅ concentration by the TEPA program and the other scenario for further emission reduction on various sources. Four scenarios had been evaluated in this study, including a basic case and three reduction scenarios (A to C). The ambient PM₂.₅ concentration was evaluated by Community Multi-scale Air Quality modelling system (CMAQ) ver. 4.7.1 along with the Weather Research and Forecasting Model (WRF) ver. 3.4.1. The grid resolutions in the modelling work are 81 km × 81 km for domain 1 (covers East Asia), 27 km × 27 km for domain 2 (covers Southeast China and Taiwan), and 9 km × 9 km for domain 3 (covers Taiwan). The result of PM₂.₅ concentration simulation in different regions of Taiwan shows that the annual average concentration of basic case is 24.9 μg/m³, and are 22.6, 18.8, and 11.3 μg/m³, respectively, for scenarios A to C. The annual average concentration of PM₂.₅ would be reduced by 9-55 % for those control scenarios. The result of scenario C (the emissions of precursors reduce to allowance levels) could improve effectively the airborne PM₂.₅ concentration to attain the air quality standard. According to the results of unit precursor reduction contribution, the allowance emissions of PM₂.₅, SOₓ, and NOₓ are 16.8, 39, and 62 thousand tons per year, respectively. In the Kao-Ping air basin, the priority for reducing precursor emissions is PM₂.₅ > NOₓ > SOₓ, whereas the priority for reducing precursor emissions is PM₂.₅ > SOₓ > NOₓ in others area. The result indicates that the target pollutants that need to be reduced in different air basin are different, and the control measures need to be adapted to local conditions.

Keywords: airborne PM₂.₅, community multi-scale air quality modelling system, control measures, weather research and forecasting model

Procedia PDF Downloads 115

Authors: Joseph C. Chen, Joshua Cox

Abstract:

This research focuses on the use of the Taguchi method to reduce the surface roughness and improve dimensional accuracy of parts machined by Wire Electrical Discharge Machining (EDM) with S7 heat treated steel material. Due to its high impact toughness, the material is a candidate for a wide variety of tooling applications which require high precision in dimension and desired surface roughness. This paper demonstrates that Taguchi Parameter Design methodology is able to optimize both dimensioning and surface roughness successfully by investigating seven wire-EDM controllable parameters: pulse on time (ON), pulse off time (OFF), servo voltage (SV), voltage (V), servo feed (SF), wire tension (WT), and wire speed (WS). The temperature of the water in the Wire EDM process is investigated as the noise factor in this research. Experimental design and analysis based on L₁₈ Taguchi orthogonal arrays are conducted. This paper demonstrates that the Taguchi-based system enables the wire EDM process to produce (1) high precision parts with an average of 0.6601 inches dimension, while the desired dimension is 0.6600 inches; and (2) surface roughness of 1.7322 microns which is significantly improved from 2.8160 microns.

Keywords: Taguchi Parameter Design, surface roughness, Wire EDM, dimensional accuracy

Procedia PDF Downloads 345

Authors: Kelechi Emmanuel, Kyaw Thein Aung, William Burch

Abstract:

Introduction: Although rewarding in more ways than one, rural medicine can be challenging. The factors that lead to the challenges experienced in rural medicine include but are not limited to scarcity of resources, poor patient education inadequately trained professionals. This is the first single center study done on the challenges of and ways to improve diagnosis in rural medicine. Materials and Methods: Questionnaires were given to providers in a single hospital in rural Tennessee USA. In which providers were asked the question ‘In the past six months, what measures have you taken to improve your diagnostic accuracy given limited resources. Results: The questionnaire was passed to ten physicians working in a two hundred and twentyfive hospital bed. Physicians who participated included physicians in hospital medicine, emergency medicine, surgery, cardiology and gastroenterology. The study found that improved physical examination skills, access to specialist especially via telemedicine and affiliation to centers with more experienced professionals improved diagnosis and overall patient outcome in rural medicine. Conclusion: From this single center study, there is evidence to show that in addition to honing physical examination skills and having access to immediate results of testing done; hospital collaborations and access to highly trained specialist via telemedicine does improve diagnosis in rural medicine.

