Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Arban Uka has a total of 64 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Artificial intelligence for brain cancer management: multimodal digital workflows for diagnosis, treatment planning, and longitudinal monitoring

    by Ganesan Saibaba, Olawumi Giwa, Thomas C. Booth, Terezia Kiskova-Simkova, Arban Uka, Florenc Skuka, Bilal Javed, Dário Ferreira, Sandra Ferreira, Jewel Sengupta, et al.

  • Open Access

    Efficient and accurate cell image segmentation through optimizer and loss function fine-tuning with quantization and pruning

    by Xhoena Polisi, Arban Uka, Daniel Avdiu, Dimitrios A Karras

    Convolutional deep learning is commonly and frequently used nowadays for high data throughput in image analysis and computer vision applications. The size of such models depends on the number of hyperparameters and the precision required for each hyperparameter. The training of these models requires enormous computational power for images of high dimensionality, resulting in trained models of large sizes. To solve this challenge, several approaches have been used, including quantization and pruning. These techniques have been proven to be effective in reducing the size of the models at the expense of lower accuracy, while in order to ensure higher performance, several hyperparameters need to be optimized. This research effort attempts to evaluate the effect of different combinations of classical and hybrid optimizer-loss functions with the U-Net convolutional deep learning model regarding their segmentation accuracy in cell imaging, while implementing both quantization and pruning. Different metrics are used to evaluate the training and testing process, such as precision, recall, F1 score, dice coefficients, Jaccard index, etc. It is found that the hybrid loss function combination of binary cross-entropy with Jaccard loss functions and RMSprop as the best optimizer results in the highest segmentation accuracy. These combinations of optimizers and loss functions are implemented with post-training quantization and quantization-aware training, and in both cases, the accuracy metrics were not compromised significantly, as they resulted in a decrease of less than 1%. Pruning, as a process to reduce the computational complexity and the size of the model, has proven to be effective by also providing the highest performance among other methods.


  • Open Access

    Optimizing combination of feature extraction and classifiers in supervised classification of cells images

    by Xhoena Polisi Duro, Arban Uka, Griselda Alushllari, Albana Ndreu Halili, Dimitrios A. Karras, Nihal Engin Vrana

    Developments in the medical field have opened the opportunity to conduct analyses on a personalized patient level. One of the important analyses that can be conducted is the cellular response to engineered materials, and the most appropriate non-invasive methods are imaging. These images of the cells are unstained brightfield images, as they are acquired from multiparametric microfluidic chambers in the presence of biomaterials and fluids that can change the optical path length over time as the cells’ health state is monitored. These experimental conditions lead to an image dataset with unique illumination, texture, and noise spectrum. This study explores the optimization of supervised cell classification by combining feature extraction architectures and machine learning classifiers, with a focus on applications in biomaterial risk assessment. Brightfield microscopy images of three cell types (A549, BALB 3T3, and THP1) were analyzed to evaluate the impact of Inception V3, Squeeze Net, and VGG16 architectures paired with classifiers including KNN, Decision Tree, Random Forest, AdaBoost, Neural Networks, and Naïve Bayes. Dimensionality reduction using Information Gain was applied to improve computational efficiency and accuracy. Butterworth filters with varying parameters were used to balance the enhancement of image features and noise removal, improving classification performance in certain cases. Experimental results demonstrate that the VGG16 architecture, when paired with Neural Networks, achieves higher classification accuracy as measured by different metrics. The improved accuracy when using Butterworth filters compared to the unfiltered dataset and the differences between various Butterworth filters indicate the importance of optimizing filter parameters for these types of images.


