Search

This study quantifies the performance of an international cohort of cornea specialists in image-based differentiation of bacterial and fungal keratitis, identifying significant regional variation and establishing a reference standard for comparison against machine learning models.

Most robotic systems consist of heavy machine elements that help reduce the vibrations generated by motion and that support good positional accuracy. Large-sized actuators are needed for such systems. Energy consumption significantly increases in oversized robotic systems. The proposed solution to which is to design robotic systems in lightweight and flexible forms. This makes possible smaller overall structure and actuator sizes. However, such systems suffer from residual vibrations. One well-known control solution to residual vibration problem could be input shaping techniques. In this study, a new Fuzzy Logic-based intelligent input shaping selection technique has been proposed. With this proposed approach, it is possible to set optimum settling time, positioning accuracy of robotic systems, and minimum residual vibrations by implementing the crucial open-loop control approach. Comprehensive experience is needed to overcome design issues on command shaping and system modeling. Based on these design criteria constraints and extensive knowledge requirements, the recommended solution is a Fuzzy Logic-based intelligent input shaping technique for accurate modeling and parameter estimation of the systems. The proposed approach is a Fuzzy Logic-based intelligent input shaper selection and input shaper setting determination algorithm. The applicability of the proposed method was verified on the Quanser Flexible Link experimental setup.

Background Testing a hypothesis for ‘factors-outcome effect’ is a common quest, but standard statistical regression analysis tools are rendered ineffective by data contaminated with too many noisy variables. Expert Systems (ES) can provide an alternative methodology in analysing data to identify variables with the highest correlation to the outcome. By applying their effective machine learning (ML) abilities, significant research time and costs can be saved. The study aims to systematically review the applications of ES in urological research and their methodological models for effective multi-variate analysis. Their domains, development and validity will be identified. Methods The PRISMA methodology was applied to formulate an effective method for data gathering and analysis. This study search included seven most relevant information sources: WEB OF SCIENCE, EMBASE, BIOSIS CITATION INDEX, SCOPUS, PUBMED, Google Scholar and MEDLINE. Eligible articles were included if they applied one of the known ML models for a clear urological research question involving multivariate analysis. Only articles with pertinent research methods in ES models were included. The analysed data included the system model, applications, input/output variables, target user, validation, and outcomes. Both ML models and the variable analysis were comparatively reported for each system. Results The search identified n = 1087 articles from all databases and n = 712 were eligible for examination against inclusion criteria. A total of 168 systems were finally included and systematically analysed demonstrating a recent increase in uptake of ES in academic urology in particular artificial neural networks with 31 systems. Most of the systems were applied in urological oncology (prostate cancer = 15, bladder cancer = 13) where diagnostic, prognostic and survival predictor markers were investigated. Due to the heterogeneity of models and their statistical tests, a meta-analysis was not feasible. Conclusion ES utility offers an effective ML potential and their applications in research have demonstrated a valid model for multi-variate analysis. The complexity of their development can challenge their uptake in urological clinics whilst the limitation of the statistical tools in this domain has created a gap for further research studies. Integration of computer scientists in academic units has promoted the use of ES in clinical urological research.

La gestión de talento humano es un factor fundamental en el éxito de las organizaciones. La inclusión en el entorno laboral de las personas con discapacidad ha ayudado a potenciar sus cualidades y a aprovechar su talento. Muchos de los sistemas de gestión de talento humano carecen de directrices para el reclutamiento y selección de una persona con discapacidad, es por eso por lo que el presente trabajo muestra el estudio realizado a estos dos procesos indicando los factores que influyen en la asignación o no de un cargo, donde de cada candidato es considerando el nivel y tipo de discapacidad, nivel de estudios, experiencia, capacitación entre otros aspectos, enfocándose en la tarea de aplicar técnicas de aprendizaje supervisado que permitan clasificar a un candidato con discapacidad para un puesto de trabajo como apto o no y técnicas de aprendizaje no supervisado como el clustering que ayuda a definir patrones ocultos en los datos si los hubiera. El resultado obtenido del estudio presenta varias técnicas de clasificación y la selección de la más adecuada para el conjunto de datos en cuestión, igualmente mediante técnicas de aprendizaje no supervisado se determina cuántos clústeres representativos se identifican en los datos. No se busca que se integren las personas a través de las discapacidades, sino todo lo contrario, que se integren las personas por medio de la potencialidad de todas sus capacidades.// Human talent management is a key factor in the success of organizations. The inclusion of people with disabilities in the work environment has helped to enhance their qualities and harness their talent. Many of the human talent management systems lack guidelines for the recruitment and selection of a person with a disability, this is why this work shows the study carried out on these two processes indicating the factors that influence the allocation or not of a position, where for each candidate it is considered the level and type of disability, level of education, experience, training among other aspects, focusing on the task of applying supervised learning techniques that enable us to classify a candidate with a disability as suitable or not for a job, and unsupervised learning techniques such as clustering that helps us define hidden patterns in the data, if any. The result obtained from the study presents some classifications techniques and the selection of the most appropriate one for the available dataset. It is not sought to integrate people through their disabilities, but quite the opposite, to integrate people based on the potential of all their abilities.

Purpose To compare the performance of a novel convolutional neural network (CNN) classifier and human graders in detecting angle closure in EyeCam (Clarity Medical Systems, Pleasanton, California, USA) goniophotographs. Design Retrospective cross-sectional study. Methods Subjects from the Chinese American Eye Study underwent EyeCam goniophotography in 4 angle quadrants. A CNN classifier based on the ResNet-50 architecture was trained to detect angle closure, defined as inability to visualize the pigmented trabecular meshwork, using reference labels by a single experienced glaucoma specialist. The performance of the CNN classifier was assessed using an independent test dataset and reference labels by the single glaucoma specialist or a panel of 3 glaucoma specialists. This performance was compared to that of 9 human graders with a range of clinical experience. Outcome measures included area under the receiver operating characteristic curve (AUC) metrics and Cohen kappa coefficients in the binary classification of open or closed angle. Results The CNN classifier was developed using 29,706 open and 2,929 closed angle images. The independent test dataset was composed of 600 open and 400 closed angle images. The CNN classifier achieved excellent performance based on single-grader (AUC = 0.969) and consensus (AUC = 0.952) labels. The agreement between the CNN classifier and consensus labels (κ = 0.746) surpassed that of all non-reference human graders (κ = 0.578-0.702). Human grader agreement with consensus labels improved with clinical experience (P = 0.03). Conclusion A CNN classifier can effectively detect angle closure in goniophotographs with performance comparable to that of an experienced glaucoma specialist. This provides an automated method to support remote detection of patients at risk for primary angle closure glaucoma.

