Prof. Lazim Abdullah
Universiti Malaysia Terengganu, Malaysia
Lazim Abdullah is a Professor of Computational Mathematics
at the Faculty of Computer Science and Mathematics,
Universiti Malaysia Terengganu. He received his Ph.D
(Information Technology) from the Universiti Malaysia
Terengganu, in 2004. His research and expertise focus on
fuzzy set theory of mathematics, decision making models,
applied statistics, and their applications to environment,
health sciences and technology management. His research
findings have been published in more than 395 publications
including refereed journals, conference proceedings,
chapters in book, monographs, and textbooks. He has been
ranked among the world’s top 2% scientists by Stanford
University in the field of artificial intelligence and image
processing since 2018. Prof Lazim is a member of the IEEE
Computational Intelligence Society, and a member of
International Society on Multiple Criteria Decision Making.
Speech Title: An Integrated Bipolar Fuzzy-DEMATEL for Elucidating Factors Influencing Customers Choice: A Case of Life Insurance Companies
Abstract: Multi-criteria decision-making (MCDM) methods have
gained substantial traction across various scientific
disciplines, with the Decision-Making Trial and Evaluation
Laboratory (DEMATEL) method being particularly prominent.
This study advances the DEMATEL framework by incorporating
bipolar fuzzy sets to better handle complex, uncertain
decision environments. The primary objectives are twofold:
(1) to propose an integrated Bipolar Fuzzy-DEMATEL model and
(2) to apply the model to identify key factors influencing
customer choice in life insurance companies. The model
introduces a novel linguistic scale for bipolar fuzzy sets,
allowing simultaneous evaluation of positive and negative
membership degrees across truth, falsity, and uncertainty
dimensions. A sensitivity analysis was also conducted to
assess the robustness of the findings. Results indicate that
the cause factors influencing customer choice include F1,
F2, F7, F8, and F9, while F3, F4, F5, F6, and F10 are
classified as effect factors. Among them, ‘F2|: Competitive
pricing and clear terms’ emerged as the most influential.
The sensitivity analysis confirmed the model’s robustness,
showing minimal impact of weight variations on factor
rankings. The stability of top-ranked factors under changing
conditions highlights the model’s reliability and its
practical relevance for strategic decision-making in the
insurance sector.
Assoc. Prof. Marko Đurasević
University of Zagreb, Croatia
Marko Đurasević is an Associate Professor at the Faculty of
Electrical Engineering and Computing (FER), University of
Zagreb. His research is centered on evolutionary computation,
particularly genetic programming and hyper-heuristics for
solving complex scheduling and optimization problems. He earned
his Ph.D. in Computer Science from FER in 2018, with a
dissertation focused on the automated design of dispatching
rules in unrelated machines environments.
Dr. Đurasević has published over 100 scientific papers in
international journals and conferences, contributing extensively
to the fields of combinatorial optimization, machine learning,
and soft computing. He is the principal investigator of two
nationally funded projects dealing with optimization of
containers in ports and routing of electric vehicles.
Furthermore, he also leads a project in collaboration with the
company AVL-AST.
His scientific excellence has been recognized with the Annual
Award for Young Researchers by the Croatian Parliament in 2023
and several other national institutions. Dr. Đurasević is an
active member of IEEE, IEEE CIS, ACM, and ACM SIGEVO, and
regularly serves as a reviewer for leading journals in
artificial intelligence and operations research.
Assoc. Prof. Mahdi Madani
Université Bourgogne Europe, France
Mahdi Madani received his Ph.D. in Electronics Systems from
the University of Lorraine on July 12, 2018. He was a temporary
research and teaching associate at IUT Auxerre, University of
Burgundy, from September 2018 to August 2020, and he was also a
temporary researcher at IETR laboratory and teaching associate
at IUT Nantes from September 2020 to August 2022. In September
2022, he joined the Université Bourgogne Europe and the CORES
team in the IMVA laboratory for the associate professor
position. His research interest is information security in new
digital networks, algorithm-architecture suitability, FPGA, and
SoC implementation of complex algorithms, applying security
techniques (confidentiality, integrity, encryption, chaotic
systems, etc.) to image, signal, and vision applications, and
exploring artificial intelligence packages for data and privacy
preserving.
Speech Title: Secure and Efficient Tele-Radiography
Based on the Fusion of a Convolutional Autoencoder and Chaotic
Latent Encryption
Abstract: This work addresses the dual challenges of
efficient compression and secure transmission for medical
images, particularly in bandwidth-constrained telemedicine
scenarios like tele-radiography. We proposed an end-to-end
pipeline combining deep learning-based compression with
chaos-based encryption. A convolutional autoencoder (CAE),
optimized with a Structural Similarity Index Measure (SSIM) loss
function and incorporating residual connections and batch
normalization, achieves an 8:1 (87.5%) compression ratio on
Chest X-ray images while maintaining a high fidelity of 96% SSIM
and 36 dB Peak signal-to-noise ratio (PSNR). To secure the
compact latent representation generated by the CAE, we introduce
a lightweight, chaos-based encryption scheme operating directly
on the latent space. This scheme utilizes a logistic map for
confusion and secure permutations for diffusion. The
experimental results confirm the effectiveness of the
compression module in preserving high-frequency details and the
encryption scheme’s resistance against statistical attacks, by
achieving high entropy (7.92), strong randomness (0.99),
correlation (close to 0 in horizontal, vertical, and diagonal
directions), and very sensitive to small changes in the key (1
single bit change conduct to a completely different keystream).
Our work offers a promising solution for secure and efficient
medical image transmission over constrained networks.
Assoc. Prof. Maciej Kusy
Rzeszow University of Technology, Poland
Maciej Kusy received his MSc degree in Electrical
Engineering from the Rzeszów University of Technology, Poland,
in 2000; his PhD in Biocybernetics and Biomedical Engineering
from the Warsaw University of Technology, Poland, in 2008; and
his DSc in Information and Communication Technology from the
Systems Research Institute of the Polish Academy of Sciences,
Warsaw, Poland, in 2019. He is currently an Associate Professor
at the Faculty of Electrical and Computer Engineering, Rzeszów
University of Technology. His research interests focus on
artificial intelligence, particularly machine learning,
generative learning, data mining, and video/image processing.
Speech Title: Task-Focused Label Selection for Improving YOLO Performance in Video Detection
Abstract: The talk will focus on enhancing urban scene
datasets by introducing critical object categories through the
use of an open-vocabulary detection model. This innovation
enables automatic annotation, eliminating the need for manual
labelling and allowing fine-tuning of real-time detection
models. To simplify the training process and improve model
performance, static or less informative categories are
selectively excluded. This targeted approach addresses class
imbalance by prioritising task-relevant elements, even when they
are underrepresented in the dataset. Through prompt-guided
detection and efficient annotation conversion, the model is
trained on a reduced label set. Evaluation results demonstrate
consistent precision, stable or improved accuracy, and minimal
recall drops for certain categories — illustrating the
effectiveness of a simplified and focused labelling strategy.