The building integrated photovoltaic (BIPV) systems are a popular option for integrating renewable energy sources in the power system, and for users to reduce energy bills. This paper analyzes the performance of inverters in BIPV systems with oversized PV configurations. Oversizing PV systems has become a common practice to optimize energy production, particularly in periods of low sunlight, but it raises concerns about efficiency, power quality, and potential economic implications. Performance analysis is performed on two inverters, one operating under an overloaded regime due to the oversized PV installation and another under normal conditions. Several performance metrics are compared, including efficiency, thermal behavior, THD, and economic factors. The results demonstrate that although oversizing can slightly increase the inverter’s temperature and affect power quality, the efficiency was better for the overloaded inverter, although the investment costs have increased. These results offer practical insights for designing PV systems, showing that oversizing can be beneficial if properly managed.

Early detection of atrial fibrillation plays a crucial role in the timely prevention, diagnosis, and treatment of cardiovascular diseases. This paper proposes two different network architectures for automated atrial fibrillation detection. In the first architecture, a 1D CNN is used as a feature extractor and classifier. In the second hybrid architecture, a 1D CNN is used only as a feature extractor from ECG time series signals that supply a KNN with the most relevant features for further classification. Experimental results showed that the hybrid architecture achieved remarkable results and outperformed a 1D CNN.

Ultrasound images are used in various branches of medicine to detect diseases. The process of obtaining this data is complex due to procedures and legal restrictions, leading to scarce datasets. Different data augmentation techniques can be employed to improve classification performance. This paper shows that augmenting the ultrasound breast cancer images dataset using generative adversarial networks (GANs) increased the classification accuracy compared to the original dataset and compared to the dataset augmented using standard techniques.

Generative AI approaches such as ChatGPT are very popular and can be used for multiple purposes. This paper explores the possibility of using ChatGPT-4o for analysing visual information about 2D objects on provided images and returning annotated image results to the user. The achieved results indicate that ChatGPT can be used for the analysis of visual data and detect approximate values of desired parameters, however its generative capabilities are lacking and often unusable.

Smart wearable devices often contain heart rate monitoring capabilities. This paper presents an experimental study that compares the accuracy of smart watches (Xiaomi Amazfit Bip 3 and GEEKIN X10) to microcontroller-based systems that use raw sensors (HW-827 and MAX30102). The achieved results indicate that the accuracy of raw sensors is lower compared to smart watches and that the level of inaccuracy depends on the level of physical activity of the test subjects.

Modern IoT devices used for remote health monitoring use basic parameters such as heart rate, skin temperature and oxygen saturation. Maximum heart rate is an important parameter used for calculating heart rate zones that is helpful in diagnosis and prevention of cardiovascular diseases. This paper presents an information system that contains an IoT subsystem for heart rate measurement, and a web-server subsystem for monitoring by doctors that includes heart rate zone monitoring.

This paper presents a method based on Artificial Neural Networks (ANN) for magnetic flux density harmonics estimation in the vicinity of overhead lines. The proposed method can be employed for the magnetic flux density estimation in the cases of excitation currents with a pure sinusoidal waveform, as well as for excitation currents with harmonically distorted waveforms. The method utilizes two ANNs that are trained in a such way that enables their application for overhead lines of arbitrary phase conductor configurations and arbitrary current harmonic spectrum. In this paper, the proposed method is applied for magnetic flux density harmonics estimation in the vicinity of a typical distribution overhead line. The proposed method is validated by comparison with Biot-Savart based method. The obtained results demonstrate not only the accuracy and effectiveness of the proposed method but also the importance of considering the magnetic flux density harmonics in the vicinity of power facilities.

Random sequential adsorption (RSA) is a broadly used model for irreversible deposition on substrates. Over the last decades, a huge number of works have been published concerning this topic. Here we give a brief review of the results for irreversible deposition on two-dimensional discrete substrates. Depositing objects are randomly and sequentially adsorbed onto the substrate, and they are not allowed to overlap, so the jamming coverage θjam is less than in close packing. The kinetics of the process is described by the time-dependence of the coverage fraction θ(t), and for the discrete substrates, this dependence was found to be of the form: θ(t)=θjam−Ae−t/σ. Another topic of interest is the percolation of the deposit that can occur at a certain coverage. The coverage of the surface is increased through the RSA process up to the percolation threshold when a cluster that extends through the whole system appears. A percolating cluster arises in the system when the opposite edges are connected via some path of nearest neighbor sites occupied by the particles. Studying percolation is of great interest due to its relevance to conductivity in composite materials, flow through porous media, polymerization, the properties of nanomaterials, etc.