Purpose Nowadays, due to the rapid development of the Internet and the rapid growth of web pages, many electronic websites are using product recommendation systems to guide users to the products that they need. Such systems usually provide a list of suggested items that the user may prefer. These systems are provided as a support tool to help users obtain information that best meets their needs. These systems can actually improve user decisions, resulting in increased sales and mutual customer satisfaction. The purpose of the paper is to improve user decisions in online shopping using fuzzy expert system. Methodology: The statistical population of this study consists of 30 experts in the field of e-commerce who were selected by combining two methods of deliberate sampling and snowball sampling. To analyze the status of improvement of users' decisions, a fuzzy expert system was created using input variables business reputation status, environmental factors status in e-commerce, online store features; product specifications; user/customer characteristics. Findings: The final results showed that there is no significant difference between the results of the created expert system and the mean of expert opinions. Originality/Value: In this paper, a conceptual Model to improve user decisions in online shopping using a fuzzy expert system is designed.

In the electronic health record, using clinical notes to identify entities such as disorders and their temporality (e.g. the order of an event relative to a time index) can inform many important analyses. However, creating training data for clinical entity tasks is time consuming and sharing labeled data is challenging due to privacy concerns. The information needs of the COVID-19 pandemic highlight the need for agile methods of training machine learning models for clinical notes. We present Trove, a framework for weakly supervised entity classification using medical ontologies and expert-generated rules. Our approach, unlike hand-labeled notes, is easy to share and modify, while offering performance comparable to learning from manually labeled training data. In this work, we validate our framework on six benchmark tasks and demonstrate Trove’s ability to analyze the records of patients visiting the emergency department at Stanford Health Care for COVID-19 presenting symptoms and risk factors., In the electronic health record, using clinical notes to identify entities such as disorders and their temporality can inform many important analyses. Here, the authors present a framework for weakly supervised entity classification using medical ontologies and expert-generated rules.

A novel, multi-dimensional protocol named GENIE has been in use for intensive insulin therapy (IIT, target glucose

Predicting an individual's risk of getting a stroke has been a research subject for many authors worldwide since it is a frequent illness and there is strong evidence that early awareness of having that risk can be beneficial for prevention and treatment. Many Governments have been collecting medical data about their own population with the purpose of using artificial intelligence methods for making those predictions. The most accurate ones are based on so called black-box methods which give little or no information about why they make a certain prediction. However, in the medical field the explanations are sometimes more important than the accuracy since they allow specialists to gain insight about the factors that influence the risk level. It is also frequent to find medical information records with some missing data. In this work, we present the development of a prediction method which not only outperforms some other existing ones but it also gives information about the most probable causes of a high stroke risk and can deal with incomplete data records. It is based on the Dempster-Shafer theory of plausibility. For the testing we used data provided by the regional hospital in Okayama, Japan, a country in which people are compelled to undergo annual health checkups by law. This article presents experiments comparing the results of the Dempster-Shafer method with the ones obtained using other well-known machine learning methods like Multilayer perceptron, Support Vector Machines and Naive Bayes. Our approach performed the best in these experiments with some missing data. It also presents an analysis of the interpretation of rules produced by the method for doing the classification. The rules were validated by both medical literature and human specialists.

The paper presents the research aimed at setting up and developing a software system for the printing process parameters identification based on modern computer and software systems, algorithmic principles, principles of expert systems construction and advanced learning. Thus, the possibilities of application of contemporary software tools were investigated, which facilitates the process and forms the program structure of the model that uses programming languages based on the expert systems construction principles and tools for the development of system model based on the principles of modern learning. For complex model development, concepts of process knowledge bases with influential process parameters of printing technique have been developed through modelling and construction based on the logic of expert systems with the presentation, use and involvement of experts knowledge in decision making with the evaluation of the impact of individual parameters. In addition to this approach, a module was developed using modern software tools based on an algorithmic principle and a module for identifying printing process parameters using modern platforms based on advanced learning. Sophisticated software model has been made through the research and developed with databases of process parameter identification systems based on modern software tools. This tool enables a significant expedition of the solution resolving, thus improving the graphical production process and the processes of acquiring and expanding knowledge. The model is based on integrative modules: a printing process parameters identification system based on algorithmic program structure systems, a printing process parameters identification system based on expert system building principles, and a printing process parameter identification system based on modern learning systems.

Data, information, and knowledge processing systems, in the domain of healthcare, are currently plagued by heterogeneity at various levels. Current solutions have focused on developing a standard-based, manual intervention mechanism, which requires a large number of human resources and necessitates the realignment of existing systems. State-of-the-art methodologies in the field of natural language processing and machine learning can help to partially automate this process, reducing the resource requirements and providing a relatively good multi-class-based classification algorithm. We present a novel methodology for bridging the gap between various healthcare data management solutions by leveraging the strength of transformer-based machine learning models, to create mappings between the data elements. Additionally, the annotated data, collected against five medical schemas and labeled by four annotators is made available for helping future researchers. Our results indicate, that for biased, dependent multi-class text classification, transformer-based models provide better results than linguistic and other classical models. In particular, the Robustly Optimized BERT Pretraining Approach (RoBERTa) provides the best schema matching performance by achieving a Cohen’s kappa score of 0.47 and Matthews Correlation Coefficient (MCC) score of 0.48, with human-annotated data.

BACKGROUND: To combat COVID-19, curb the pandemic, and manage containment, governments around the world are turning to data collection and population monitoring for analysis and prediction. The massive data generated through the use of big data and artificial intelligence can play an important role in addressing this unprecedented global health and economic crisis. OBJECTIVES: The objective of this work is to develop an expert system that combines several solutions to combat COVID-19. The main solution is based on a new developed software called General Guide (GG) application. This expert system allows us to explore, monitor, forecast, and optimize the data collected in order to take an efficient decision to ensure the safety of citizens, forecast, and slow down the spread’s rate of COVID-19. It will also facilitate countries’ interventions and optimize resources. Moreover, other solutions can be integrated into this expert system, such as the automatic vehicle and passenger sanitizing system equipped with a thermal and smart High Definition (HD) cameras and multi-purpose drones which offer many services. All of these solutions will facilitate lifting COVID-19 restrictions and minimize the impact of this pandemic. METHODS: The methods used in this expert system will assist in designing and analyzing the model based on big data and artificial intelligence (machine learning). This can enhance countries’ abilities and tools in monitoring, combating, and predicting the spread of COVID-19. RESULTS: The results obtained by this prediction process and the use of the above mentioned solutions will help monitor, predict, generate indicators, and make operational decisions to stop the spread of COVID-19. CONCLUSIONS: This developed expert system can assist in stopping the spread of COVID-19 globally and putting the world back to work.

Introduction. The paper considers problems of creating information support for solving the task of assessing the maturity level of an organization. It is proposed to use intelligent information systems, i.e. expert systems. Substantive aspects of various stages of creating such systems are briefly described; the expert system architecture, which is based on using a fuzzy expert knowledge base, is given. The work objective was to create new software to solve the problem of assessing the maturity level of an organization. Materials and Methods. Previously performed modeling of the subject domain under consideration allowed us to create a knowledge base in the form of production memory, which is the basis of the fuzzy inference mechanism. The software is written in PHP and is suitable for embedding in complex web applications. The software system is a web application written primarily in PHP and JavaScript. The software works in all modern web browsers, which accelerates significantly the implementation and deployment based on both the parent-enterprise and its subsidiaries. Results. New software has been created to automate the processing of questionnaires during the organization’s selfassessment based on key indicators, as well as considering 6 main groups of the quality management system indicators. Application of the program will significantly speed up the process of input and processing of expert information required for self-assessment. The program provides organizations to get an adequate idea of the opportunities and prospects for improving the organization’s quality management system. Some fragments of the software system interface are given. Discussion and Conclusions. The proposed software can be used to determine the level of maturity of an organization. The application of Web-technologies improves usability, reduces software support costs. The software can be both deployed in the existing network infrastructure of a customer and used by all the functionality through connecting to a remote server. The software is optimized for various screen resolutions, which allows you to use it not only at the central office, but also when analyzing the quality management system of corporate customers. The traffic generated by the web application is optimized for working with mobile devices with a low-speed Internet connection. Application of the program will significantly reduce the time for users to enter and process expert information required for the problem solving and to eliminate duplication of information.