Keywords: rural medicine, diagnostic accuracy, diagnosis, telemedicine

Procedia PDF Downloads 51

Authors: Robert Long, Hiroaki Watanabe

Abstract:

This longitudinal study examines the grammatical errors of Japanese university students’ dialogues with a native speaker over an academic year. The L2 interactions of 15 Japanese speakers were taken from the JUSFC2018 corpus (April/May 2018) and the JUSFC2019 corpus (January/February). The corpora were based on a self-introduction monologue and a three-question dialogue; however, this study examines the grammatical accuracy found in the dialogues. Research questions focused on a possible significant difference in grammatical accuracy from the first interview session in 2018 and the second one the following year, specifically regarding errors in clauses per 100 words, global errors and local errors, and with specific errors related to parts of speech. The investigation also focused on which forms showed the least improvement or had worsened? Descriptive statistics showed that error-free clauses/errors per 100 words decreased slightly while clauses with errors/100 words increased by one clause. Global errors showed a significant decline, while local errors increased from 97 to 158 errors. For errors related to parts of speech, a t-test confirmed there was a significant difference between the two speech corpora with more error frequency occurring in the 2019 corpus. This data highlights the difficulty in having students self-edit themselves.

Keywords: clause analysis, global vs. local errors, grammatical accuracy, L2 output, longitudinal study

Procedia PDF Downloads 108

Authors: Keyi Wang

Abstract:

Similar to the touchscreen, hand gesture based human-computer interaction (HCI) is a technology that could allow people to perform a variety of tasks faster and more conveniently. This paper proposes a training method of an image-based hand gesture image and video clip recognition system using a CNN (Convolutional Neural Network) with a dataset. A dataset containing 6 hand gesture images is used to train a 2D CNN model. ~98% accuracy is achieved. Furthermore, a 3D CNN model is trained on a dataset containing 4 hand gesture video clips resulting in ~83% accuracy. It is demonstrated that a Cozmo robot loaded with pre-trained models is able to recognize static and dynamic hand gestures.

Keywords: deep learning, hand gesture recognition, computer vision, image processing

Procedia PDF Downloads 114

Authors: Hichem Sassi, Tawfik Najeh, Noureddine Liouane

Abstract:

The localization information is crucial for the operation of WSN. There are principally two types of localization algorithms. The Range-based localization algorithm has strict requirements on hardware; thus, it is expensive to be implemented in practice. The Range-free localization algorithm reduces the hardware cost. However, it can only achieve high accuracy in ideal scenarios. In this paper, we locate unknown nodes by incorporating the advantages of these two types of methods. The proposed algorithm makes the unknown nodes select the nearest anchor using the Received Signal Strength Indicator (RSSI) and choose two other anchors which are the most accurate to achieve the estimated location. Our algorithm improves the localization accuracy compared with previous algorithms, which has been demonstrated by the simulating results.

Keywords: WSN, localization, DV-Hop, RSSI

Procedia PDF Downloads 336

Authors: Jyh Sheen, Yong-Lin Wang

Abstract:

This research dedicates to find a different measurement procedure of microwave dielectric properties of ceramic materials with high dielectric constants. For the composite of ceramic dispersed in the polymer matrix, the dielectric constants of the composites with different concentrations can be obtained by various mixture equations. The other development of mixture rule is to calculate the permittivity of ceramic from measurements on composite. To do this, the analysis method and theoretical accuracy on six basic mixture laws derived from three basic particle shapes of ceramic fillers have been reported for dielectric constants of ceramic less than 40 at microwave frequency. Similar researches have been done for other well-known mixture rules. They have shown that both the physical curve matching with experimental results and low potential theory error are important to promote the calculation accuracy. Recently, a modified of mixture equation for high dielectric constant ceramics at microwave frequency has also been presented for strontium titanate (SrTiO3) which was selected from five more well known mixing rules and has shown a good accuracy for high dielectric constant measurements. However, it is still not clear the accuracy of this modified equation for other high dielectric constant materials. Therefore, the five more well known mixing rules are selected again to understand their application to other high dielectric constant ceramics. The other high dielectric constant ceramic, TiO2 with dielectric constant 100, was then chosen for this research. Their theoretical error equations are derived. In addition to the theoretical research, experimental measurements are always required. Titanium dioxide is an interesting ceramic for microwave applications. In this research, its powder is adopted as the filler material and polyethylene powder is like the matrix material. The dielectric constants of those ceramic-polyethylene composites with various compositions were measured at 10 GHz. The theoretical curves of the five published mixture equations are shown together with the measured results to understand the curve matching condition of each rule. Finally, based on the experimental observation and theoretical analysis, one of the five rules was selected and modified to a new powder mixture equation. This modified rule has show very good curve matching with the measurement data and low theoretical error. We can then calculate the dielectric constant of pure filler medium (titanium dioxide) by those mixing equations from the measured dielectric constants of composites. The accuracy on the estimating dielectric constant of pure ceramic by various mixture rules will be compared. This modified mixture rule has also shown good measurement accuracy on the dielectric constant of titanium dioxide ceramic. This study can be applied to the microwave dielectric properties measurements of other high dielectric constant ceramic materials in the future.

