ABSTRACT
It is important to evaluate medical imaging artificial intelligence (AI) models for possible implicit discrimination (ability to distinguish between subgroups not related to the specific clinical task of the AI model) and disparate impact (difference in outcome rate between subgroups). We studied potential implicit discrimination and disparate impact of a published deep learning/AI model for the prediction of ICU admission for COVID-19 within 24 hours of imaging. The IRB-approved, HIPAA-compliant dataset contained 8,357 chest radiography exams from February 2020-January 2022 (12% ICU admission within 24 hours) and was separated by patient into training, validation, and test sets (64%, 16%, 20% split). The AI output was evaluated in two demographic categories: sex assigned at birth (subgroups male and female) and self-reported race (subgroups Black/African-American and White). We failed to show statistical evidence that the model could implicitly discriminate between members of subgroups categorized by race based on prediction scores (area under the receiver operating characteristic curve, AUC: median [95% confidence interval, CI]: 0.53 [0.48, 0.57]) but there was some marginal evidence of implicit discrimination between members of subgroups categorized by sex (AUC: 0.54 [0.51, 0.57]). No statistical evidence for disparate impact (DI) was observed between the race subgroups (i.e. the 95% CI of the ratio of the favorable outcome rate between two subgroups included one) for the example operating point of the maximized Youden index but some evidence of disparate impact to the male subgroup based on sex was observed. These results help develop evaluation of implicit discrimination and disparate impact of AI models in the context of decision thresholds © COPYRIGHT SPIE. Downloading of the is permitted for personal use only.
ABSTRACT
Purpose: This paper aims to illustrate the potential of high-frequency data for tourism and hospitality analysis, through two research objectives: First, this study describes and test a novel high-frequency forecasting methodology applied on big data characterized by fine-grained time and spatial resolution;Second, this paper elaborates on those estimates' usefulness for visitors and tourism public and private stakeholders, whose decisions are increasingly focusing on short-time horizons. Design/methodology/approach: This study uses the technical communications between mobile devices and WiFi networks to build a high frequency and precise geolocation of big data. The empirical section compares the forecasting accuracy of several artificial intelligence and time series models. Findings: The results robustly indicate the long short-term memory networks model superiority, both for in-sample and out-of-sample forecasting. Hence, the proposed methodology provides estimates which are remarkably better than making short-time decision considering the current number of residents and visitors (Naive I model). Practical implications: A discussion section exemplifies how high-frequency forecasts can be incorporated into tourism information and management tools to improve visitors' experience and tourism stakeholders' decision-making. Particularly, the paper details its applicability to managing overtourism and Covid-19 mitigating measures. Originality/value: High-frequency forecast is new in tourism studies and the discussion sheds light on the relevance of this time horizon for dealing with some current tourism challenges. For many tourism-related issues, what to do next is not anymore what to do tomorrow or the next week. Plain Language Summary: This research initiates high-frequency forecasting in tourism and hospitality studies. Additionally, we detail several examples of how anticipating urban crowdedness requires high-frequency data and can improve visitors' experience and public and private decision-making. (PsycInfo Database Record (c) 2023 APA, all rights reserved)
ABSTRACT
Artificial intelligence is increasingly applied in many fields, specially in medicine to assist patients and physicians. Growing datasets provide a sound basis to adapt machine learning methods to identify and detect some diseases. These later, are often very similar which make difficult their identification by chest X-ray images. In this paper, we introduce a diagnostic AI model that allow to separate, diagnose and classify three various diseases: tuberculosis, covid19 and Pneumonia. The proposed model is based on a combination of Deep Learning using the deep SqueezeNet model and Machine Learning: SVM, KNN, Logistic Regression, decision tree and Naive Bayes. The model is applied to a chest X-ray dataset containing images for each type of disease. To train and test our model, we split the image dataset into two training and test subsets in order to differentiate between different disease types. The accuracy show clearly that our model provides better results of diagnosis and identification. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
Researchers and investors have been paying close attention to the application of Artificial Intelligence models to the economics, agriculture and other fields in recent years. This study uses a Multilayer Perceptron Artificial Neural Network to anticipate the effect of covid-19 on crude-oil prices, continuing the deep learning trend and also applied the use of time series model known as Autoregressive Integrated Moving Average (ARIMA) to validate the result gotten from MLP-ANN. The results produced accurately predicted crude oil prices, and covid-19 data was also analyzed, as well as the association between crude-oil prices and covid-19. Because of the substantial causative association between the coronavirus (number of confirmed cases), crude oil prices, this study is intriguing. Ten years forecast was done using both MLP-ANN and ARIMA and from result gotten, MLP-ANN has accuracy of 96% while ARIMA has 39% accuracy. © 2022 IEEE.