The paper focuses on the development of a reliable medical expert system for diagnosis of low back pain (LBP) by proposing an efficient frame-based knowledge representation scheme and a suitable resolution logic with conflicts in outcomes being resolved using Bayesian network. Considering that LBP is classified into many diseases based on different pain generators, the proposed methodology infers non-conflicting LBP diseases sorted according to their chances of occurrence. A satisfactory clinical efficacy (average relative error - 0.09, recall 74.44%, precision 76.67%, accuracy 71.11%, and F1-score 73.88%) of the proposed methodology has been found after validating the design with empirically selected thirty LBP patient cases. Constraining that an inferred disease having chance of occurrence, prior to pathological investigations, below 0.75 (as set by four pain specialists) is not accepted clinically; the design can correctly identify, on average, 74.44% of actual diagnosis; and 76.67% of inferred diagnosis is included in actual diagnosis. With the predicted chance of occurrence being lower than 0.75 by a fraction of 0.09 on average, the proposed design performs well for 73.88% cases detecting 71.11% inferred outcomes as accurate. The design offers homogeneity to the actual outcomes, with the chi-squared static being calculated as 11.08 having 12 as degree of freedom. Graphical abstract.

The use of expert systems in the world of civil engineering, and in particular for roads, has become a necessity for the reason of the particularity, complexity, and diversity of the influencing parameters at the level of the design calculation, the latter of which represents the major source of subsequent degradation. This system consists of proposing a tool for helping the user firstly to size the body of the roadway, with several analytical methods and models (Pre-project, Boussinesq, Westgaard, and Burmister), and secondly, to offer different design possibilities (thickness and type of the material) that make up the layers. Lastly, it is to calculate the stresses and strains in order to compare them with admissible limits. The management of a knowledge base of complex natures (words, sentences, numbers, symbols, tables, calculation methods, equations, conditions, etc.) requires an adequate methodology which goes beyond the simple use of the technology but enables you to imagine the process of regrouping this mass of complex data and classifying the data, which can then be integrated into a database or spreadsheets and external programs designed with code compatible with the expert system generator. Our contribution relates initially to the formulation, organization, and preparation of the algorithms, and then the starting of the programs in order to conceive fully executable programs, the latter of which we can call the expert system. The validation of such a system was made as the work progressed, changes were made in the formulation of the rules, and the order and orientation of the data in the knowledge that the advantage of this type of system is the possibility of permanently reinforcing the database with human expertise in the field, or in books, especially so that we can avoid data loss due to illnesses, retirement, etc.

Monitoring current communication networks and services is an increasingly complex task as a result of a growth in the number and variety of components involved. Moreover, different perspectives on network monitoring and optimisation policies must be considered to meet context-dependent monitoring requirements. To face these demanding expectations, this article proposes a semantic-based approach to support the flexible configuration of context-aware network monitoring, where traffic sampling is used to improve efficiency. Thus, a semantic layer is proposed to provide with a standard and interoperable description of the elements, requirements and relevant features in the monitoring domain. On top of this description, semantic rules are applied to make decisions regarding monitoring and auditing policies in a proactive and context-aware manner. Use cases focusing on traffic accounting and traffic classification as monitoring tasks are also provided, demonstrating the expressiveness of the ontology and the contribution of smart SWRL rules for recommending optimised configuration profiles., This work is financed by National Funds through the Portuguese funding agency, FCT FundacAo para a Ciencia e Tecnologia within Project UIDB /50014/2020 and also by FCT within the R&D Units Project Scope: UIDB/00319/2020.

BACKGROUND: There are limited data about Xpert-Ultra performance in different settings, in HIV-infected persons, in those with a history of previous TB, and with trace readouts. METHODS: We evaluated the relative accuracy of Xpert-MTB/RIF and Xpert-Ultra in 272 selected but well-characterized archived sputum samples. Of these, 168 were culture-positive (64/168 smear-positive and 104/168 smear-negative), and 104 were culture-negative (102/104 from patients with previous TB and 2/104 from patients without a TB history). Assay-specific limit-of-detection (LOD) experiments were conducted using serial dilutions of Mycobacterium tuberculosis H37Rv. RESULTS: Overall sensitivity (95%CI) in smear-negative culture-positive samples for Xpert-MTB/RIF and Xpert-Ultra were 71.2% (62.5-79.9) and 77% (68.9-85.1), respectively (and in HIV-infected persons: 63.5% (50-76.1) and 73.1% (61.1-85.2), respectively). The LOD for Xpert-Ultra was lower (9 versus 184 CFU/ml). There were a total of 9/272 (3.3%) Xpert Ultra trace readouts (6/104 [5.8%]) in smear-negative culture-positive persons, and 3/102 (3%) in culture-negative non-TB persons with a history of previous TB). CONCLUSIONS: Xpert-Ultra had a lower LOD compared to Xpert-MTB/RIF. A small proportion of samples (

The developing countries are still starving for the betterment of health sector. The disease commonly found among the women is breast cancer, and past researches have proven results that if the cancer is detected at a very early stage, the chances to overcome the disease are higher than the disease treated or detected at a later stage. This article proposed cloud-based intelligent BCP-T1F-SVM with 2 variations/models like BCP-T1F and BCP-SVM. The proposed BCP-T1F-SVM system has employed two main soft computing algorithms. The proposed BCP-T1F-SVM expert system specifically defines the stage and the type of cancer a person is suffering from. Expert system will elaborate the grievous stages of the cancer, to which extent a patient has suffered. The proposed BCP-SVM gives the higher precision of the proposed breast cancer detection model. In the limelight of breast cancer, the proposed BCP-T1F-SVM expert system gives out the higher precision rate. The proposed BCP-T1F expert system is being employed in the diagnosis of breast cancer at an initial stage. Taking different stages of cancer into account, breast cancer is being dealt by BCP-T1F expert system. The calculations and the evaluation done in this research have revealed that BCP-SVM is better than BCP-T1F. The BCP-T1F concludes out the 96.56 percentage accuracy, whereas the BCP-SVM gives accuracy of 97.06 percentage. The above unleashed research is wrapped up with the conclusion that BCP-SVM is better than the BCP-T1F. The opinions have been recommended by the medical expertise of Sheikh Zayed Hospital Lahore, Pakistan, and Cavan General Hospital, Lisdaran, Cavan, Ireland.