Keywords: microwave measurement, dielectric constant, mixture rules, composites

Procedia PDF Downloads 341

Authors: Hossein Shokouhmand, Majid Hasanpour

Abstract:

A new modification to the Strelow method of chevron-type plate heat exchangers (PHX) modeling is proposed. The effects of maldistribution are accounted in the resulting equation. The results of calculations are validated by reported experiences. The good accuracy of heat transfer performance prediction is shown. The results indicate that considering flow maldistribution improve the accuracy of predicting the flow and thermal behavior of the plate exchanger. Additionally, a wide range of the parametric study has been presented which brings out the effects of chevron angle of PHE on its thermal efficiency with considering maldistribution effect. In addition, the thermally optimal corrugation discussed for the chevron-type PHEs.

Keywords: chevron angle, plate heat exchangers, maldistribution, strelow method

Procedia PDF Downloads 167

Authors: Beghdadi Lotfi, Bouziane Abdelhafid

Abstract:

A finite volume method for quadrilaterals unstructured mesh is developed to predict the two dimensional steady-state solutions of conduction equation. In this scheme, based on the integration around the polygonal control volume, the derivatives of conduction equation must be converted into closed line integrals using same formulation of the Stokes theorem. To valid the accuracy of the method two numerical experiments s are used: conduction in a regular block (with known analytical solution) and conduction in a rotated block (case with curved boundaries).The numerical results show good agreement with analytical results. To demonstrate the accuracy of the method, the absolute and root-mean square errors versus the grid size are examined quantitatively.

Keywords: Stokes theorem, unstructured grid, heat transfer, complex geometry

Procedia PDF Downloads 261

Authors: Jeonghwan Jeon, Mintak Han, Youngjun Kim

Abstract:

Due to the rapid development of information and communication technology (ICT), a substantial transformation is currently happening in the society. As the range of intelligent technologies and services is continuously expanding, ‘things’ are becoming capable of communicating one another and even with people. However, such “Internet of Things” has the technical weakness so that a great amount of such information transferred in real-time may be widely exposed to the threat of security. User’s personal data are a typical example which is faced with a serious security threat. The threats of security will be diversified and arose more frequently because next generation of unfamiliar technology develops. Moreover, as the society is becoming increasingly complex, security vulnerability will be increased as well. In the existing literature, a considerable number of private and public reports that forecast future society have been published as a precedent step of the selection of future technology and the establishment of strategies for competitiveness. Although there are previous studies that forecast security technology, they have focused only on technical issues and overlooked the interrelationships between security technology and social factors are. Therefore, investigations of security threats in the future and security technology that is able to protect people from various threats are required. In response, this study aims to derive potential security threats associated with the development of technology and to explore the security technology that can protect against them. To do this, first of all, private and public reports that forecast future and online documents from technology-related communities are collected. By analyzing the data, future issues are extracted and categorized in terms of STEEP (Society, Technology, Economy, Environment, and Politics), as well as security. Second, the components of potential security threats are developed based on classified future issues. Then, points that the security threats may occur –for example, mobile payment system based on a finger scan technology– are identified. Lastly, alternatives that prevent potential security threats are proposed by matching security threats with points and investigating related security technologies from patent data. Proposed approach can identify the ICT-related latent security menaces and provide the guidelines in the ‘problem – alternative’ form by linking the threat point with security technologies.

Keywords: future society, information and communication technology, security technology, technology forecasting

Procedia PDF Downloads 443

Authors: Timothy Kayode Samson, Adedoyin Isola Lawal

Abstract:

The past five years have shown a sharp increase in public interest in the crypto market, with its market capitalization growing from $100 billion in June 2017 to $2158.42 billion on April 5, 2022. Despite the outrageous nature of the volatility of cryptocurrencies, the use of skewed error innovation distributions in modelling the volatility behaviour of these digital currencies has not been given much research attention. Hence, this study models the volatility of 5 largest cryptocurrencies by market capitalization (Bitcoin, Ethereum, Tether, Binance coin, and USD Coin) using four variants of GARCH models (GJR-GARCH, sGARCH, EGARCH, and APARCH) estimated using three skewed error innovation distributions (skewed normal, skewed student- t and skewed generalized error innovation distributions). Daily closing prices of these currencies were obtained from Yahoo Finance website. Finding reveals that the Binance coin reported higher mean returns compared to other digital currencies, while the skewness indicates that the Binance coin, Tether, and USD coin increased more than they decreased in values within the period of study. For both Bitcoin and Ethereum, negative skewness was obtained, meaning that within the period of study, the returns of these currencies decreased more than they increased in value. Returns from these cryptocurrencies were found to be stationary but not normality distributed with evidence of the ARCH effect. The skewness parameters in all best forecasting models were all significant (p<.05), justifying of use of skewed error innovation distributions with a fatter tail than normal, Student-t, and generalized error innovation distributions. For Binance coin, EGARCH-sstd outperformed other volatility models, while for Bitcoin, Ethereum, Tether, and USD coin, the best forecasting models were EGARCH-sstd, APARCH-sstd, EGARCH-sged, and GJR-GARCH-sstd, respectively. This suggests the superiority of skewed Student t- distribution and skewed generalized error distribution over the skewed normal distribution.

Keywords: skewed generalized error distribution, skewed normal distribution, skewed student t- distribution, APARCH, EGARCH, sGARCH, GJR-GARCH

Procedia PDF Downloads 71

Authors: Eyob Seife, Tesfalem Teshome, Bereket Seyoum, Behailu Getachew, Yohans Demis

Abstract:

Developing countries continue to face preventable communicable diseases, such as vaccine-preventable diseases. The Expanded Programme on Immunization (EPI) was established by the World Health Organization in 1974 to control these diseases. Health data use is crucial in decision-making, but ensuring data quality remains challenging. The study aimed to assess the accuracy ratio, timeliness, and quality index of regular immunization programme data in the Birkod district of the Somali Region, Ethiopia. For poor data quality, technical, contextual, behavioral, and organizational factors are among contributors. The study used a quantitative cross-sectional design conducted in September 2022GC using WHO-recommended data quality self-assessment tools. The accuracy ratio and timeliness of reports on regular immunization programmes were assessed for two health centers and three health posts in the district for one fiscal year. Moreover, the quality index assessment was conducted at the district level and health facilities by trained assessors. The study found poor data quality in the accuracy ratio and timeliness of reports at all health units, which includes zeros. Overreporting was observed for most facilities, particularly at the health post level. Health centers showed a relatively better accuracy ratio than health posts. The quality index assessment revealed poor quality at all levels. The study recommends that responsible bodies at different levels improve data quality using various approaches, such as the capacitation of health professionals and strengthening the quality index components. The study highlighted the need for attention to data quality in general, specifically at the health post level, and improving the quality index at all levels, which is essential.

Keywords: Birkod District, data quality, quality index, regular immunization programme, Somali Region-Ethiopia

Procedia PDF Downloads 47

Authors: Jamal A. Gledan, Othman A. Azzeidani

Abstract:

During the last decade, Libya established a new Geodetic Datum called Libyan Geodetic Datum 2006 (LGD 2006) by using GPS, whereas the ground traversing method was used to establish the last Libyan datum which was called the Europe Libyan Datum 79 (ELD79). The current research paper introduces ELD79 to LGD2006 coordinate transformation technique, the accurate comparison of transformation between multiple regression equations and the three-parameters model (Bursa-Wolf). The results had been obtained show that the overall accuracy of stepwise multi regression equations is better than that can be determined by using Bursa-Wolf transformation model.

Keywords: geodetic datum, horizontal control points, traditional similarity transformation model, unconventional transformation techniques

Procedia PDF Downloads 278

Authors: S. Haider, B. Bhushan

Abstract:

Eye Gaze behaviour of individuals with and without intellectual disability are investigated in an eye tracking study in terms of sensitivity to Nonaccidental (NAPs) and Metric (MPs) shape properties. Total fixation time is used as an indirect measure of attention allocation. Studies have found Mean reaction times for non accidental properties (NAPs) to be shorter than for metric (MPs) when the MP and NAP differences were equalized. METHODS: Twenty-five individuals with intellectual disability (mild and moderate level of Mental Retardation) and twenty-seven normal individuals were compared on mean total fixation duration, accuracy level and mean reaction time for mild NAPs, extreme NAPs and metric properties of images. 2D images of cylinders were adapted and made into forced choice match-to-sample tasks. Tobii TX300 Eye Tracker was used to record total fixation duration and data obtained from the Areas of Interest (AOI). Variable trial duration (total reaction time of each participant) and fixed trail duration (data taken at each second from one to fifteen seconds) data were used for analyses. Both groups did not differ in terms of fixation times (fixed as well as variable) across any of the three image manipulations but differed in terms of reaction time and accuracy. Normal individuals had longer reaction time compared to individuals with intellectual disability across all types of images. Both the groups differed significantly on accuracy measure across all image types. Normal individuals performed better across all three types of images. Mild NAPs vs. Metric differences: There was significant difference between mild NAPs and metric properties of images in terms of reaction times. Mild NAPs images had significantly longer reaction time compared to metric for normal individuals but this difference was not found for individuals with intellectual disability. Mild NAPs images had significantly better accuracy level compared to metric for both the groups. In conclusion, type of image manipulations did not result in differences in attention allocation for individuals with and without intellectual disability. Mild Nonaccidental properties facilitate better accuracy level compared to metric in both the groups but this advantage is seen only for normal group in terms of mean reaction time.

