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 102

Some of the metrics are blocked by your
consent settings
Privacy preserving synchronization of directed dynamical networks with periodic data-sampling
(Elsevier BV, 2025-01)
Jia, Qiang
;
Yao, Xinyi
;
Data privacy has become a key issue in networked systems, but few effort was devoted to privacy preservation in synchronization of nonlinear dynamical networks when data sampling is involved. This work focuses on the privacy preserving synchronization in a type of nonlinear dynamical network with sampled data. In order to preserve their private initial states, the nodes conceal the sampled data via certain deterministic perturbation, and exchange the masked data with their neighbors via the communication network. A novel privacy-preserving protocols with sampled data is developed, which differs from existing designs with continuous data, and a commonly used restriction on the nodes’ neighbor sets is unnecessary herein. By establishing a new Halanay-type inequality with decaying perturbation, some sufficient criteria are derived to guarantee synchronization without disclosing the nodes’ privacy, revealing how the decaying rate of the masking functions, the topology and the sampling period influence synchronization. Furthermore, in order to reduce the control update, the analogue of the above design with event-trigger is also given, leading to another useful condition for privacy preserving synchronization. Some numerical examples are finally given to validate the theoretical results and demonstrate the effectiveness of the proposed designs.
Some of the metrics are blocked by your
consent settings
A Comparison of GEC Tools for Grammatical Error Correction in English
(IEEE, 2025-06-02)
Virtanen, Johanna
;
Using the Building Educational Application (BEA) benchmark11https://codalab.lisn.upsaclay.fr/competitions/4057, this study compares the capabilities of Google Gemini22https://gemini.google.com/, ChatGPT33https://openai.com/chatgpt, DeepSeek44https://www.deepseek.com/en, and the builtin grammar checkers in Google Docs and Microsoft Word for grammatical error correction (GEC). These tools correct a variety of errors, some of which overlap. Based on the BEA benchmark evaluation results, Google Gemini and the Google Docs grammar checker achieve the best F0.5 scores of 60.2 and 65.86, respectively. Google Docs grammar checker is easy to use and, according to this evaluation, performs well, thus proving to be a viable option for GEC. However, standard grammar checkers are not typically designed for rewriting text to the same extent as GenAI tools; hence, it may be advisable especially for non-native speakers to combine traditional and GenAI grammar correction for the best possible results. However, it is necessary to check the grammatical corrections of LLMs, since generative AI tools suffer from hallucinations, which refers to their tendency to generate information that can be factually incorrect [1].
Some of the metrics are blocked by your
consent settings
Improving Atrial Fibrillation Detection with Machine Learning Models
(IEEE, 2025-06-02)
Tudjarski, Stojancho
;
;
Madevska Bogdanova, Ana
;
Stankovski, Aleksandar
In this paper, we focus on feature engineering to detect Atrial Fibrillation by determining whether the heart rhythm has irregularities without patterns. We experiment with a broad spectrum of features derived from the duration of heartbeat-to-heartbeat intervals in the benchmark electrocardiogram databases MIT-BIH Arrhythmia Database, MIT-BIH Atrial Fibrillation Database, and Long Term AF Database. The experiments included position-based features, fluctuation indices, standard deviations, mean average values, Shannon entropy, and statistical measures, such as compressed time series data length. The research questions are to detect the most influential features that result in the best-performing model and the impact of the training dataset. Our approach is to evaluate the model on a completely different dataset from the one it was trained on. Achieved F1 scores vary between 62,32% and 85.08%. The results prove that positional features increase the model's performance.
Some of the metrics are blocked by your
consent settings
Scalability Evaluation of HPC Multi-GPU Training for ECG-based LLMs
(IEEE, 2025-06-02)
;
Petrovski, Nikola
;