The paper considers the analysis of the development of artificial intelligence systems. There were methods of analysis, comparison and deduction applied. Despite important achievements at the solution of some private tasks and big financing, the industry in general faces serious problems of development. Already in the nearest future the shortage of computing power will become serious restriction of deep machine learning. A review of literature showed that the commercialization of artificial intelligence and big data negatively affects the solution of fundamental problems in the industry development. The author shows that priorities of developers of systems of artificial intelligence are more and more displaced towards realization of simple consumer services while the solution of tasks, really important for all mankind, is not implemented. Dangers of misunderstanding by people of mechanisms of decisions development are revealed by intellectual computer systems. It is specified the signs allowing to assume possibility of global economic problems because of the developing pessimism of investors concerning prospects of the companies engaged in this field. The principal results of the study are recommended to specialists in the development of artificial intelligence systems as a part of a big data creation.

Introduction . A mathematical model of the industrial safety monitoring system in mechanical engineering is investigated. The work objective was to create a mathematical model based on expert assessments of workplace safety parameters with a calculated and experimental justification of its applicability to the “STRAZH” expert security monitoring system. Materials and Methods . The classification of expert systems for engineering enterprises is proposed. The stages of creating expert systems are considered. A methodology for assessing the consistency of experts as a basis for models of expert systems in the field of mechanical facilities safety is presented. Results . The basic safety parameters of the workplace are identified. A matrix of expert evaluation of parameters based on the opinion of leading experts in the field of engineering is created. The results of modeling the expert system “STRAZH” with the calculated and empirical support of the mathematical model validity are presented. The advantages of implementing expert systems to increase the level of personnel safety are proved. Discussion and Conclusions . The results obtained have a high degree of expert coordination and can be used in the development of expert safety monitoring systems for engineering enterprises.

Blood pressure (BP) variability is one of the important risk factors of cardiovascular disease (CVD). “Surge BP,” which represents short-term BP variability, is defined as pathological exaggerated BP increase capable of triggering cardiovascular events. Surge BP is effectively evaluated by our new BP monitoring device. To the best of our knowledge, we are the first to develop an algorithm for the automatic detection of surge BP from continuous “beat-by-beat” (BbB) BP measurements. It enables clinicians to save significant time identifying surge BP in big data from their patients’ continuous BbB BP measurements. A total of 94 subjects (74 males and 20 females) participated in our study to develop the surge BP detection algorithm, resulting in a total of 3272 surges collected from the study subjects. The surge BP detection algorithm is a simple classification model based on supervised learning which formulates shape of surge BP as detection rules. Surge BP identified with our algorithm was evaluated against surge BP manually labeled by experts with 5-fold cross validation. The recall and precision of the algorithm were 0.90 and 0.64, respectively. Processing time on each subject was 11.0 ± 4.7 s. Our algorithm is adequate for use in clinical practice and will be helpful in efforts to better understand this unique aspect of the onset of CVD. Graphical abstractSurge blood pressure (surge BP) which is defined as pathological short-term (several tens of seconds) exaggerated BP increase capable of triggering cardiovascular events. We have already developed a wearable continuous beat-by-beat (bBb) BP monitoring device and observed surge BPs successfully in obstructive sleep apnea patients. In this, we developed an algorithm for the automatic detection of surge BP from continuous BbB BP measurements to save significant time identifying surge BP among > 30,000 BbB BP measurements. Our result shows this algorithm can correctly detect surge BPs with a recall of over 0.9. Electronic supplementary material The online version of this article (10.1007/s11517-020-02162-4) contains supplementary material, which is available to authorized users.

Classification of high dimensional data suffers from curse of dimensionality and over-fitting. Neural tree is a powerful method which combines a local feature selection and recursive partitioning to solve these problems, but it leads to high depth trees in classifying high dimensional data. On the other hand, if less depth trees are used, the classification accuracy decreases or over-fitting increases. This paper introduces a novel Neural Tree exploiting Expert Nodes (NTEN) to classify high-dimensional data. It is based on a decision tree structure, whose internal nodes are expert nodes performing multi-dimensional splitting. Any expert node has three decision-making abilities. Firstly, they can select the most eligible neural network with respect to the data complexity. Secondly, they evaluate the over-fitting. Thirdly, they can cluster the features to jointly minimize redundancy and overlapping. To this aim, metaheuristic optimization algorithms including GA, NSGA-II, PSO and ACO are applied. Based on these concepts, any expert node splits a class when the over-fitting is low, and clusters the features when the over-fitting is high. Some theoretical results on NTEN are derived, and experiments on 35 standard data show that NTEN reaches good classification results, reduces tree depth without over-fitting and degrading accuracy.

Objective: Statistical knowledge about many patients could be exploited using machine learning to provide supporting information to otolaryngologists and other hearing health care professionals, but needs to be made accessible. The Common Audiological Functional Parameters (CAFPAs) were recently introduced for the purpose of integrating data from different databases by providing an abstract representation of audiological measurements. This paper aims at collecting expert labels for a sample database and to determine statistical models from the labelled data set. Design: By an expert survey, CAFPAs as well as labels for audiological findings and treatment recommendations were collected for patients from the database of Hörzentrum Oldenburg. Study sample: A total of 287 single patient cases were assessed by twelve highly experienced audiological experts. Results: The labelled data set was used to derive probability density functions for categories given by the expert labels. The collected data set is suitable for estimating training distributions due to realistic variability contained in data for different, distinct categories. Suitable distribution functions were determined. The derived training distributions were compared regarding different audiological questions. Conclusions: The method-expert survey, sorting data into categories, and determining training distributions - could be extended to other data sets, which could then be integrated via the CAFPAs and used in a classification task.

Accurate and early detection of machine faults is an important step in the preventive maintenance of industrial enterprises. It is essential to avoid unexpected downtime as well as to ensure the reliability of equipment and safety of humans. In the case of rotating machines, significant information about machine’s health and condition is present in the spectrum of its vibration signal. This work proposes a fault detection system of rotating machines using vibration signal analysis. First, a dataset of 3-dimensional vibration signals is acquired from large induction motors representing healthy and faulty states. The signal conditioning is performed using empirical mode decomposition technique. Next, multi-domain feature extraction is done to obtain various combinations of most discriminant temporal and spectral features from the denoised signals. Finally, the classification step is performed with various kernel settings of multiple classifiers including support vector machines, K-nearest neighbors, decision tree and linear discriminant analysis. The classification results demonstrate that a hybrid combination of time and spectral features, classified using support vector machines with Gaussian kernel achieves the best performance with 98.2% accuracy, 96.6% sensitivity, 100% specificity and 1.8% error rate.

This paper considers the use of a post metadata-based approach to identifying intentionally deceptive online content. It presents the use of an inherently explainable artificial intelligence technique, which utilizes machine learning to train an expert system, for this purpose. It considers the role of three factors (textual context, speaker background, and emotion) in fake news detection analysis and evaluates the efficacy of using key factors, but not the inherently subjective processing of post text itself, to identify deceptive online content. This paper presents initial work on a potential deceptive content detection tool and also, through the networks that it presents for this purpose, considers the interrelationships of factors that can be used to determine whether a post is deceptive content or not and their comparative importance.