Keywords: eye gaze fixations, eye movements, intellectual disability, stimulus properties

Procedia PDF Downloads 531

Authors: Dalia Rimawi, Walid Salameh, Amal Al-Omari, Hadeel AbdelKhaleq

Abstract:

Predication of Survival time of patients with cancer, is a core factor that influences oncologist decisions in different aspects; such as offered treatment plans, patients’ quality of life and medications development. For a long time proportional hazards Cox regression (ph. Cox) was and still the most well-known statistical method to predict survival outcome. But due to the revolution of data sciences; new predication models were employed and proved to be more flexible and provided higher accuracy in that type of studies. Artificial neural network is one of those models that is suitable to handle time to event predication. In this study we aim to compare ph Cox regression with artificial neural network method according to data handling and Accuracy of each model.

Keywords: Cox regression, neural networks, survival, cancer.

Procedia PDF Downloads 169

Authors: Katsumi Hirata

Abstract:

Certain systems can function well only if they recognize the sound environment as humans do. In this research, we focus on sound classification by adopting a convolutional neural network and aim to develop a method that automatically classifies various environmental sounds. Although the neural network is a powerful technique, the performance depends on the type of input data. Therefore, we propose an approach via a slice bispectrogram, which is a third-order spectrogram and is a slice version of the amplitude for the short-time bispectrum. This paper explains the slice bispectrogram and discusses the effectiveness of the derived method by evaluating the experimental results using the ESC‑50 sound dataset. As a result, the proposed scheme gives high accuracy and stability. Furthermore, some relationship between the accuracy and non-Gaussianity of sound signals was confirmed.

Keywords: environmental sound, bispectrum, spectrogram, slice bispectrogram, convolutional neural network

Procedia PDF Downloads 101

Authors: Muneeb Ullah, Daishihan, Xiadong Young

Abstract:

Deep learning plays a crucial role in identifying COVID-19 and preventing its spread. To improve the accuracy of COVID-19 diagnoses, it is important to have access to a sufficient number of training images of CXRs (chest X-rays) depicting the disease. However, there is currently a shortage of such images. To address this issue, this paper introduces COVID-19 GAN, a model that uses generative adversarial networks (GANs) to generate realistic CXR images of COVID-19, which can be used to train identification models. Initially, a generator model is created that uses digressive channels to generate images of CXR scans for COVID-19. To differentiate between real and fake disease images, an efficient discriminator is developed by combining the dense connectivity strategy and instance normalization. This approach makes use of their feature extraction capabilities on CXR hazy areas. Lastly, the deep regret gradient penalty technique is utilized to ensure stable training of the model. With the use of 4,062 grape leaf disease images, the Leaf GAN model successfully produces 8,124 COVID-19 CXR images. The COVID-19 GAN model produces COVID-19 CXR images that outperform DCGAN and WGAN in terms of the Fréchet inception distance. Experimental findings suggest that the COVID-19 GAN-generated CXR images possess noticeable haziness, offering a promising approach to address the limited training data available for COVID-19 model training. When the dataset was expanded, CNN-based classification models outperformed other models, yielding higher accuracy rates than those of the initial dataset and other augmentation techniques. Among these models, ImagNet exhibited the best recognition accuracy of 99.70% on the testing set. These findings suggest that the proposed augmentation method is a solution to address overfitting issues in disease identification and can enhance identification accuracy effectively.

Keywords: classification, deep learning, medical images, CXR, GAN.

