Abstract
Coronary artery disease (CAD) remains a leading cause of morbidity and mortality worldwide, underscoring the need for accurate and reliable diagnostic tools. While AI- driven models have shown significant promise in identifying CAD through imaging techniques, their 'black box' nature often hinders clinical adoption due to a lack of interpretability. In response, this paper proposes a novel approach to image captioning specifically tailored for CAD diagnosis, aimed at enhancing the transparency and usability of AI systems. Utilizing the COCA dataset, which comprises gated coronary CT images along with Ground Truth (GT) segmentation annotations, we introduce a hybrid model architecture that combines a Vision Transformer (ViT) for feature extraction with a Generative Pretrained Transformer (GPT) for generating clinically relevant textual descriptions. This work builds on a previously developed 3D Convolutional Neural Network (CNN) for coronary artery segmentation, leveraging its accurate delineations of calcified regions as critical inputs to the captioning process. By incorporating these segmentation outputs, our approach not only focuses on accurately identifying and describing calcified regions within the coronary arteries but also ensures that the generated captions are clinically meaningful and reflective of key diagnostic features such as location, severity, and artery involvement. This methodology provides medical practitioners with clear, context-rich explanations of AI-generated findings, thereby bridging the gap between advanced AI technologies and practical clinical applications. Furthermore, our work underscores the critical role of Explainable AI (XAI) in fostering trust, improving decision- making, and enhancing the efficacy of AI-driven diagnostics, paving the way for future advancements in the field.

Abstract

Abstract
Surface plasmon resonance (SPR) conventionally occurs at the interface of a thin metallic film and an external dielectric medium in fiber optics through core-guided light. However, this work introduces theoretical and experimental evidence suggesting that the SPR in optical fibers can also be induced through light scattering from Au nanoparticles (NPs) on the thin metallic film, defined as nanoparticle-induced SPR (NPI-SPR). This method adheres to phase-matching conditions between SPR dispersion curves and the wave vectors of scattered light from Au NPs. Experimentally, these conditions are met on an etched optical fiber, enabling direct interaction between light and immobilized Au NPs. Compared to SPR, NPI-SPR exhibits stronger field intensity in the external region and wavelength tuning capabilities (750 to 1250 nm) by varying Au NP diameters (20 to 90 nm). NPI-SPR demonstrates refractive index sensitivities of 4000 to 4416 nm per refractive index unit, nearly double those of typical SPR using the same optical fiber configuration sans Au NPs. Additionally, NPI-SPR fiber configuration has demonstrated its applicability for developing biosensors, achieving a remarkable limit of detection of 0.004 nm for thrombin protein evaluation, a twenty-fold enhancement compared to typical SPR. These findings underscore the intrinsic advantages of NPI-SPR for sensing. Surface plasmon resonance (SPR) typically occurs at the interface of a thin metallic film and a dielectric medium in fiber optics. This work presents evidence of nanoparticle-induced SPR (NPI-SPR) in optical fibers through light scattering from Au nanoparticles on the thin metallic film. NPI-SPR offers stronger field intensity, wavelength tuning, and enhanced refractive index sensitivities, making it advantageous for biosensing applications. image

Abstract
This study introduces a method to assess the quality of Explainable Artificial Intelligence (XAI) algorithms in dynamic data streams, concentrating on the fidelity and stability of feature-importance and rule-based explanations. We employ XAI metrics, such as fidelity and Lipschitz Stability, to compare explainers between each other and introduce the Comparative Expert Stability Index (CESI) for benchmarking explainers against domain knowledge. We adopted the aforementioned metrics to the streaming data scenario and tested them in an unsupervised classification scenario with simulated distribution shifts as different classes. The necessity for adaptable explainers in complex scenarios, like failure detection is underscored, stressing the importance of continued research into versatile explanation techniques to enhance XAI system robustness and interpretability.

Abstract
Learning environments unquestionably enable learners to develop their pedagogical and scientific processes efficiently and effectively. Thus, considering the impossibility of not having conditions of autonomy over the routine underlying the studies and, consequently, not having guarantees of the learning carried out makes the learners experience gaps in the domain of materials adequate to their actual needs. The paper's objective is to present the relevance of the applicability of Artificial Intelligence in Recommendation Systems, reinforced through the Assurance of Learning, oriented towards adaptive-personalized practice in corporate e-learning contexts. The research methodology underlying the work fell on Design Science Research, as it is considered adequate to support the research, given the need to carry out the design phases, development, construction, evaluation, validation of the artefact and, finally, communication of the results. The main results instigate the development of an Adaptive-Personalized Learning framework for corporate e-learning, provided with models of Artificial Intelligence and guided using the Assurance of Learning process. It becomes central that learners can enjoy adequate academic development. In this sense, the framework has an implicit structure that promotes the definition of personalized attributes, which involves recommendations and customizations of content per profile, including training content that will be suggested and learning activity content that will be continuously monitored, given the specific needs of learners.

Abstract
Since digital media has become increasingly popular, video processing has expanded in recent years. Video processing systems require high levels of processing, which is one of the challenges in this field. Various approaches, such as hardware upgrades, algorithmic optimizations, and removing unnecessary information, have been suggested to solve this problem. This study proposes a video saliency map based method that identifies the critical parts of the video and improves the system's overall performance. Using an image registration algorithm, the proposed method first removes the camera's motion. Subsequently, each video frame's color, edge, and gradient information are used to obtain a spatial saliency map. Combining spatial saliency with motion information derived from optical flow and color-based segmentation can produce a saliency map containing both motion and spatial data. A nonlinear function is suggested to properly combine the temporal and spatial saliency maps, which was optimized using a multi-objective genetic algorithm. The proposed saliency map method was added as a preprocessing step in several Human Action Recognition (HAR) systems based on deep learning, and its performance was evaluated. Furthermore, the proposed method was compared with similar methods based on saliency maps, and the superiority of the proposed method was confirmed. The results show that the proposed method can improve HAR efficiency by up to 6.5% relative to HAR methods with no preprocessing step and 3.9% compared to the HAR method containing a temporal saliency map.