The purpose of the present study was to assess the agreement at the categorical level between the Vitek 2 system and the Colibri coupled to the Radian under real routine laboratory conditions. The 675 nonduplicate clinical strains included in this study (249 Enterobacterales isolates, 198 Pseudomonas aeruginosa, 107 Staphylococcus aureus, 78 coagulase-negative staphylococci, 38 Enterococcus faecalis, and 5 Enterococcus faecium) were isolated from nonconsecutive clinical samples referred to our laboratory between June and November 2020. In addition, 43 carbapenemase-producing Enterobacterales (CPE) formerly identified and stored in our laboratory were added to the panel, for a total of 718 strains. The overall categorical agreements between the two compared methods were 99.3% (4,350/4,380; 95% CI 99% to 99.5%); 98.6% (2,147/2,178; 95% CI 98.0% to 99.0%); 99.4% (1,839/1,850; 95% CI 98.9% to 99.7%); and 99.4% (342/344; 95% CI 97.9% to 99.8%) for Enterobacterales, P. aeruginosa, Staphylococcus spp., and Enterococcus spp., respectively. The most important cause of the very major errors encountered on the Vitek 2 for P. aeruginosa (62%, 13/21) was related to the presence of heteroresistant populations. Among the 43 CPE included in this study, one OXA-48-like, and one OXA-181-like were missed by the Vitek 2, even by rigorously applying the CPE screening cutoffs defined by EUCAST. The Colibri coupled to the Radian provide a fully automated solution for antimicrobial disk diffusion susceptibility testing with an accuracy that is equal to or better than that of the Vitek 2 system.

This paper examines how expert system may negatively or positively influence ethical based decision making process in an organization. Expert system ethical characteristics are chosen; including lack of human intelligence, lack of emotions, accidental bias and lack of values. Depending on quantitative approach; and through distributing a questionnaire on (132) GM, Deputy GM, project manager and officer, it appeared that expert systems' characteristics negatively influenced on the degree of ethics within the organizational setting. According to analysis, it appears that lack of values and human intelligence were among the characteristics that hinder the ethical stream adoption within an organization leading to ethics problems and mal-functioning on the decision making process. On the other hand, lack of emotions appeared to have good impact on the ethical efforts within an organization. Researchers recommended taking extra measures of surveillance in terms of ethics for individuals who are supposed to develop and monitor expert systems.

High complexity, meaning a model in which components interact in multiple ways and follow certain local rules, is a huge challenge for brain research. This paper presents a semantic vector-driven closed-loop model, namely THINKING-LOOP, for brain computing to improve the understanding and development of complex cognition. The proposed model is a three-layer fusion of data, information and knowledge with human intelligence, which exploits ontological knowledge modeling, rule-based reasoning and a human-computer interaction mechanism. The interaction and collaboration within the model depend on a pair of complementary schemes in a loop: the top-down scheme from the knowledge layer to the data layer that is used to search for stable cognitive patterns; and the bottom-up scheme from the data layer to the knowledge layer that is used to deeply analyze cognitive functions. As a key factor, human beings participate in the whole learning process of the model, which in turn assists human beings to make decisions. To verify the applicability of the present model in cognitive research, a series of fMRI experiments and analytic methods (e.g. statistical tests and network topology analysis) were conducted. The results show that the proposed model is able to take into account the characteristics of different types of brain patterns and cognitive functions, thereby achieving reasonable decision-making level.

This paper proposes the creation of an intelligent system for assessing the levels of professional competencies of students in e-learning. Mathematical models and methods for assessing the formation of professional competencies of students at the level of disciplines, modules and the entire educational program are built. Algorithms for calculating the assessment of levels of competency in the discipline, module, and educational program are developed. These estimates can be used in the formation of inference rules for making decisions on changing the structure and content of disciplines, modules and educational programs.

Medicated chewing gums represent an orally administered dosage form with promising potential for local and systemic drug delivery. However, compared to other solid oral dosage forms, formulation development and release mechanism of medicated chewing gums are extremely complex, and thus only few products reached the approval for the market so far. Therefore, Quality by Design (QbD) approaches for rational formulation development of medicated chewing gums are needed to utilize their full potential. For chewing gum tablets, which are manufactured by direct compression, QbD approaches derived from tableting processes might be exerted. In this context, the SeDeM-Diagram-Expert-System implements the QbD approach while indicating whether a blend is suitable for direct compression and comprises powder properties, which need to be improved to facilitate the formulation development. Here, we present the successful application of the SeDeM-Diagram-Expert-System to the formulation development of medicated chewing gum tablets manufactured by direct compression. Furthermore, limitations of the SeDeM-System for medicated chewing gum tablets are evaluated and potential modifications of the system are suggested and discussed for future use.

Objective To identify factors that prevent and promote uptake of medication-related computerized decision support systems (CDSS) in hospitals, based on the perceptions of prescribers. Materials and methods Databases Medline, Embase, CINAHL, PubMed and PsycINFO and the top five medical informatics journals were searched. English papers published after 2002, which used a qualitative approach to examine prescriber views of CDSS in hospitals were included. Qualitative data were extracted and mapped to the three domains of the HOT-fit framework (human, organization, and technology). Results Factors preventing CDSS uptake were perceived threats to autonomy, CDSS conflicting with personal prescribing preferences, and mistrust of CDSS information. Factors promoting CDSS uptake were perceptions that CDSS improves safety and efficiency, and is easy to use. With respect to medication alerts, large numbers of irrelevant alerts reportedly led to alerts being ignored. When using order sentences/order sets, long lists of options led to excessive scrolling or clicks, and resulted in users opting for free text ordering. Discussion and conclusion To promote medication-related CDSS uptake, it is recommended that prescribers’ perspectives on CDSS usability and integration into workflow be sought during the design phase, that evidence on CDSS’ effectiveness to improve safety be provided to prescribers, and that system information be kept up to date. To improve alert uptake, organizations should ensure that alerts and minimal and relevant. To improve uptake of order sentences/order sets, organizations should minimise the number of options available to prescribers. Future work should focus on exploring prescriber perceptions of other types of CDSS.

Background and objective: Ectopic pregnancy is an important cause of morbidity and mortality worldwide. An early diagnosis, as well as the choice of the most suitable treatment for the patient is crucial to avoid possible complications. According to different factors an ectopic pregnancy must be treated from surgery, using a pharmacological treatment or following a conservative treatment. In this paper, a clinical decision support systems based on artificial intelligence algorithms has been developed to help clinicians to choose the initial treatment to be followed by the patient. Methods: A decision support system based on a three stages classifier has been developed. Each stage acts as a filter and allows re-evaluation of the classification made in the previous stage in order to find diagnostic errors. This classifier has been implemented and tested for four different aid algorithms: Multilayer Perceptron, Deep Learning, Support Vector Machine and Naives Bayes. Results: The results prove that the evaluated algorithms Support Vector Machine and Multilayer Perceptron can be useful to help gynecologists in their decisions about initial treatment, especially with Support Vector Machine that presents accuracy, sensitivity and specificity outcomes about 96.1%, 96% and 98%, respectively. Conclusions: According to the results, it is feasible to develop a clinical decision support system using the algorithms that present a higher precision. This system would help gynecologists to take the most accurate decision about the initial treatment, thus avoiding future complications. This work has been granted by the Ministerio de Economá y Competitividad of the Spanish Government (ref. TIN2014-53067-C3-1-R) and Alberto De Ramón Fernández is supported by grant BES-2015-073611.