Training large language models requires extensive processing, made possible by many high-performance computing resources. This study compares multi-node and multi-GPU environments for training large language models using electrocardiogram data. It provides a detailed mapping of current frameworks for distributed deep learning in multi-node and multi-GPU settings, including Horovod from Uber, DeepSpeed from Microsoft, and the built-in distributed capabilities of PyTorch and TensorFlow. We compare various multi-GPU setups for different dataset configurations, utilizing multiple HPC nodes independently and focusing on scalability, speedup, efficiency, and overhead. The analysis leverages HPC infrastructure with SLURM, Apptainer (Singularity) containers, CUDA, PyTorch, and shell scripts to support training workflows and automation. We achieved a sub-linear speedup when scaling the number of GPUs, with values of 1.6x for two GPUs and 1.9x for four GPUs.
Some of the metrics are blocked by your
consent settings
Efficient Time-Series Heart Rate Variability Metrics in C++
(IEEE, 2025-06-02)
Shekerov, A.
;
Angjelevska, A.
;
Time-Series Heart Rate Variability is an essential set of metrics in modern medicine and cardiology. Algorithms for HRV calculations exhibit quadratic time complexity in the worst cases, as metrics are calculated at many pre-defined time intervals with different widths and offsets, especially for patients with potentially week-long recordings. In this paper, we aim to develop an efficient software solution for electrocardiograms from wearable sensors, which requires careful preprocessing and filtering steps to eliminate noise and erroneous values, detect heartbeats, and classify arrhythmia. This paper describes the utilized method and presents an efficient C++ implementation. We evaluate efficiency by comparing the new solution with the existing Python implementation and conclude that the new implementation performs exceptionally better, with a median speedup of 81.4 for single-day patient recordings and an average speedup of 11.9 for multi-day patient recordings. The results are presented for an extensive set of preprocessed patient databases, discussing the benefits and drawbacks.
Some of the metrics are blocked by your
consent settings
Benchmarking Functions Versus Containers for Compute-Intensive Workloads
(IEEE, 2025-11-25)
Krajchevska, Evgenija
;
Platform choices for compute-intensive jobs are often made without a disciplined, workload-specific comparison. This paper examines two dominant execution models, Functions-as-a-Service and Containers-as-a-Service, using a simple CPUbound kernel. We implement functionally equivalent services on both models, vary input sizes and resource allocations, and exercise sequential and bursty request patterns at comparable resource tiers to characterize performance, cost trade-offs, and scaling behavior. The study provides a controlled, reproducible protocol for comparing the models on compute-bound tasks and reports how memory and concurrency influence latency. The methodology is extensible, meaning the kernel can be swapped for other algorithms while preserving the measurement pipeline.
Some of the metrics are blocked by your
consent settings
Scalability Analysis of SaaS Infomatrix ERP
(IEEE, 2025-11-25)
Ordanoski, Petar
;
his paper presents the implementation and performance evaluation of scalability in the Infomatrix ERP system, a true cloud-based SaaS developed by Infoproject LLC. Through architectural analysis and performance testing, the system demonstrates horizontal scalability. A shared database supports up to 5,000 concurrent users while maintaining 3 -second response times. Results confirm efficient scaling, cost reduction, and the effectiveness of true SaaS architecture in enterprise environments.
Some of the metrics are blocked by your
consent settings
Overview of Interpretable and Explainable Artificial Intelligence for Atrial Fibrillation
(IEEE, 2025-11-25)
Tudjarski, Stojancho
;
Angjelevska, Ana
;
;
Madevska Bogdanova, Ana
Accurate and interpretable detection of irregular work of the heart, such as atrial fibrillation, from electrocardiogram (ECG) signals is crucial for timely diagnosis and effective patient management. While machine learning (ML) models, particularly deep learning architectures, have achieved state-of-the-art performance in ECG arrhythmia classification, their black-box nature limits clinical adoption. This paper explores explainable artificial intelligence (XAI) techniques applicable to ML models trained on ECG data, highlighting both global and local interpretability approaches. We provide an overview of posthoc methods, including SHAP, LIME, PFI, and LIG, among others, treating various types of ECG recordings. As a practical case study, we present our findings analyzing the results of PFI and LIG methods applied to a transformer-based model fine-tuned for atrial fibrillation detection and explain its decision process. Our findings underscore the value of integrating XAI into ECG analysis pipelines to enhance transparency, foster clinician trust, and support more informed decision-making in cardiovascular diagnostics.
