Abstract
Artificial intelligence (AI) models have received considerable attention in recent years for their ability to identify optical coherence tomography (OCT) biomarkers with clinical diagnostic potential and predict disease progression. This study aims to externally validate a deep learning (DL) algorithm by comparing its segmentation of retinal layers and fluid with a gold-standard method for manually adjusting the automatic segmentation of the Heidelberg Spectralis HRA + OCT software Version 6.16.8.0. A total of sixty OCT images of healthy subjects and patients with intermediate and exudative age-related macular degeneration (AMD) were included. A quantitative analysis of the retinal thickness and fluid area was performed, and the discrepancy between these methods was investigated. The results showed a moderate-to-strong correlation between the metrics extracted by both software types, in all the groups, and an overall near-perfect area overlap was observed, except for in the inner segment ellipsoid (ISE) layer. The DL system detected a significant difference in the outer retinal thickness across disease stages and accurately identified fluid in exudative cases. In more diseased eyes, there was significantly more disagreement between these methods. This DL system appears to be a reliable method for accessing important OCT biomarkers in AMD. However, further accuracy testing should be conducted to confirm its validity in real-world settings to ultimately aid ophthalmologists in OCT imaging management and guide timely treatment approaches.

Abstract
This paper introduces a mathematical model designed to optimise the operation of natural gas distribution networks, considering the injection of hydrogen in multiple nodes. The model is designed to optimise the quantity of hydrogen injected to maintain pressure, gas flows, and gas quality indexes (Wobbe index (WI) and higher heating value (HHV)) within admissible limits. This study also presents the maximum injection allowable of hydrogen correlated with the gas quality index variation. The model has been applied to a case study of a gas network with four distinct scenarios and implemented using Python. The findings of the case study quantify the maximum permitted volume of hydrogen in the network, the total savings in natural gas, and the reduction in carbon dioxide emissions. Lastly, a sensitivity analysis of injected hydrogen as a function of the Wobbe index (WI) and Higher Heating Value (HHV) limits relaxation.

Abstract
This study aimed to test an Instructional Design model prototype for Virtual Reality (VR) in Higher Education. Aqualitative research methodologywas used, employing questionnaires and observations for data collection. The research had three main objectives: (1) to identify the applicability and effectiveness of the VR Instructional Design model, (2) to evaluate participants' experience with immersion, interactivity, and usability of the VR environment, and (3) to obtain feedback from participants about their VR experience. The study involved two sessions. In the first session, participants were introduced to the VR environment, and their initial adaptation difficulties were observed. Informal interviews and a questionnaire collected feedback on immersion, interactivity, interest, and educational potential of VR. The second session indicated the need for revisions in applicability and ease of use. Based on student feedback, session planning should consider initial adaptation, teacher training, equipment availability, interaction elements, resources, realism, immersion, safety, comfort, session duration, communication, collaboration, and clear content delivery. Providing alternative plans for technical failures is essential. Despite these challenges, participants expressed interest in participating in VR sessions and activities.

Abstract

Abstract
This paper introduces ROSAR, a novel framework enhancing the robustness of deep learning object detection models tailored for side-scan sonar (SSS) images, generated by autonomous underwater vehicles using sonar sensors. By extending our prior work on knowledge distillation (KD), this framework integrates KD with adversarial retraining to address the dual challenges of model efficiency and robustness against SSS noises. We introduce three novel, publicly available SSS datasets, capturing different sonar setups and noise conditions. We propose and formalize two SSS safety properties and utilize them to generate adversarial datasets for retraining. Through a comparative analysis of projected gradient descent (PGD) and patch-based adversarial attacks, ROSAR demonstrates significant improvements in model robustness and detection accuracy under SSS-specific conditions, enhancing the model's robustness by up to 1.85%. ROSAR is available at https://github.com/remaro-network/ROSAR-framework.

Abstract