ABSTRACT
Given the infection's wide growth, one of the biggest challenges on the planet right now is identifying Corona Virus Disease 2019 (COVID-19). Recent findings show that, with over 225M confirmed instances, the number of people who have been diagnosed with COVID-19 is drastically increasing;Around the world, the sickness is affecting several countries. In this study, the global COVID-19 circulation incidence is briefly examined, and a deep convolutional neural network (CNN) artificial intelligence model is developed to identify COVID19 patients using real-world information. To find such patients, the model looks at chest CT scan images. The results show that such an approach is helpful in diagnosing COVID-19 since CT scans are easily accessible fast and inexpensively. This suggested approach is effective at detecting COVID-19 and achieves an F-measure range of 95-99%, according to empirical findings from 100 CT scan pictures of actual patients. The suggested model has a considerable impact in identifying sick individuals. © 2022 IEEE.
ABSTRACT
Purpose: The purpose of this study was to demonstrate a cloud business intelligence model for industrial SMEs. An initial model was developed to accomplish this, followed by validation and finalization of the cloud business intelligence model. Additionally, this research employs a mixed-techniques approach, including both qualitative and quantitative methods. This paper aims to achieve the following objectives: (1) Recognize the Cloud business intelligence concepts. (2) Identify the role of cloud BI in SMEs. (3) Identify the factors that affect the design and presenting a Cloud business intelligence model based on critical factors affecting SMEs during pandemic COVID-19. (4) Discuss the importance of Cloud BI in pandemic COVID-19 for SMEs. (5) Provide managerial implications for using Cloud BI effectively in Iran's SMEs. Design/methodology/approach: In the current study, an initial model was first proposed, and the cloud business intelligence model was then validated and finalized. Moreover, this study uses a mixed-methods design in which both qualitative and quantitative methods are used. The fuzzy Delphi Method has been applied for parameter validation purposes, and eventually, the Cloud business intelligence model has been presented through exploiting the interpretive structural modeling. The partial least squares method was also applied to validate the model. Data were also analyzed using the MAXQDA and Smart PLS software package. Findings: In this research, from the elimination of synonym and frequently repeated factors and classification of final factors, six main factors, 24 subfactors and 24 identifiers were discovered from the texts of the relevant papers and interviews conducted with 19 experts in the area of BI and Cloud computing. The main factors of our research include drivers, enablers, competencies, critical success factors, SME characteristics and adoption. The subfactors of included competitors pressure, decision-making time, data access, data analysis and calculations, budget, clear view, clear missions, BI tools, data infrastructure, information merging, business key sector, data owner, business process, data resource, data quality, IT skill, organizational preparedness, innovation orientation, SME characteristics, SME activity, SME structure, BI maturity, standardization, agility, balances between BI systems and business strategies. Then, the quantitative part continued with the fuzzy Delphi technique in which two factors, decision-making time and agility, were deleted in the first round, and the second round was conducted for the rest of the factors. In that step, 24 factors were assessed based on the opinions of 19 experts. In the second round, none of the factors were removed, and thus the Delphi analysis was concluded. Next, data analysis was carried out by building the structural self-interaction matrix to present the model. According to the results, adoptability is a first-level or dependent variable. Regarding the results of interpretive structural modeling (ISM), the variable of critical success factors is a second-level variable. Enablers, competencies and SME characteristics are the third-level and most effective variables of the model. Accordingly, the initial model of Cloud BI for SMEs is presented as follows: The results of ISM revealed the impact of SME characteristics on BI critical success factors and adoptability. Since this category was not an underlying category of BI;thus, it played the role of a moderating variable for the impact of critical success factors on adoptability in the final model. Research limitations/implications: Since this study is limited to about 100 SMEs in the north of Iran, results should be applied cautiously to SMEs in other countries. Generalizing the study's results to other industries and geographic regions should be done with care since management perceptions, and financial condition of a business vary significantly. Additionally, the topic of business intelligence in SMEs constrained the sample from the start since not all SMEs use business int lligence systems, and others are unaware of their advantages. BI tools enable the effective management of companies of all sizes by providing analytic data and critical performance indicators. In general, SMEs used fewer business intelligence technologies than big companies. According to studies, SMEs understand the value of simplifying their information resources to make critical business choices. Additionally, they are aware of the market's abundance of business intelligence products. However, many SMEs lack the technical knowledge necessary to choose the optimal tool combination. In light of the frequently significant investment required to implement BI approaches, a viable alternative for SMEs may be to adopt cloud computing solutions that enable organizations to strengthen their systems and information technologies on a pay-per-use basis while also providing access to cutting-edge BI technologies at a reasonable price. Practical implications: Before the implementation of Cloud BI in SMEs, condition of driver, competency and critical success factor of SMEs should also be considered. These will help to define the significant resources and skills that form the strategic edge and lead to the success of Cloud BI projects. Originality/value: Most of the previous studies have been focused on factors such as critical success factors in cloud business intelligence and cloud computing in small and medium-sized enterprises, cloud business intelligence adoption models, the services used in cloud business intelligence, the factors involved in acceptance of cloud business intelligence, the challenges and advantages of cloud business intelligence, and drivers and barriers to cloud business intelligence. None of the studied resources proposed any comprehensive model for designing and implementing cloud business intelligence in small and medium-sized enterprises;they only investigated some of the aspects of this issue. © 2021, Emerald Publishing Limited.