Some of the metrics are blocked by your
consent settings
Prompt-to-Pill: Multi-Agent Drug Discovery and Clinical Simulation Pipeline
(Oxford University Press (OUP), 2025-12-23)
Vichentijevikj, Ivana
;
;
;
Yamanishi, Yoshihiro
Summary This study presents a proof-of-concept, comprehensive, modular framework for AI-driven drug discovery (DD) and clinical trial simulation, spanning from target identification to virtual patient recruitment. Synthesized from a systematic analysis of 51 large language model (LLM)-based systems, the proposed Prompt-to-Pill architecture and corresponding implementation leverages a multi-agent system (MAS) divided into DD, preclinical and clinical phases, coordinated by a central Orchestrator. Each phase comprises specialized LLM for molecular generation, toxicity screening, docking, trial design, and patient matching. To demonstrate the full pipeline in practice, the well-characterized target Dipeptidyl Peptidase 4 (DPP4) was selected as a representative use case. The process begins with generative molecule creation and proceeds through ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) evaluation, structure-based docking, and lead optimization. Clinical-phase agents then simulate trial generation, patient eligibility screening using electronic health records (EHRs), and predict trial outcomes. By tightly integrating generative, predictive, and retrieval-based LLM components, this architecture bridges drug discovery and preclinical phase with virtual clinical development, offering a demonstration of how LLM-based agents can operationalize the drug development workflow in silico. Availability and implementation The implementation and code are available at: https://github.com/ChatMED/Prompt-to-Pill.
Some of the metrics are blocked by your
consent settings
Evaluating a Nationally Localized AI Chatbot for Personalized Primary Care Guidance: Insights from the HomeDOCtor Deployment in Slovenia
(MDPI AG, 2025-07-29)
Gams, Matjaž
;
Horvat, Tadej
;
Kolar, Žiga
;
Kocuvan, Primož
;
Background/Objectives: The demand for accessible and reliable digital health services has increased significantly in recent years, particularly in regions facing physician shortages. HomeDOCtor, a conversational AI platform developed in Slovenia, addresses this need with a nationally adapted architecture that combines retrieval-augmented generation (RAG) and a Redis-based vector database of curated medical guidelines. The objective of this study was to assess the performance and impact of HomeDOCtor in providing AI-powered healthcare assistance. Methods: HomeDOCtor is designed for human-centered communication and clinical relevance, supporting multilingual and multimedia citizen inputs while being available 24/7. It was tested using a set of 100 international clinical vignettes and 150 internal medicine exam questions from the University of Ljubljana to validate its clinical performance. Results: During its six-month nationwide deployment, HomeDOCtor received overwhelmingly positive user feedback with minimal criticism, and exceeded initial expectations, especially in light of widespread media narratives warning about the risks of AI. HomeDOCtor autonomously delivered localized, evidence-based guidance, including self-care instructions and referral suggestions, with average response times under three seconds. On international benchmarks, the system achieved ≥95% Top-1 diagnostic accuracy, comparable to leading medical AI platforms, and significantly outperformed stand-alone ChatGPT-4o in the national context (90.7% vs. 80.7%, p = 0.0135). Conclusions: Practically, HomeDOCtor eases the burden on healthcare professionals by providing citizens with 24/7 autonomous, personalized triage and self-care guidance for less complex medical issues, ensuring that these cases are self-managed efficiently. The system also identifies more serious cases that might otherwise be neglected, directing them to professionals for appropriate care. Theoretically, HomeDOCtor demonstrates that domain-specific, nationally adapted large language models can outperform general-purpose models. Methodologically, it offers a framework for integrating GDPR-compliant AI solutions in healthcare. These findings emphasize the value of localization in conversational AI and telemedicine solutions across diverse national contexts.

Faculty of Computer Science and Engineering

Browse

Filters

Settings

Sort By

Results per page

Search Results