  • Open Access

    Fourier Ptychography Microscopy Resolution Improvement Employing Refocusing

    by Arban Uka, Gent Imeraj, Bjorna Qesaraku, Besmir Shehu


  • Open Access

    The Effect of Students’ Experience with the Transition from Primary to Secondary School on Self-Regulated Learning and Motivation

    by Ana Uka, Arban Uka

    The transition from primary to secondary school is more successful when students’ learning is consistent. Students are also more likely to enjoy school, engage with learning, and have a high academic achievement in secondary school when they feel motivated. This is a critical aspect, especially in cases in which global pandemic situations allow only online schooling opportunities. Students that are away from school lack the traditional sources of motivation and self-regulated learning skills; thus, research is needed to identify other important factors that can be developed in remote settings. The aim of this study was to find out how students perceive their experience with the transition from primary to secondary school and how such a transition influences students’ self-regulated learning (SRL) and motivation. Self-reported data were collected during the COVID-19 breakout from a total of n = 80 sixth and seventh grade students aged 12–14 years old. The results showed that students had a successful transition, especially when they were supported by their parents and teachers. Next, bivariate Pearson correlation analysis indicated that students’ perceptions about their experience with the transition from primary to secondary school, their self-regulated learning, and their motivation were significantly correlated. No gender differences were found among any of the main study variables. Teachers can foster students’ SRL skills by implementing effective teaching methods and by guiding them towards SRL-enhancing techniques.


  • Open Access

    The Effect of Students’ Experience with the Transition from Primary to Secondary School on Self-Regulated Learning and Motivation

    by Ana Uka, Arban Uka

    Transition from primary to secondary school is more successful when students’ learning is consistent. Students are also more likely to enjoy the school, engage with learning, and have a high academic achievement in the secondary school when they feel motivated. This is a critical aspect especially in cases when global pandemics situations allow only the online schooling opportunity. Students that are away from school lack the traditional sources of motivation and self-regulated learning skills, thus research is needed to identify other important factors that can be developed in remote settings. The aim of this study was to find out how students perceive their experience with the transition from primary to secondary school and how such a transition influences students’ self-regulated learning (SRL) and motivation. Self-reported data were collected during the COVID-19 breakout from a total of N=80, 6th and 7th grade students aged 12-14 years old. Results showed that students had a successful transition, especially when they are supported by their parents and teachers. Next, Bivariate Pearson Correlation analysis indicated that students’ perceptions about their experience with the transition from primary to secondary school and their self-regulated learning and motivation are significantly correlated. No gender differences were found among all main study variables.


  • Effect of Preprocessing on Performance of Neural Networks for Microscopy Image Classification

    by Arban Uka, Xhoena Polisi, Julien Barthes, Albana Ndreu Halili, Florenc Skuka, Nihal Engin Vrana

    Medical field depends heavily on understanding and analyzing microscopy images of cells to better diagnose diseases, to evaluate the effectiveness of various medical treatments and to determine their health under stress. The amount of data that needs to be analyzed has increased and computer assisted analysis has become crucial as it would be very labor intensive for the medical practitioners otherwise. Many of the images are acquired using brightfield microscopy with no staining in order to avoid all the side effects. The unstained images have some associating challenges as they suffer from random nonuniform illumination, low contrast, relatively high transparency of the cytoplasm. The initial challenge of the large amount of data calls for the use of deep learning algorithms, whereas the other structural challenges call for the need to carefully train the convolutional neural networks in order to have a reliable system of evaluation. We have prepared a dataset of 20.000 images and we have tested the trained models on datasets with different number of images (N=300-8000). Here is this work we present classification of the cell health using convolutional neural networks and monitor the effect of the preprocessing steps on the overall accuracy.


  • Iris Recognition Performance Analysis for Noncooperative Conditions

    by Oktay Koc, Arban Uka, Maaruf Ali, Klevis Muda, Orges Balla, Albana Roci

    A biometric system is presented using the human iris to help determine the authenticity of an individual. The system extracts the unique features of the iris that are recorded in templates. These templates are then compared with other irides utilising Daugman’s method. This follows a strict procedure (including segmentation, normalization, encoding and matching) over which a user has complete control. Often the recognition phase is crucial in nonoptimal or noncooperative conditions. In this work, a comparison is made of the relative accuracy of utilizing noisy iris datasets. The performance is analysed for a different number of iris images per person, for different number of individuals, for different noise levels using three different segmentations and three different encoding schemes. Adjustment of the Gabor filters’ bandwidth used in the encoding stage proves to be decisive in improving the accuracy for higher noise levels.