Abstract
Colorectal cancer is the second most common cancer and the second with the highest associated deaths in the world. Methods used in clinical practice for colon cancer diagnosis are fairly effective but quite unpleasant and not always applicable in situations where the patient has symptoms of colonic obstruction. This problem can be solved by the use of optical methods that can be applied less invasively. This study presents the results of classification of cancerous and healthy colon tissue absorption coefficient spectra. The absorption coefficient was measured using direct calculations from the total reflectance and total transmittance spectra obtained ex vivo. Classification was performed using support vector machine, multilayer perceptron and linear discriminant analysis.

Abstract
Encouraging pre-university students to pursue engineering courses at the university level is essential to meet the industry’s escalating demand for engineers. Each year, universities host hundreds of secondary students who tour their facilities to get a feel for the academic environment. This paper discusses an educational experiment designed as part of a semester-long undergraduate project in Informatics Engineering. The project involves tailoring a Dobot Magician robot, equipped with a standard webcam, to engage in a game of tic-tac-toe against a human user. The camera stream is continuously processed by a computer vision algorithm to detect the pieces placement in the game board. The paper outlines the project development stages, the elements involved, and presents preliminary test results. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Abstract
We explore deep generative models to generate case-based explanations in a medical federated learning setting. Explaining AI model decisions through case-based interpretability is paramount to increasing trust and allowing widespread adoption of AI in clinical practice. However, medical AI training paradigms are shifting towards federated learning settings in order to comply with data protection regulations. In a federated scenario, past data is inaccessible to the current user. Thus, we use a deep generative model to generate synthetic examples that protect privacy and explain decisions. Our proof-of-concept focuses on pleural effusion diagnosis and uses publicly available Chest X-ray data. © 2024 IEEE.

Abstract
Nearest neighbor search (NNS) is one of the main concerns in data stream applications since similarity queries can be used in multiple scenarios. Online NNS is usually performed on a sliding window by lazily scanning every element currently stored in the window. This paper proposes Sliding Window-based Incremental Nearest Neighbors (SWINN), a graph-based online search index algorithm for speeding up NNS in potentially never-ending and dynamic data stream tasks. Our proposal broadens the application of online NNS-based solutions, as even moderately large data buffers become impractical to handle when a naive NNS strategy is selected. SWINN enables efficient handling of large data buffers by using an incremental strategy to build and update a search graph supporting any distance metric. Vertices can be added and removed from the search graph. To keep the graph reliable for search queries, lightweight graph maintenance routines are run. According to experimental results, SWINN is significantly faster than performing a naive complete scan of the data buffer while keeping competitive search recall values. We also apply SWINN to online classification and regression tasks and show that our proposal is effective against popular online machine learning algorithms.

Abstract
Despite the ubiquitous term climate neutral cities, there is a distinct lack of quantifiable and meaningful municipal decarbonization goals in terms of the targeted energy balance and composition that collectively connect to national scenarios. In this paper we present a simple but useful allocation approach to derive municipal targets for energy demand reduction and renewable expansion based on national energy transition strategies in combination with local potential estimators. The allocation uses local and regional potential estimates for demand reduction and the expansion of renewables and differentiates resulting municipal needs of action accordingly. The resulting targets are visualized and opened as a decision support system (DSS) on a web-platform to facilitate the discussion on effort sharing and potential realization in the decarbonization of society. With the proposed framework, different national scenarios, and their implications for municipal needs for action can be compared and their implications made explicit.

Abstract
Solar wind forecasting is a core component of Space Weather, a field that has been the target of many novel machine-learning approaches. The continuous monitoring of the Sun has provided an ever-growing ensemble of observations, facilitating the development of forecasting models that predict solar wind properties on Earth and other celestial objects within the solar system. This enables us to prepare for and mitigate the effects of solar wind-related events on Earth and space. The performance of some simulation-based solar wind models depends heavily on the quality of the initial guesses used as initial conditions. This work focuses on improving the accuracy of these initial conditions by employing a Recurrent Neural Network model. The study's findings confirmed that Recurrent Neural Networks can generate better initial guesses for the simulations, resulting in faster and more stable simulations. In our experiments, when we used predicted initial conditions, simulations ran an average of 1.08 times faster, with a statistically significant improvement and reduced amplitude transients. These results suggest that the improved initial conditions enhance the numerical robustness of the model and enable a more moderate integration time step. Despite the modest improvement in simulation convergence time, the Recurrent Neural Networks model's reusability without retraining remains valuable. With simulations lasting up to 12 h, an 8% gain equals one hour saved per simulation. Moreover, the generated profiles closely match the simulator's, making them suitable for applications with less demanding physical accuracy.