The task of biomedical diagnostics is to determine the dependence of the conclusion/diagnosis on the sets of parameters that characterize the state of the biosystem/patient. The performed analytical review of modern scientific literature permitted us to determine that Bayesian, regression and correlation analyzes and logical programming are most often used for biomedical diagnostics purposes. At the same time, their informativeness can only be realized for the solution of those diagnostic tasks in which quantitative parameters are analyzed. Qualitative and binary information provides an opportunity to find out more about features of the biosystem’s state. However, its use is limited, since the results obtained are presented in words (that is, in a linguistic form) that cannot be processed by means of traditional (digital) mathematical analysis. The objective of this work was determining the capabilities of the mathematical apparatus to deepen the study of hormonopoiesis in the thyroid gland. The object of the study was electron micrographs of ultrathin tissue sections, its subject was the features of correlations between ultrastructural cell elements which carry out the processes of synthesis and secretion in follicular thyrocytes. In the context of studying the features of synthetic and secretory activity of follicular thyrocytes of the thyroid glands in white male rats, it was shown that the objectification of non-numerical information about the state of cells allows us to use linguistic information about changes in their morphofunctional state. Implementation of correlation analysis for studying the relationships and interdependencies between organelles which implement synthesis and secretion of the hormonal product in the main structural unit of the thyroid gland – follicular thyrocyte – allows us to determine, study, analyze and generalize peculiarities of both changes in individual ultrastructures and their functional complexes (clusters) in response to the actions of various factors and to trace the interdependencies and mutual interactions existing between them, as well as to deepen the idea of the intimate mechanism features of the specific directions of follicular thyrocyte activity, which substantially expands the research platform in cytophysiology and cytomorphology.

For very serious crimes, reporting scientists often have to contend with complex cases where literally hundreds of items are submitted by investigators for analysis. In order to efficiently expedite the challenge of comparing reference profiles to evidence profiles, many of which are mixtures, we have developed an investigative open source expert system CaseSolver. We have analysed a real case based on GlobalFiler involving 119 evidence profiles and 3 reference profiles. To provide a demonstration of the power of the system we also added the three references to a fictive large database of 1 million individuals in order to test subsequent recovery of the presumed true contributors. CaseSolver was used on a Fusion 6C validation study involving 25 two- to four-person mixture profiles based on 14 reference profiles. The sequential use of simple allele comparison, the qualitative model ( forensim) and the quantitative model ( EuroForMix) makes the analysis very fast and accurate – and finally, the software generates a list of potential match candidates which can be exported as a report. From these two studies we found that the resolution of match candidates from CaseSolver was the same as that reported by a scientist who worked manually through the samples, except that CaseSolver highlighted two manual errors. For the validation study we found low template DNA samples giving negative results, which demonstrate the limitations of the tool; but overall our assessment shows that CaseSolver will benefit all analyses involving mixture interpretation and screening. Importantly, CaseSolver removes the very time-consuming aspect of manual comparison and gives improved quality by preventing manual errors.

Objective To examine performance of a sepsis surveillance system in a simulated environment where modifications to parameters and settings for identification of at-risk patients can be explored in-depth. Materials and Methods This was a multiple center observational cohort study. The study population comprised 14 917 adults hospitalized in 2016. An expert-driven rules algorithm was applied against 15.1 million data points to simulate a system with binary notification of sepsis events. Three system scenarios were examined: a scenario as derived from the second version of the Consensus Definitions for Sepsis and Septic Shock (SEP-2), the same scenario but without systolic blood pressure (SBP) decrease criteria (near SEP-2), and a conservative scenario with limited parameters. Patients identified by scenarios as being at-risk for sepsis were assessed for suspected infection. Multivariate binary logistic regression models estimated mortality risk among patients with suspected infection. Results First, the SEP-2-based scenario had a hyperactive, unreliable parameter SBP decrease >40 mm Hg from baseline. Second, the near SEP-2 scenario demonstrated adequate reliability and sensitivity. Third, the conservative scenario had modestly higher reliability, but sensitivity degraded quickly. Parameters differed in predicting mortality risk and represented a substitution effect between scenarios. Discussion Configuration of parameters and alert criteria have implications for patient identification and predicted outcomes. Conclusion Performance of scenarios was associated with scenario design. A single hyperactive, unreliable parameter may negatively influence adoption of the system. A trade-off between modest improvements in alert reliability corresponded to a steep decline in condition sensitivity in scenarios explored.

Objective: Celiac disease is a genetic disease affecting people of all ages, resulting in small intestine enteropathy. It is considered to be a clinical chameleon. Average prevalence of cel...

BACKGROUND Screening tools are critical in the early prevention of cervical cancers. Acetic-acid based colposcopy test provides an affordable solution but is often challenged by the lack of experienced doctors in under-developed districts. OBJECTIVE To evaluate the feasibility of an internet-based expert system to better utilize the expert knowledge. METHODS A centralized internet-based expert system was developed to upload acetowhite images acquired by an integrated colposcopy device at remote healthcare centers. The detection rates of high grade cervical intraepithelial neoplasia (CIN) during two consecutive years (installed in the second year) were collected at two study sites and twelve control sites. RESULTS No difference in the detection rate was observed between study and control sites at baseline (year 1); however, greater detection rates were found at study sites after introducing the internet-based expert system (year 2). Comparing the detection rates at each site between year 1 and year 2, the study site exhibited improved detection rates. User feedbacks suggested "great learning experience from expert feedback" as the most recognized benefits. CONCLUSIONS This study shows the feasibility of an internet-based expert system in screening cervical cancers and lays the foundation to future population-level application in under-developed districts.