Procedia PDF Downloads 57

Authors: Seulbi Choi, Hyunchul Ahn

Abstract:

Collaborative filtering (CF) algorithm has been popularly used for recommender systems in both academic and practical applications. It basically generates recommendation results using users’ numeric ratings. However, the additional use of the information other than user ratings may lead to better accuracy of CF. Considering that a lot of people are likely to share their honest opinion on the items they purchased recently due to the advent of the Web 2.0, user's review can be regarded as the new informative source for identifying user's preference with accuracy. Under this background, this study presents a hybrid recommender system that fuses CF and user's review mining. Our system adopts conventional memory-based CF, but it is designed to use both user’s numeric ratings and his/her text reviews on the items when calculating similarities between users.

Keywords: Recommender system, Collaborative filtering, Text mining, Review mining

Procedia PDF Downloads 312

Authors: Guofan Shao, Lina Tang, Hao Zhang

Abstract:

The values of image classification accuracy are affected by class size distributions and classification schemes, making it difficult to compare the performance of classification algorithms across different remote sensing data sources and classification systems. Based on the term efficacy from medicine and pharmacology, we have developed the metrics of image classification efficacy at the map and class levels. The novelty of this approach is that a baseline classification is involved in computing image classification efficacies so that the effects of class statistics are reduced. Furthermore, the image classification efficacies are interpretable and comparable, and thus, strengthen the assessment of image data classification methods. We use real-world and hypothetical examples to explain the use of image classification efficacies. The metrics of image classification efficacy meet the critical need to rectify the strategy for the assessment of image classification performance as image classification methods are becoming more diversified.

Keywords: accuracy assessment, efficacy, image classification, machine learning, uncertainty

Procedia PDF Downloads 181

Authors: Ahmadreza Bagheri, Seyyed S. Eftekhari, Shervin Rashidinia

Abstract:

Background: Various stress tests have been proposed yet to assess patients with suspected coronary artery disease. However, their diagnostic properties in terms of sensitivity, specificity, and accuracy are variable and their applicability remained somewhat vague. The aim of this study is to validate per-vessel diagnostic properties of 3 types of stress myocardial perfusion scintigraphy in gated SPECT (Single-Photon Emission Computed Tomography) using either exercise or pharmacological stress testing with dipyridamole or dobutamine. Materials and Methods: Hospital records of 314 patients who referred to Imam Khomeini hospital of Tehran between September 2015 and January 2017 were completely reviewed in this study. All patients underwent coronary angiography within 3 months after stress myocardial perfusion scan. Eventually, the results were analyzed in per-vessel basis to find the proper modality for each involved vessel or scanned site. Results: The mean age of patients was 62.15 ± 4.94 years (30-85) and 35.03% were women. The overall sensitivity, specificity, and accuracy were calculated as 56.59%, 54.24%, and 55.09%, respectively. These values were 56.43% and 53.25%, 54.46% and 47.36%, 56.75% and 54.83% for dipyridamole and exercise, respectively. Ischemia of the anterior wall through exercise stress testing has the highest diagnostic accuracy in detecting LAD (Left Anterior Descending artery) involvement. Inferior wall hypokinesia and anterolateral wall ischemia during exercise stress testing have the highest diagnostic accuracy in detecting RCA (Right Coronary Artery) and LCX artery (Left Circumflex Artery) stenosis, respectively. Conclusion: Stress myocardial perfusion scan should be carried out on the basis of the findings of the preliminary investigations on suspicion of a specific coronary artery or involved myocardial wall.

Keywords: dipyridamole, dobutamine, single-photon emission computed tomography, stress myocardial perfusion scintigraphy

Procedia PDF Downloads 129

Authors: Osama Terra, Hatem Hussein, Adriaan Van Brakel

Abstract:

Dense wavelength division multiplexing (DWDM) technology requires tight specification and therefore measurement of wavelength accuracy and stability of the telecommunication lasers. Thus, calibration of the used Optical Spectrum Analyzers (OSAs) that are used to measure wavelength is of a great importance. Proficiency testing must be performed on such measuring activity to insure the accuracy of the measurement results. In this paper, a new comparison scheme is introduced to test the performance of such calibrations. This comparison scheme is implemented between NIS-Egypt and NMISA-South Africa for the calibration of the wavelength scale of an OSA. Both institutes employ reference gas cell to calibrate OSA according to the standard IEC/ BS EN 62129 (2006). The result of this comparison is compiled in this paper.

Keywords: OSA calibration, HCN gas cell, DWDM technology, wavelength measurement

Procedia PDF Downloads 278

Authors: Danni Cong, Meiping Wu, Xiaofeng He, Junxiang Lian, Juliang Cao, Shaokuncai, Hao Qin

Abstract:

Gravity gradient instrument (GGI) is the core of the gravity gradiometer, so the structural error of the sensor has a great impact on the measurement results. In order not to affect the aimed measurement accuracy, limit error is required in the installation of the accelerometer. In this paper, based on the established measuring principle model, the radial installation limit error is calibrated, which is taken as an example to provide a method to calculate the other limit error of the installation under the premise of ensuring the accuracy of the measurement result. This method provides the idea for deriving the limit error of the geometry structure of the sensor, laying the foundation for the mechanical precision design and physical design.