ABSTRACT
The pandemic caused by COVID-19 has changed the way people live, forcing them to adopt measures to avoid transmission, so much so that countries have had to develop containment strategies because this virus continues to spread throughout the planet. It is important to define strategies that support physicians to prevent and improve the incidence of cases and thus avoid the collapse of health systems. The implementation of different technologies is convenient because it allows monitoring and prevention to be done more quickly. This paper analyzes emerging technologies such as mobile applications, devices (IoT) and artificial intelligence models as alternatives to traditional processes. © 2022 IEEE.
ABSTRACT
The recent COVID-19 (novel coronavirus disease) pandemic induced a deep polarization among regional as well as global communities. The sentiments regarding the pandemic and its impact on lifestyle and economy, often expressed via social networks, are regarded as critical metrics for capturing such polarization and formulating appropriate intervention by the relevant authorities. While there exist a myriad of Natural Language Processing (NLP) models for mining social media data, we demonstrate the shortcomings of the individual models in this paper, and explore how to improve the COVID-19 sentiment analysis in social media network data via two hybrid predictive models based on a Long-Short-Term-Memory (LSTM)-based autoencoder and a Convolutional Neural Network (CNN) model coupled with a bi-directional LSTM. Through extensive experiments on the recently acquired Twitter dataset, we compare the COVID-19 sentiments exhibited in the USA and Canada using our proposed hybrid predictive models and demonstrate their superiority over individual Artificial Intelligence (AI) models. © 2022 IEEE.
ABSTRACT
During the COVID-19 pandemic, face masks detection has become a critical responsibility. Since the pandemic is still ongoing, and workplaces are opening up it is absolutely necessary that all COVID-19 rules & regulations issued by the government which require all employees in a workplace to wear a face mask to be followed. In order to enforce this rule with severity a face mask detection system can be set up in the workplace. This is achieved with the help of artificial intelligence models, face recognition based neural network and transfer learning approach. Once the model is developed it will be integrated with an application which will keep reminding the employee to wear a mask and also keep records and track of the times the employee was liable to pay a penalty. © 2022 IEEE.
ABSTRACT
For decades artificial intelligence (AI) has been used for various applications in the healthcare industry. Machine learning and artificial intelligence algorithms allow us to diagnose and customize medical care and follow-up plans to get better results, and during the covid19 pandemic, it was found that AI models have been using to predict the Covid-19 symptoms, understanding how it spreads, speeding up research and treatment using medical data. However, it is very challenging to make a robust AI model and use it in a real-time and real-world environment since most organizations do not want to share their data with other third parties due to privacy concerns, furthermore, it is difficult to build a generalized prediction model because of the fragmented nature of the patient data across the healthcare system. To solve the above problems, this paper presents a solution based on blockchain and AI technologies. The blockchain will securely protect the data access and AI-based federated learning for building a robust model for global and real-time usage. © 2022 Global IT Research Institute-GiRI.
ABSTRACT
There are multiple papers published about different AI models for the COVID-19 diagnosis with promising results. Unfortunately according to the reviews many of the papers do not reach the level of sophistication needed for a clinically usable model. In this paper I go through multiple review papers, guidelines, and other relevant material in order to generate more comprehensive requirements for the future papers proposing a AI based diagnosis of the COVID-19 from chest X-ray data (CXR). Main findings are that a clinically usable AI needs to have an extremely good documentation, comprehensive statistical analysis of the possible biases and performance, and an explainability module. © 2021 IEEE.
ABSTRACT
The quality of the dataset affects the accuracy of the artificial intelligence model, but it is a lot of work to manually detect errors related to the quality evaluation of the dataset, and it may not be possible to perform quality evaluation through simple viewing. Therefore, we propose an image dataset quality measurement model, including nine evaluation metrics, and analyze the evaluation metrics from three aspects: definition, calculation formula and description. Based on the label file, the quality of the dataset file and the content of the dataset is evaluated, and the evaluation standard is given to judge whether the quality of the dataset is qualified. The measurement model and evaluation criteria proposed in this article were verified against the Cifar-10 dataset and the COVID-CT dataset, and the problems of label accuracy and label category imbalance were found, which proved the effectiveness of the method in this paper. © 2021 IEEE.