Abstract
[No abstract available]

Abstract

Abstract
The optimal management of distributed energy resources (DERs) and renewable-based generation in multi -energy systems (MESs) is crucial as it is expected that these entities will be the backbone of future energy sys-tems. To optimally manage these numerous and diverse entities, an aggregator is required. This paper proposes the self-scheduling of a DER aggregator through a hybrid Info-gap Decision Theory (IGDT)-stochastic approach in an MES. In this approach, there are several renewable energy resources such as wind and photovoltaic (PV) units as well as multiple DERs, including combined heat and power (CHP) units, and auxiliary boilers (ABs). The approach also considers an EV parking lot and thermal energy storage systems (TESs). Moreover, two demand response (DR) programs from both price-based and incentive-based categories are employed in the microgrid to provide flexibility for the participants. The uncertainty in the generation is addressed through stochastic pro-gramming. At the same time, the uncertainty posed by the energy market prices is managed through the application of the IGDT method. A major goal of this model is to choose the risk measure based on the nature and characteristics of the uncertain parameters in the MES. Additionally, the behavior of the risk-averse and risk -seeking decision-makers is also studied. In the first stage, the sole-stochastic results are presented and then, the hybrid stochastic-IGDT results for both risk-averse and risk-seeker decision-makers are discussed. The pro-posed problem is simulated on the modified IEEE 15-bus system to demonstrate the effectiveness and usefulness of the technique.

Abstract
Two-dimensional echocardiography is the most widely used non-invasive imaging modality due to its fast acquisition time, low cost, and high temporal resolution. Accurate segmentation of the left ventricle in echocardiography is vital for ensuring the accuracy of subsequent diagnosis. Currently, numerous efforts have been made to automatize this task and various public datasets have been released in recent decades to further develop present research. However, medical datasets acquired at different institutions have inherent bias caused by various confounding factors, such as operation policies, machine protocols, treatment preference, etc. As a result, models trained on one dataset, regardless of volume, cannot be confidently utilized for the others. In this study, we investigated model robustness to dataset bias using two publicly available echocardiographic datasets. This work validates the efficacy of a supervised deep learning model for left ventricle segmentation and ejection fraction prediction, outside the dataset on which it was trained. The exposure of this model to unseen, but related samples without additional training maintained a good performance. However, a performance decrease from the original results can be observed, while the impact of quality is also noteworthy with lower quality data leading to decreased performance.

Abstract
This paper presents the main contributions of the VCMI Team to the ImageCLEFmedical GANs 2023 task. This task aims to evaluate whether synthetic medical images generated using Generative Adversarial Networks (GANs) contain identifiable characteristics of the training data. We propose various approaches to classify a set of real images as having been used or not used in the training of the model that generated a set of synthetic images. We use similarity-based approaches to classify the real images based on their similarity to the generated ones. We develop autoencoders to classify the images through outlier detection techniques. Finally, we develop patch-based methods that operate on patches extracted from real and generated images to measure their similarity. On the development dataset, we attained an F1-score of 0.846 and an accuracy of 0.850 using an autoencoder-based method. On the test dataset, a similarity-based approach achieved the best results, with an F1-score of 0.801 and an accuracy of 0.810. The empirical results support the hypothesis that medical data generated using deep generative models trained without privacy constraints threatens the privacy of patients in the training data. © 2023 Copyright for this paper by its authors.

Abstract
To design and deploy their supply chains, companies must naturally take quite different decisions, some being strategic or tactical, and others of an operational nature. This work resulted in a decision support system for optimising a biomass supply chain in Portugal, allowing a more efficient operations management, and enhancing the design process. Uncertainty and variability in the biomass supply chain is a critical issue that needs to be considered in the production planning of bioenergy plants. A simulation/optimisation framework was developed to support decision-making, by combining plans generated by a resource allocation optimisation model with the simulation of disruptive wildfire scenarios in the forest biomass supply chain. Different scenarios have been generated to address uncertainty and variability in the quantity and quality of raw materials in the different supply nodes. Computational results show that this simulation/optimisation approach can have a significant impact in the operations efficiency, particularly when disruptions occur closer to the end of the planning horizon. The approach seems to be easily scalable and easy to extend to other sectors.