Keywords: gravity gradient sensor, radial installation limit error, accelerometer, uniaxial rotational modulation

Procedia PDF Downloads 399

Authors: Chonlakorn Wiboonsiriruk, Ekachai Phaisangittisagul, Chadchai Srisurangkul, Itsuo Kumazawa

Abstract:

Public transportation is a crucial aspect of passenger transportation, with buses playing a vital role in the transportation service. Passenger counting is an essential tool for organizing and managing transportation services. However, manual counting is a tedious and time-consuming task, which is why computer vision algorithms are being utilized to make the process more efficient. In this study, different object detection algorithms combined with passenger tracking are investigated to compare passenger counting performance. The system employs the EfficientDet algorithm, which has demonstrated superior performance in terms of speed and accuracy. Our results show that the proposed system can accurately count passengers in varying conditions with an accuracy of 94%.

Keywords: computer vision, object detection, passenger counting, public transportation

Procedia PDF Downloads 126

Authors: Adel Khorramrouz, Monireh Ahmadi Bani, Ehsan Norouzi, Morvarid Lalenoor

Abstract:

Background: Plantar pressure distribution measurement has been used for a long time to assess foot disorders. Plantar pressure is an important component affecting the foot and ankle function and Changes in plantar pressure distribution could indicate various foot and ankle disorders. Morphologic and mechanical properties of the foot may be important factors affecting the plantar pressure distribution. Accurate and early measurement may help to reduce the prevalence of pes planus. With recent developments in technology, new techniques such as machine learning have been used to assist clinicians in predicting patients with foot disorders. Significance of the study: This study proposes a neural network learning-based flat foot classification methodology using static foot pressure distribution. Methodologies: Data were collected from 895 patients who were referred to a foot clinic due to foot disorders. Patients with pes planus were labeled by an experienced physician based on clinical examination. Then all subjects (with and without pes planus) were evaluated for static plantar pressures distribution. Patients who were diagnosed with the flat foot in both feet were included in the study. In the next step, the leg length was normalized and the network was trained for plantar pressure mapping images. Findings: From a total of 895 image data, 581 were labeled as pes planus. A computational neural network (CNN) ran to evaluate the performance of the proposed model. The prediction accuracy of the basic CNN-based model was performed and the prediction model was derived through the proposed methodology. In the basic CNN model, the training accuracy was 79.14%, and the test accuracy was 72.09%. Conclusion: This model can be easily and simply used by patients with pes planus and doctors to predict the classification of pes planus and prescreen for possible musculoskeletal disorders related to this condition. However, more models need to be considered and compared for higher accuracy.

Keywords: foot disorder, machine learning, neural network, pes planus

Procedia PDF Downloads 329

Authors: Ali Kirkbas

Abstract:

R peaks in an electrocardiogram (ECG) are signs of cardiac activity in individuals that reveal valuable information about cardiac abnormalities, which can lead to mortalities in some cases. This paper examines the problem of detecting R-peaks in ECG signals, which is a two-class pattern classification problem in fact. To handle this problem with a reliable high accuracy, we propose to use the common vector approach which is a successful machine learning algorithm. The dataset used in the proposed method is obtained from MIT-BIH, which is publicly available. The results are compared with the other popular methods under the performance metrics. The obtained results show that the proposed method shows good performance than that of the other. methods compared in the meaning of diagnosis accuracy and simplicity which can be operated on wearable devices.

Keywords: ECG, R-peak classification, common vector approach, machine learning

Procedia PDF Downloads 31

Authors: Vikrum Malhotra, Manpreet Kaur, Ayesha Ghotto

Abstract:

Background: Cardiac registries such as ACC Percutaneous Coronary Intervention Registry require high quality data to be abstracted, including data elements such as nuclear cardiology, diagnostic coronary angiography, and PCI. Introduction: The audit tool created is used by data abstractors to provide data audits and assess the accuracy and inter-rater reliability of abstraction performed by the abstractors for a health system. This audit tool solution has been developed across 13 registries, including ACC/NCDR registries, PCI, STS, Get with the Guidelines. Methodology: The data audit tool was used to audit internal registry abstraction for all data elements, including stress test performed, type of stress test, data of stress test, results of stress test, risk/extent of ischemia, diagnostic catheterization detail, and PCI data elements for ACC/NCDR PCI registries. This is being used across 20 hospital systems internally and providing abstraction and audit services for them. Results: The data audit tool had inter-rater reliability and accuracy greater than 95% data accuracy and IRR score for the PCI registry in 50 PCI registry cases in 2021. Conclusion: The tool is being used internally for surgical societies and across hospital systems. The audit tool enables the abstractor to be assessed by an external abstractor and includes all of the data dictionary fields for each registry.

