Periodicals of Engineering and Natural Sciences (PEN - International University of Sarajevo)
Not a member yet
1290 research outputs found
Sort by
Enhanced feature selection algorithm for pneumonia detection
Get PDFPneumonia is a type of lung disease that can be detected using X-ray images. The analysis of chest X-ray images is an active research area in medical image analysis and computer-aided radiology. This research aims to improve the accuracy and efficiency of radiologists\u27 work by providing a technique for identifying and categorizing diseases. More attention should be given to applying machine learning approaches to develop a robust chest X-ray image classification method. The typical method for detecting Pneumonia is through chest X-ray images but analyzing these images can be complex and requires the expertise of a radiographer. This paper demonstrates the feasibility of detecting the disease using chest X-ray images as datasets and a Support Vector Machine combined with a Naive Bayesian classifier, with PCA and GA as feature selection methods. The selected features are essential for training many classifiers. The proposed system achieved an accuracy of 92.26%, using 91% of the principal component. The study\u27s result suggests that using PCA and GA for feature selection in chest X-ray image classification can achieve a good accuracy of 97.44%. Further research is needed to explore the use of other data mining models and care components to improve the accuracy and effectiveness of the system
Design of automatic speech recognition in noisy environments enhancement and modification
Get PDFRecurrent neural networks (RNN) and feed-forward multi-layer perceptron’s have been proposed for determining the absence and presence of speech in continuous voice signals when there is a variety of background noise levels present. The Aurora2 and Aurora3 were used to conduct detailed performance evaluations on vocal activity detection. When a Recurrent neural network feeds on automatic speech recognition particular features and acoustic features, the best outcomes can be achieved, according to this study. Aurora2 and the French, Romanian and Norway portions of the Aurora3 corpus is also proposed for detailed studies of ASR. When noise presence probability is utilized to change for encoding speech, phone subsequent probabilities are employed; the WER is reduced by 10.3 percent
Digital leadership and organizational capabilities in manufacturing industry: A study in Malaysian context
Get PDFThe research is conducted to study the outcome of digital leadership on dynamic capabilities, innovation capabilities, and alliances capabilities in manufacturing industries within the Malaysian context. Today, developing corporations and industries, at the least, require a virtual transformation to have greater organizational abilities in shaping and growing their new and present commercial enterprise to healthy the brand-new generation paradigm. A cross-sectional quantitative method has been used in this study with a sample of 132 respondents with different industry back grounds. These respondents are organizations, which are based on the nature of a business role like authorized representative, distributor, importer, manufacturer, combination authorized representative, distributor and importer, and combination distributor and importer that located in Selangor, Malaysia. The research used the SMART PLS software to analyze and interpret the results. There main hypotheses are proposed and tested. The results showed that digital leadership positively affects dynamic capabilities, innovation capabilities, and alliances capabilities
A novel secure artificial bee colony with advanced encryption standard technique for biomedical signal processing
Get PDFOver the years, the privacy of a biomedical signal processing is protected using the encryption techniques design and meta-heuristic algorithms which are significant domain and it will be more significant shortly. Present biomedical signal processing research contained security because of their critical role in any developing technology that contains applications of cryptography and health deployment. Furthermore, implementing public-key cryptography in biomedical signal processing sequence testing equipment needs a high level of skill. Whatever key is being broken with enough computing capabilities using brute-force attack. As a result, developing a biomedical signal processing cryptography model is critical for improving the connection between existing and emerging technology. Furthermore, public-key cryptography implementation for meta-heuristic-based bio medical signal processing sequence test equipment necessitates a high level of skill. The suggested novel technique can be used to develop a secure algorithm of artificial bee colony, which depend on the advanced encryption standard (AES). AES can be used to reduce the encryption time and to increase the protection capacity for health systems. The novel secure can protect the biomedical signal processing against plain text attacks
The shading quality of tree species and their influence on the microclimate of the immediate surroundings in urban environments
Get PDFThe purpose of this paper was to assess the quality and impact of the shading offered by two types of shading trees Samanea saman (rain tree) and Searsia Pendulina (Wit Karee) of the environmental variables Dry Bulb Temperature (TBS) and Radiant Thermal Load (CTR) by selected species trees and increasing their relative air humidity (RH) on the climatic conditions of the University of Baghdad, and Baghdad Park in An Nasiriyah, Iraq. Data related to the variables described were collected by means of two sets of thermometers placed: the one in tree shade (5 m) and the second in trunk (full sun) at 10 m, from June to August 2020, for each single tree, for 3 days each hour from 10:00 am to 14:00 pm. The calculation of the percentage of the Relatively Variated Values, at 5 m and 10 m, with respect to the values obtained at 5 m for quantification of shading contribution to the attenuation and augmentation of environmental variables took place. The data were analysed by testing the following hypotheses: I TBS shade attenuation > TBS attenuation at 10 m (ii) shade CTR attenuation > CTR attenuation at 10 m and (iii) shade RH increased > RH increase at 10 m. Increased shade of TBS at 10 m. The results show the good impact on TBS and CTR mitigation and RH increases. Searsia Pendulina was the most prominent species that exhibited a general TBS attenuation range from 5% to 10%, rain tree and Karee in the UR variable with increments over 30% and Karee in the CTR shadow rates with attenuations in the order of 15%. The comparison of results in this sector with the criterion of comfort has proved the impact on improving the microclimate of the local environment of the researched arboreal species
Analyzing and processing medical images with increased performance using fractal geometry
