Faculty of Computer Science and Engineering

Permanent URI for this communityhttps://repository.ukim.mk/handle/20.500.12188/5

The Faculty of Computer Science and Engineering (FCSE) within UKIM is the largest and most prestigious faculty in the field of computer science and technologies in Macedonia, and among the largest faculties in that field in the region. The FCSE teaching staff consists of 50 professors and 30 associates. These include many “best in field” personnel, such as the most referenced scientists in Macedonia and the most influential professors in the ICT industry in the Republic of Macedonia.

Browse

Search Results

Now showing 1 - 10 of 29

Some of the metrics are blocked by your
consent settings
Deep Learning Image Embeddings for Ventricular Beat Classification
(IEEE, 2025-11-25)
Tudjarski, Stojancho
;
Petrovski, Nikola
;
;
Ventricular arrhythmias pose significant risks to cardiovascular health, necessitating accurate detection from electrocardiogram signals. This study investigates the effectiveness of vector embeddings derived from various convolutional neural networks and contrastive language-image pretraining-based models for classifying ventricular heartbeats presented as images. A comparative analysis revealed that the combination of the ResNet50 model and support vector machines achieved the top F1 score of 79.30%. The results provide insights into how deep neural network models can produce effective vector embeddings of heartbeats to train heartbeat classification models.
Some of the metrics are blocked by your
consent settings
Benchmarking Parallel Electrocardiogram Compression Based on Successive Differences
(IEEE, 2024-11-26)
Shekerov, A
;
;
Gusev, Marjan
We focus on parallelization methods for an electrocardiogram data compression algorithm based on successive differences to gain insights into the advantages and disadvantages of parallel implementations. The experimental methodology exposes a comprehensive and systematic benchmarking process with varying input file sizes, hosting machine characteristics, and two popular parallelization approaches: OpenMP and MPI. We check the research hypothesis to see if parallelizing the compression algorithm can reduce the runtime while keeping the original algorithm’s compression results. Our analysis and discussion show that OpenMP outperforms MPI. An OpenMP implementation with 12 threads on a processor with six cores achieves the highest average speedup of 7 versus a single-thread implementation. Performance gains depend heavily on the utilized hardware and the degree of parallelism.
Some of the metrics are blocked by your
consent settings
Atrial Fibrillation Detection using the Stars 2D Convolutional Neural Network
(IEEE, 2024-05-20)
Petrovski, N
;
Gusev, Marjan
;
Atrial Fibrillation is one of the riskiest potentials of heart failure in the recent escalating prevalence of cardiovascular diseases. Detecting such an irregularly irregular heart rhythm is a paramount challenge in modern biomedical computation. This research contributes to its resolution by introducing an advanced 2D Convolutional Neural Network model trained on the Poincare plot of the differences between consecutive instantaneous heart rates instead of beat-to-beat time intervals. Furthermore, we used the Gaussian Blur technique to enhance the model’s capacity to generalize and augment its accuracy by forming a star-like plot inspired by Van Gogh’s famous Starry Nights canvas. Model development uses different benchmark datasets to train and test the model. Applying the standard Machine Learning window label comparison method, our model achieves an impressive F1 Score of 96.34% and a remarkable F1 Score of 94.85% when evaluated by the duration-based assessment method.
Some of the metrics are blocked by your
consent settings
Container vs Function as a Service: Impact on Cloud Deployment for Real-World Applications
(IEEE, 2024-05-20)
Petrovski, Toshe
;
Gusev, Marjan