Keywords: abstraction, cardiac registry, cardiovascular registry, registry, data

Procedia PDF Downloads 81

Authors: Rujia Chen, Akbar Ghobakhlou, Ajit Narayanan

Abstract:

This study investigates the task of identifying musical instruments in polyphonic compositions using Convolutional Neural Networks (CNNs) from spectrogram inputs, focusing on binary classification. The model showed promising results, with an accuracy of 97% on solo instrument recognition. When applied to polyphonic combinations of 1 to 10 instruments, the overall accuracy was 64%, reflecting the increasing challenge with larger ensembles. These findings contribute to the field of Music Information Retrieval (MIR) by highlighting the potential and limitations of current approaches in handling complex musical arrangements. Future work aims to include a broader range of musical sounds, including electronic and synthetic sounds, to improve the model's robustness and applicability in real-time MIR systems.

Keywords: binary classifier, CNN, spectrogram, instrument

Procedia PDF Downloads 14

Authors: Jonathan Sohn

Abstract:

Deep neural networks (DNNs) have shown state-of-the-art performance for many applications, including computer vision, natural language processing, and speech recognition. Recently, adversarial attacks have been studied in the context of deep neural networks, which aim to alter the results of deep neural networks by modifying the inputs slightly. For example, an adversarial attack on a DNN used for object detection can cause the DNN to miss certain objects. As a result, the reliability of DNNs is undermined by their lack of robustness against adversarial attacks, raising concerns about their use in safety-critical applications such as autonomous driving. In this paper, we focus on studying the adversarial attacks and defenses on DNNs for image classification. There are two types of adversarial attacks studied which are fast gradient sign method (FGSM) attack and projected gradient descent (PGD) attack. A DNN forms decision boundaries that separate the input images into different categories. The adversarial attack slightly alters the image to move over the decision boundary, causing the DNN to misclassify the image. FGSM attack obtains the gradient with respect to the image and updates the image once based on the gradients to cross the decision boundary. PGD attack, instead of taking one big step, repeatedly modifies the input image with multiple small steps. There is also another type of attack called the target attack. This adversarial attack is designed to make the machine classify an image to a class chosen by the attacker. We can defend against adversarial attacks by incorporating adversarial examples in training. Specifically, instead of training the neural network with clean examples, we can explicitly let the neural network learn from the adversarial examples. In our experiments, the digit recognition accuracy on the MNIST dataset drops from 97.81% to 39.50% and 34.01% when the DNN is attacked by FGSM and PGD attacks, respectively. If we utilize FGSM training as a defense method, the classification accuracy greatly improves from 39.50% to 92.31% for FGSM attacks and from 34.01% to 75.63% for PGD attacks. To further improve the classification accuracy under adversarial attacks, we can also use a stronger PGD training method. PGD training improves the accuracy by 2.7% under FGSM attacks and 18.4% under PGD attacks over FGSM training. It is worth mentioning that both FGSM and PGD training do not affect the accuracy of clean images. In summary, we find that PGD attacks can greatly degrade the performance of DNNs, and PGD training is a very effective way to defend against such attacks. PGD attacks and defence are overall significantly more effective than FGSM methods.

Keywords: deep neural network, adversarial attack, adversarial defense, adversarial machine learning

Procedia PDF Downloads 162

Authors: Syed Islam, Yiyun Huang, Sebastian Magierowski, Ebrahim Ghafar-Zadeh

Abstract:

This paper presents system level CMOS solid-state nanopore techniques enhancement for speedup next generation molecular recording and high throughput channels. This discussion also considers optimum number of base-pair (bp) measurements through channel as an important role to enhance potential read accuracy. Effective power consumption estimation offered suitable rangeof multi-channel configuration. Nanopore bp extraction model in statistical method could contribute higher read accuracy with longer read-length (200 < read-length). Nanopore ionic current switching with Time Multiplexing (TM) based multichannel readout system contributed hardware savings.

Keywords: DNA, nanopore, amplifier, ADC, multichannel

Procedia PDF Downloads 428