Get PDFThe research relied on the application of a series of steps to analyze medical images, and to basically achieve this goal, a set of techniques were made from both fractal engineering and tissue analysis by improving the studied image and then analyzing the studied image texture in the fractal dimension and propose a hybrid method for segmenting images of complex situations and structures based on the geometric patterns that are repeated and represented by the fractal filter (Hurst), which is one of the modern techniques used in the field of digital image processing. Using fractal methods, that is, a specific application through real fractal structures of medical images and measuring their fractal dimensions and in capturing the exact features based on the scale in dimensional fractions, where the accuracy rate reached )98%( in diagnosing pathological conditions with an error rate close to zero. Also, the coefficients of multiple fractals were calculated (α) ,with a threshold factor of (4.5), the texture is also classified based on the fractal algorithm and Gray-Level Co-Occurrence Matrices (GLCM) and according to the experimental results performed on the medical images, the classification method provides a classification rate of 95%. To increase the accuracy, the lacunarity was calculated in the healthy medical images by applying fractal theorem filters where the gap ratio was close to (1) in the lacunarity size. The results also showed that the decrease in the contrast of the image with the continuation of the smoothing process or the decrease in the intensity levels of the image causes a significant decrease in the contrast of the image, especially in the areas of the edges
A stand-alone hydrogen photovoltaic fuel cell hybrid system for efficient renewable energy generation
Get PDFToday the main concern for World is energy and power age. By and by, out of around 7 billion populaces, just 65-69% approaches power. Essentially to carry the populaces into the office of power access however much as could be expected inside the restricted assets, we have used the regular assets like sun oriented and wind to satisfy this assumption. Utilizing sun based and wind energy in relationship with the power gadgets, we can supply the power to the buyers inside their capacity and we will want to limit the power issue as could really be expected. Hydrogen Photovoltaic Fuel (HPF) cell is the mix of force gadgets which lessens the major sun-oriented emergency of expenses, where expenses are the enormous issue for non-industrial nations. Presently a-days, the coordinated circuits (IC) are entirely solid and modest, to the point that make the conveying and reversing or changing over components simplest than the massive and expensive instruments utilized in the traditional power supply framework. The examination expects that the lattice joining of the environmentally friendly power assets utilizing HPF inverter might cause a colossal comment in satisfying the absence of force use across the world. Solar energy is a rapidly growing resource, already providing 4.5% of electricity in the World and projected to supply up to 35% by 2050. On the other hand, the default model’s predictions were far from the actual metered HPF data. For renewability, the simulated renewable energy consumption with modified inputs is 3.9% below of actual metered renewable data while the default model’s prediction was more than 52% below actual renewable use. Using PV-HPF hybrid model indices to represent how well a simulated model describes the variability in the measured data; the modified model has achieved accurate renewability results; with a Solar of 10.99 % and Wind of 9.90%, while the hybrid model has a solar of 57.16% and a Wind of 57.20% in renewable energy comparison being performed in MATLAB. 
Enhancement of the efficiency of solar energy cells by selecting suitable places based on the simulation of PV System
Get PDFAt present, the increasing demand for electrical energy and the presence of renewable sources in various forms in the world and, particularly in Iraq, such as solar energy and wind energy, have become the focus of researchers\u27 attention. Huge efforts are focused on finding ways to use ecologically friendly energy to generate electricity and eliminate fossil fuels. In this study paper, we propose the use of a simulation program to discover the ideal location for a solar cell and the amount of time to be exposed to the sun\u27s rays, so that a home powered by solar energy can be built. Also, through this program, the losses were calculated that accompany the conversion of light energy into electrical energy to find the necessary solutions to make the solar cell work with high efficiency
Clinical decision making for prediction of otitis using machine learning approach
Get PDFThis study investigates the relationship between autoimmune disease otitis and gut microbial community abundance by using machine learning as an aid in the medical decision-making process. Stool samples of healthy and otitis diseased infants were obtained from the curatedMetagenomicData package. Class imbalance present in the dataset was handled by oversampling a minority class. Afterwards, we built several machine learning models (support vector machine, k-nearest neighbour, artificial neural networks, random forest and gradient boosting) to predict otitis from gut microbial samples. The best overall accuracy was obtained by the random forest classifier, 0.99, followed by support vector machine and gradient boosting algorithms, both achieving 0.96 overall accuracy. We also obtained the most informative predictors as potential microbial biomarkers for the otitis disease. The obtained results showed better accuracy in prediction of otitis from microbial metagenome than previously proposed methods found in literature
Blockchain-based student certificate management and system sharing using hyperledger fabric platform
Get PDFOne of the major capabilities of blockchain technology is the sharing of data in verifiable ways without losing control of information possession. Issuing and verifying student certifications for higher study applications or job recruitment require many steps that take days to complete and are considered time-consuming. Most universities around the world use centralized systems to control the entire procedure when a graduate applies for a job or postgraduate studies. Applying blockchain technology to certificate verification protocols through a comprehensive architecture provides authenticity and reduces time significantly. In this paper, a framework has been proposed to issue student certifications locally in addition to sharing them across the internet while maintaining control and ownership of the certifications. This framework leverages the advantages of blockchain technology to electronic certification sharing and verification. Applying the proposed blockchain-based certification system in universities will provide low latency for issuing, sharing, and verification of these certifications. The paper presents the proposed blockchain-based framework for e-certification sharing and an evaluation of the framework, which consists of measuring the average time to issue a certificate and transaction latency time