Background Healthcare institutions throughout the world rely on medical devices to provide their services reliably and effectively. However, medical devices can, and do sometimes fail. These failures pose significant risk to patients. Objective One way to address these issues is through the use of artificial intelligence for the detection of medical device failure. This goal of this study was to develop automated systems utilising machine learning algorithms to predict patient monitor performance and potential failures based on data collected during regular safety and performance inspections. Methods The system developed in this study utilised machine learning techniques as its core. Throughout the study four algorithms were utilised. These algorithms include Decision Tree, Random Forest, Linear Regression and Support Vector Machines. Results Final results showed that Random Forest algorithms had the best performance on various metrics among the four developed models. It achieved accuracy of 94% and precision and recall of 70% and 93% respectively. Conclusion This study shows that use of systems like the one developed in this study have the potential to improve management and maintenance of medical devices.

Background Analysis of data from incident registries such as MAUDE has identified the need to improve surveillance and maintenance strategies for infusion pumps to enhance patient and healthcare staff safety. Objective The ultimate goal is to enhance infusion pump management strategies in healthcare facilities, thus transforming the current reactive approach to infusion pump management into a proactive and predictive one. Method: This study utilized real data collected from 2015 to 2021 through the inspection of infusion pumps in Bosnia and Herzegovina. Inspections were conducted by the national laboratory in accordance with the Legal Metrology Framework, accredited to ISO 17020 standard. Out of 988 samples, 790 were used for model training, while 198 samples were set aside for validation (20% of the dataset). Various machine learning algorithms for binary classification of samples (pass/fail status) were considered, including Logistic Regression, Decision Tree, Random Forest, Naive Bayes, and Support Vector Machine. These algorithms were chosen for their ability to handle large datasets and potential for high prediction accuracy. Results Through detailed analysis of the achieved results, it was found that all applied machine learning methods yielded satisfactory results, with accuracy ranging from 0.98% to 1.0%, precision from 0.99% to 1%, sensitivity from 0.98% to 1.0%, and specificity from 0.87% to 1.0%. However, Decision Tree and Random Forest methods proved to be the best, both due to their maximum achieved values of accuracy, precision, sensitivity, and specificity, and due to result interpretability. Conclusion It has been established that machine learning methods are capable of identifying potential issues before they become critical, thus playing a crucial role in predicting the performance of infusion pumps, potentially enhancing the safety, reliability, and efficiency of healthcare delivery. Further research is needed to explore the potential application of machine learning algorithms in various healthcare domains and to address practical issues related to the implementation of these algorithms in real clinical settings.

The project phase is where the life-cycle of a building starts. The best decisions are made during the design or pre-project step. In terms of both economic resources and time, changes to specific design decisions made at this step are inexpensive compared to subsequent steps of architectural planning, not to mention the course of the construction's operation itself. The choices made during the design phase determine to a large extent whether the architectural design decisions of a building are achieved, whether the building and site can be used appropriately, and whether the project is economically viable. With BIM, building spatial planning is possible. As a result, architects can evaluate the proposed structure, its impact on the ecology, and the ecology's impact on the structure more comprehensively and at an earlier stage. This research proposes an energy modeling approach for the BIM-based spatial planning phase of a construction. The proposed method will result in an energy model for specific sites and building resolutions when utilized to create a spatial modelling for a construction. The energy model can then be used for new architectural creations. In this study, 36 different alternative scenarios were designed in terms of the rate of construction height to construction spacing, orientation factor, and form factor. With the help of BIM and GBS softwares, the energy consumption values of the alternative scenarios in cooling and heating load conditions were compared, and the alternative scenario with the minimum energy consumption was tried to be determined with spatial planning parameters.Ključne riječi

Nanotechnology has seen significant growth in the past few decades, with the use of nanomaterials reaching a wide scale. Given that antimicrobial resistance is peaking, nanotechnology holds distinct potential in this area. This review discusses recent applications of metal and metal oxide nanoparticles as antibacterial, antifungal, and antiviral agents, particularly focusing on their topical applications and their role in chronic wound therapy. We explore their use in various forms, including coated, encapsulated, and incorporated in hydrogels or as complexes, proposing them as topical antimicrobials with promising properties. Some studies have shown that metal and metal oxide nanoparticles can exhibit cytotoxic and genotoxic effects, while others have found no such properties. These effects depend on factors such as nanoparticle size, shape, concentration, and other characteristics. It is essential to establish the dose or concentration associated with potential toxic effects and to investigate the severity of these effects to determine a threshold below which metal or metal oxide nanoparticles will not produce negative outcomes. Therefore, further research should focus on safety assessments, ensuring that metal and metal oxide nanoparticles can be safely used as therapeutics in biomedical sciences.