Background: The rise of biomedical expert heuristic knowledge-based approaches for computational modeling and problem solving, for scientific inquiry and medical decision-making, and for consultation in the 1970’s led to a major change in the paradigm that affected all of artificial intelligence (AI) research. Since then, AI has evolved, surviving several “winters”, as it has oscillated between relying on expensive and hard-to-validate knowledge-based approaches, and the alternative of using machine learning methods for inferring classification rules from labelled datasets. In the past couple of decades, we are seeing a gradual but progressive intertwining of the two. Objectives: To give an overview of early directions in AI in medicine and threads of some subsequent developments motivated by the very different goals of scientific inquiry for biomedical research, and for computational modeling of clinical reasoning and more general healthcare problem solving from the perspective of today’s “AI-Deep Learning Boom”. To show how, from the beginning, AI was central to Biomedical and Health Informatics (BMHI), as a field investigating how to understand intelligent thinking in dealing professionally with the practice for healthcare, developing mathematical models, technology, and software tools to aid human experts in biomedicine, despite many previous bouts of “exuberant optimism” about the methodologies deployed. Methods: An overview and commentary on some of the early research and publications in AI in biomedicine, emphasizing the different approaches to the modeling of problems involved in clinical practice in contrast to those of biomedical science. A concluding reflection of a few current challenges and pitfalls of AI in some biomedical applications. Conclusion: While biomedical knowledge-based systems played a critical role in influencing AI in its early days, 50 years later they have taken a back seat behind “Deep Learning” which promises to discover knowledge structures for inference and prediction, both in science and for clinical decision-support. Early work on AI for medical consultation turned out to be more useful for explanation and teaching than for clinical practice, as had been originally intended. Today, despite the many reported successes of deep learning, fundamental scientific challenges arise in drawing on models of brain science, cognition, and language, if AI is to augment and complement rather than replace human judgment and expertise in biomedicine while also incorporating these advances for translational medicine. Understanding clinical phenotypes and how they relate to precision and personalization of care requires not only scientific inquiry, but also humanistic models of treatment that respond to patient and practitioner narrative exchanges, since it is the stories and insights of human experts which encourage what Norbert Weiner termed the ethical “human use of human beings”, so central to adherence to the Hippocratic Oath

Objective Clinician information overload is prevalent in critical care settings. Improved visualization of patient information may help clinicians cope with information overload, increase efficiency, and improve quality. We compared the effect of information display interventions with usual care on patient care outcomes. Materials and Methods We conducted a systematic review including experimental and quasi-experimental studies of information display interventions conducted in critical care and anesthesiology settings. Citations from January 1990 to June 2018 were searched in PubMed and IEEE Xplore. Reviewers worked independently to screen articles, evaluate quality, and abstract primary outcomes and display features. Results Of 6742 studies identified, 22 studies evaluating 17 information displays met the study inclusion criteria. Information display categories included comprehensive integrated displays (3 displays), multipatient dashboards (7 displays), physiologic and laboratory monitoring (5 displays), and expert systems (2 displays). Significant improvement on primary outcomes over usual care was reported in 12 studies for 9 unique displays. Improvement was found mostly with comprehensive integrated displays (4 of 6 studies) and multipatient dashboards (5 of 7 studies). Only 1 of 5 randomized controlled trials had a positive effect in the primary outcome. Conclusion We found weak evidence suggesting comprehensive integrated displays improve provider efficiency and process outcomes, and multipatient dashboards improve compliance with care protocols and patient outcomes. Randomized controlled trials of physiologic and laboratory monitoring displays did not show improvement in primary outcomes, despite positive results in simulated settings. Important research translation gaps from laboratory to actual critical care settings exist.

Artificial intelligence is one of the fastest-developing areas of science that covers a remarkably wide range of problems to be solved. It has found practical application in many areas of human activity, also in medicine. One of the directions of cooperation between computer science and medicine is to assist in diagnosing and proposing treatment methods with the use of IT tools. This study is the result of collaboration with the Children’s Memorial Health Institute in Warsaw, from where a database containing information about patients suffering from Bruton’s disease was made available. This is a rare disorder, difficult to detect in the first months of life. It is estimated that one in 70,000 to 90,000 children will develop Bruton’s disease. But even these few cases need detailed attention from doctors. Based on the data contained in the database, data mining was performed. During this process, knowledge was discovered that was presented in a way that is understandable to the user, in the form of decision trees. The best models obtained were used for the implementation of expert systems. Based on the data introduced by the user, the system conducts expertise and determines the severity of the course of the disease or the severity of the mutation. The CLIPS language was used for developing the expert system. Then, using this language, software was developed producing six expert systems. In the next step, experimental verification was performed, which confirmed the correctness of the developed systems.

Similar to global trends, the incidence rate of tuberculosis (TB) in China declined from 2000 to 2018. In this study, we aimed to evaluate TB trends in northern Guizhou Province and identify risk factors associated with rifampicin-resistant (RR) and concurrent extrapulmonary TB (EPTB). We analyzed data of TB patients hospitalized in Affiliated Hospital of Zunyi Medical University from 2011 to 2018, and assessed correlations between demographic characteristics of patients and RR-TB as well as concurrent EPTB. Our results showed that numbers of new, retreated, RR-TB and concurrent EPTB cases increased gradually from 2011 to 2018. Retreated patients had the highest odds of RR-TB but a lower likelihood of concurrent EPTB compared to new patients. Patients between 21 and 40 years of age had a higher likelihood of RR-TB compared to those 20 years and younger. Female patients and patients from Bijie city as well as the Miao ethnic minority had higher odds of concurrent EPTB. In summary, our data demonstrate upward trends in new, rifampicin-resistant and concurrent extrapulmonary TB cases in northern Guizhou Province of China, which should not be overlooked especially during and post the COVID-19 pandemic because TB is a greater long-term global health threat than COVID-19.

Drug-excipient compatibility study is the essential basis for excipient selection at the pre-formulation stage. According to the pharmaceutical Quality by Design (QbD) principles, a comprehensive understanding of the ingredients' physicochemical properties and a theoretical evaluation of the interaction risk between the drugs and excipients are required for conducting rational compatibility experimental design. Currently, there is an urgent need to establish an artificial intelligence system for researchers to easily get through the problem because it is very inconvenient and hard to utilize those drug-excipient incompatibility data scattered in scientific literature. Here, we designed a knowledge-driven expert system named PharmDE for drug-excipient incompatibility risk evaluation. PharmDE firstly developed an information-rich database to store incompatibility data, covering 532 data items from 228 selected articles. Then, 60 drug-excipient interaction rules were created based on our knowledge and formulation research experiences. Finally, the expert system was developed by organically integrating the database searching and rule-based incompatibility risk prediction, which resulted in four main functionalities: basic search of incompatibility database, data matching by similarity search, drug incompatibility risk evaluation, and formulation incompatibility risk evaluation. PharmDE is expected to be a useful tool for drug-excipient compatibility study and accelerate drug formulation design. It is now freely available at https://pharmde.computpharm.org.

As far back as the industrial revolution, significant development in technical innovation has succeeded in transforming numerous manual tasks and processes that had been in existence for decades where humans had reached the limits of physical capacity. Artificial Intelligence (AI) offers this same transformative potential for the augmentation and potential replacement of human tasks and activities within a wide range of industrial, intellectual and social applications. The pace of change for this new AI technological age is staggering, with new breakthroughs in algorithmic machine learning and autonomous decision-making, engendering new opportunities for continued innovation. The impact of AI could be significant, with industries ranging from: finance, healthcare, manufacturing, retail, supply chain, logistics and utilities, all potentially disrupted by the onset of AI technologies. The study brings together the collective insight from a number of leading expert contributors to highlight the significant opportunities, realistic assessment of impact, challenges and potential research agenda posed by the rapid emergence of AI within a number of domains: business and management, government, public sector, and science and technology. This research offers significant and timely insight to AI technology and its impact on the future of industry and society in general, whilst recognising the societal and industrial influence on pace and direction of AI development.