This paper compares Container as a Service (CaaS) and Function as a Service (FaaS) as different cloud deployment strategies for a multitenant ride-sharing web application Trek. The application is highly network-intensive, relying extensively on third-party APIs for collecting location data about cities and driving routes. Additionally, it is CPU-intensive due to real-time data processing of driver locations while the vehicle moves towards the destination city. This study evaluates both approaches concerning variable workloads and their impact on system responsiveness, resource utilization, and performance during peak and offpeak periods. It closely examines crucial performance metrics such as latency and throughput to analyze the implications of both approaches for end-user experience and overall operational efficiency. Additionally, our study analyzes several critical factors like scalability and deployment complexity. We conducted experiments on AWS ECS (CaaS) and AWS Lambda (FaaS) to gain actionable insights into the trade-offs, benefits, and limitations of each platform, enabling informed decision-making for cloud-based application deployment.
Some of the metrics are blocked by your
consent settings
Leveraging Dataframe-Based Operations for Calculation of Heart Rate Variability
(IEEE, 2024-05-20)
Temelkov, G
;
Gusev, Marjan
This paper introduces a novel Heart Rate Variability (HRV) calculation strategy, diverging from traditional divide-and-conquer methodologies to dataframes utilizing the Polars library as an advanced data manipulation tool. We demonstrate significant improvements in computational performances, where the dataframe approach outperforms the iterative approach with speedup factors beyond 98 times for short-term HRV calculations and substantial reductions in processing times across various test cases. These enhancements underscore the potential of tailored dataframe manipulations in enhancing performance and adapting to complex data analysis challenges in HRV assessments.
Some of the metrics are blocked by your
consent settings
Speeding up Dense Optical Flow Estimation with CUDA
(IEEE, 2024-11-26)
Stameski, Kristijan
;
Gusev, Marjan
Optical flow is the perceived movement of a pixel within the video. It is inherently helpful for motion tracking and may also be used as a preprocessing for other computer vision algorithms or machine learning. Algorithmic optical flow estimation is slow and resource-intensive, but real-time performance can be achieved by using GPUs. This paper discusses implementing and optimizing a pyramidal Lucas-Kanade optical flow algorithm in CUDA.
Some of the metrics are blocked by your
consent settings
Evaluating Azure Container, Backend, and Function as a Service
(IEEE, 2024-11-26)
Ristovski, Daniel
;
Gusev, Marjan
In this research paper, we investigate the performance of Container, Backend, and Function as a Service as various Microsoft cloud service models for specific computational and memory-bound use cases. The research method involves practical experiments to present insights into functionalities, advantages, and limitations. Our research shows that AKS performs the best compared to Web Apps and Azure Functions, reaching the highest throughput for a quick sort application.
Some of the metrics are blocked by your
consent settings
Classification of Hemoglobin A1c from Long and Extra-long Term Heart Rate Variability
(IEEE, 2024-05-20)
Angjelevska, A
;
Gusev, Marjan
;
Gjorgjieva, S
This study classifies Hemoglobin A1c (HbA1c) concentration from long- and extra-long-term heart rate variability (HRV) measurements and various machine learning (ML) models utilizing different datasets. Key metrics include SDNN, RMSSD(NN), NN50, and PNN50 under detailed window-oriented calculations, employing Average, Standard Deviation, and Concatenated methods for feature extraction. A comprehensive pre-processing phase within the ML pipeline ensures analytical robustness. The study systematically conducts patient-wise data splits and evaluates classification performance across various ML models, contributing to a thorough analysis. Evaluation metrics such as sensitivity, specificity, precision, and different F1 scores guide this research in advancing the understanding of HbA1c regulation. The study aspires to establish optimal ML model training and evaluation configurations, contributing to the broader discourse on HbA1c classification. The best-performing model reaches an F1 Score of 93.20%, and F1M of 92.76%, demonstrating its robustness and effectiveness over baseline models.
Some of the metrics are blocked by your
consent settings
Parallelism in Signature Based Virus Scanning with CUDA
(Springer International Publishing, 2019)
Dimitrioski, Andrej
;
Gusev, Marjan
;
Some of the metrics are blocked by your
consent settings
Calculating average shortest path length using Compute Unified Design Architecture (CUDA)
(IEEE, 2019-05)
Petrushevski, Stefan
;
Gusev, Marjan
;

Faculty of Computer Science and Engineering

Browse

Filters

Settings

Sort By

Results per page

Search Results