Urinary bladder cancer is one of the most common cancers of the urinary tract. This cancer is characterized by its high metastatic potential and recurrence rate. Due to the high metastatic potential and recurrence rate, correct and timely diagnosis is crucial for successful treatment and care. With the aim of increasing diagnosis accuracy, artificial intelligence algorithms are introduced to clinical decision making and diagnostics. One of the standard procedures for bladder cancer diagnosis is computer tomography (CT) scanning. In this research, a transfer learning approach to the semantic segmentation of urinary bladder cancer masses from CT images is presented. The initial data set is divided into three sub-sets according to image planes: frontal (4413 images), axial (4993 images), and sagittal (996 images). First, AlexNet is utilized for the design of a plane recognition system, and it achieved high classification and generalization performances with an $\overline{AU{C}_{micro}}$ of 0.9999 and $\sigma \left(AU{C}_{micro}\right)$ of 0.0006. Furthermore, by applying the transfer learning approach, significant improvements in both semantic segmentation and generalization performances were achieved. For the case of the frontal plane, the highest performances were achieved if pre-trained ResNet101 architecture was used as a backbone for U-net with $\overline{DSC}$ up to 0.9587 and $\sigma \left(DSC\right)$ of 0.0059. When U-net was used for the semantic segmentation of urinary bladder cancer masses from images in the axial plane, the best results were achieved if pre-trained ResNet50 was used as a backbone, with a $\overline{DSC}$ up to 0.9372 and $\sigma \left(DSC\right)$ of 0.0147. Finally, in the case of images in the sagittal plane, the highest results were achieved with VGG-16 as a backbone. In this case, $\overline{DSC}$ values up to 0.9660 with a $\sigma \left(DSC\right)$ of 0.0486 were achieved. From the listed results, the proposed semantic segmentation system worked with high performance both from the semantic segmentation and generalization standpoints. The presented results indicate that there is the possibility for the utilization of the semantic segmentation system in clinical practice.

Izvantjelesno mrvljenje kamenaca (ESWL) je novija terapijska metoda u liječenju urolitjaze. Temelji se na uporabi šok-valova koji se stvaraju u generatoru, fokusiraju i usmjeruju na kamenac u tijelu pacijenta. Razbijanje kamenca je...

Bladder cancer is one of the most common malignancies of the urinary tract. It is characterized by high metastatic potential and a high recurrence rate, which significantly complicates diagnosis and treatment. In order to increase the accuracy of the diagnostic procedure, algorithms based on artificial intelligence are introduced. This paper presents the principle of selection of convolutional neural network (CNN) models based on a multi-objective approach that maximizes classification and generalization performance. Model selection is performed on two standard CNN architectures, AlexNet and VGG-16. Classification performances are measured by using ROC analysis and the resulting AUC value. On the other hand, generalization performances are evaluated by using a 5-fold cross-validation procedure. By using these two metrics, a multi-objective fitness function, used in meta-heuristic algorithms, is designed. The multi-objective search was performed using a Genetic algorithm (GA) and a Discrete Particle Swarm (D-PS) algorithm. From obtained results, it can be noticed that such an approach has resulted in CNN models that are defined with high classification and generalization performances. When a GA-based approach is used, fitness values up to 0.97 are achieved. On the other hand, by using the D-PS approach, fitness values up to 0.99 are achieved pointing towards the conclusion that such an approach has provided models with higher classification and generalization performances.

Današnja klinička praksa, ali i istraživanja, temelje se na alo(homo)transplantaciji. Najzastupljeniji organ po broju u transplantacijskoj medicini svakako je bubreg. Transplantacija bubrega uvriježen je, najbolji i najekonomičniji...

This study evaluates the accuracy of medical-report interpretations generated by large language models in comparison to medically verified sources, with a particular focus on urology. The main goal is to examine the extent to which LLMs can reliably and precisely explain medical findings, with emphasis on expert urological terminology. For the evaluation of interpretation accuracy, real clinical documentation was used, specifically a dataset of 40 urological reports from the Clinical Hospital Center Rijeka. The generated explanations were compared with medically correct definitions using quantitative metrics assessing interpretability, specificity, and clinical accuracy. The results revealed a clear stratification of model performance, with DeepSeek-V3.2, GLM-4.6, MiMo-V2-Flash, and Qwen2.5-7B-Instruct achieving the highest semantic accuracy and consistency, while several models exhibited unstable behavior characterized by occasional catastrophic failures. Overall, the findings indicate that although LLMs can produce clinically coherent explanations of urological findings, their variability and susceptibility to hallucinations necessitate human oversight, supporting their use primarily as decision-support tools rather than autonomous clinical interpreters.