Individually tailoring education over time may help more patients, especially racial/ethnic minorities, get waitlisted and pursue deceased and living donor kidney transplant (DDKT and LDKT, respectively). We enrolled 802 patients pursuing transplant evaluation at the University of California, Los Angeles Transplant Program into a randomized education trial. We compared the effectiveness of Your Path to Transplant (YPT), an individually tailored coaching and education program delivered at 4 time points, with standard of care (SOC) education on improving readiness to pursue DDKT and LDKT, transplant knowledge, taking 15 small transplant-related actions, and pursuing transplant (waitlisting or LDKT rates) over 8months. Survey outcomes were collected prior to evaluation and at 4 and 8 months. Time to waitlisting or LDKT was assessed with at least 18months of follow-up. At 8months, compared to SOC, the YPT group demonstrated increased LDKT readiness (47% vs 33%, P=.003) and transplant knowledge (effect size [ES]=0.41, P&nbsp

The novel 2019 Coronavirus (COVID-19) infection has spread worldwide and is currently a major healthcare challenge around the world. Chest computed tomography (CT) and X-ray images have been well recognized to be two effective techniques for clinical COVID-19 disease diagnoses. Due to faster imaging time and considerably lower cost than CT, detecting COVID-19 in chest X-ray (CXR) images is preferred for efficient diagnosis, assessment, and treatment. However, considering the similarity between COVID-19 and pneumonia, CXR samples with deep features distributed near category boundaries are easily misclassified by the hyperplanes learned from limited training data. Moreover, most existing approaches for COVID-19 detection focus on the accuracy of prediction and overlook uncertainty estimation, which is particularly important when dealing with noisy datasets. To alleviate these concerns, we propose a novel deep network named RCoNet $^{k}_{s}$ for robust COVID-19 detection which employs Deformable Mutual Information Maximization (DeIM), Mixed High-order Moment Feature (MHMF), and Multiexpert Uncertainty-aware Learning (MUL). With DeIM, the mutual information (MI) between input data and the corresponding latent representations can be well estimated and maximized to capture compact and disentangled representational characteristics. Meanwhile, MHMF can fully explore the benefits of using high-order statistics and extract discriminative features of complex distributions in medical imaging. Finally, MUL creates multiple parallel dropout networks for each CXR image to evaluate uncertainty and thus prevent performance degradation caused by the noise in the data. The experimental results show that RCoNet $^{k}_{s}$ achieves the state-of-the-art performance on an open-source COVIDx dataset of 15 134 original CXR images across several metrics. Crucially, our method is shown to be more effective than existing methods with the presence of noise in the data.

Artificial Intelligence (AI) as a technology has existed for less than a century. In spite of this, it has managed to achieve great strides. The rapid progress made in this field has aroused the curiosity of many technologists around the globe and many companies across various domains are curious to explore its potential. For a field that has achieved so much in such a short duration, it is imperative that people who aim to work in Artificial Intelligence, study its origins, recent developments, and future possibilities of expansion to gain a better insight into the field. This paper encapsulates the notable progress made in Artificial Intelligence starting from its conceptualization to its current state and future possibilities, in various fields. It covers concepts like a Turing machine, Turing test, historical developments in Artificial Intelligence, expert systems, big data, robotics, current developments in Artificial Intelligence across various fields, and future possibilities of exploration.

To properly and quickly evaluate an object&rsquo, s shape, in a manner that is suitable for real-time applications, a set of parameters has been created and the shape parametric description (SPD) has been elaborated. This solution is focused on the classification of amber gemstones according to shape. To improve the results obtained by SPD, the most popular machine learning classification algorithms were applied and tested. The proposed method (i.e., SPD) achieved the fastest classification, requiring the least computational resources, while providing an accuracy of approximately 80%. The best results were achieved when the SPD parameters were used in a feedforward neural network (FFNN), and an accuracy of 91.5% was obtained, while the time required for the computations remained in a range that is acceptable for real-time applications.

Background Due to an aging population, the demand for many services is exceeding the capacity of the clinical workforce. As a result, staff are facing a crisis of burnout from being pressured to deliver high-volume workloads, driving increasing costs for providers. Artificial intelligence (AI), in the form of conversational agents, presents a possible opportunity to enable efficiency in the delivery of care. Objective This study aims to evaluate the effectiveness, usability, and acceptability of Dora agent: Ufonia’s autonomous voice conversational agent, an AI-enabled autonomous telemedicine call for the detection of postoperative cataract surgery patients who require further assessment. The objectives of this study are to establish Dora’s efficacy in comparison with an expert clinician, determine baseline sensitivity and specificity for the detection of true complications, evaluate patient acceptability, collect evidence for cost-effectiveness, and capture data to support further development and evaluation. Methods Using an implementation science construct, the interdisciplinary study will be a mixed methods phase 1 pilot establishing interobserver reliability of the system, usability, and acceptability. This will be done using the following scales and frameworks: the system usability scale; assessment of Health Information Technology Interventions in Evidence-Based Medicine Evaluation Framework; the telehealth usability questionnaire; and the Non-Adoption, Abandonment, and Challenges to the Scale-up, Spread and Suitability framework. Results The evaluation is expected to show that conversational technology can be used to conduct an accurate assessment and that it is acceptable to different populations with different backgrounds. In addition, the results will demonstrate how successfully the system can be delivered in organizations with different clinical pathways and how it can be integrated with their existing platforms. Conclusions The project’s key contributions will be evidence of the effectiveness of AI voice conversational agents and their associated usability and acceptability. International Registered Report Identifier (IRRID) PRR1-10.2196/27227

The proposed method introduces algorithms for the preprocessing of normal, COVID-19, and pneumonia X-ray lung images which promote the accuracy of classification when compared with raw (unprocessed) X-ray lung images. Preprocessing of an image improves the quality of an image increasing the intersection over union scores in segmentation of lungs from the X-ray images. The authors have implemented an efficient preprocessing and classification technique for respiratory disease detection. In this proposed method, the histogram of oriented gradients (HOG) algorithm, Haar transform (Haar), and local binary pattern (LBP) algorithm were applied on lung X-ray images to extract the best features and segment the left lung and right lung. The segmentation of lungs from the X-ray can improve the accuracy of results in COVID-19 detection algorithms or any machine/deep learning techniques. The segmented lungs are validated over intersection over union scores to compare the algorithms. The preprocessed X-ray image results in better accuracy in classification for all three classes (normal/COVID-19/pneumonia) than unprocessed raw images. VGGNet, AlexNet, Resnet, and the proposed deep neural network were implemented for the classification of respiratory diseases. Among these architectures, the proposed deep neural network outperformed the other models with better classification accuracy.

The importance and benefits of healthcare mobile applications is increasing rapidly, especially when such applications are connected to the internet of things (IoT). This paper describes a smart knowledge-based system (KBS) that helps patients showing symptoms of Influenza verify being infected with Coronavirus, commonly known as COVID-19. In addition to the systems’ diagnostic functionality, it helps these patients get medical assistance fast by notifying medical authorities using the IoT. This system displays patient’s location, phone number, date and time of examination. During the applications’ development, the developers used Twilio, short message service (SMS), WhatsApp, and Google map applications.