Understanding how localized anatomical variations influence brain age predictions is critical for clinical interpretability. However, existing explainability methods primarily capture associations and do not support controlled interventions to assess how specific morphometric changes affect model output. These approaches lack the ability to simulate coherent “what-if” scenarios at the regional level. We present a simulation framework that enables localized morphometric interventions with biologically consistent propagation of effects. When a single feature (e.g., gray matter volume) is perturbed, the framework automatically adjusts the remaining features within the same brain region to preserve statistical dependencies, estimated from a reference population. This produces realistic, covarying input profiles that any pretrained brain age model can evaluate without retraining. The method is tested on the publicly available OpenBHB dataset, focusing on two regions of interest selected for their contrasting correlation with age. Controlled perturbations of 1

Traditional evaluation methods for Explainable AI (XAI) often emphasize perceived clarity or task performance, overlooking the cognitive processes by which domain experts construct and refine their understanding. In this work, we propose a mental model-based evaluation framework that, grounded in cognitive science and Human-Computer Interaction (HCI), defines five analytical dimensions to assess how end users engage with XAI systems. Our goal is to support the design and evaluation of XAI interfaces that enable informed, trustworthy, and reflective decision-making in professional practice. We illustrate this perspective outlining how the mental model-based framework could be used to assess explanation effectiveness of Brain Age Predictor, an interactive interface for brain age prediction based on structural Magnetic Resonance Imaging (MRI) data.

The rapid advancement of digital technologies continues to reshape the way humans interact with software systems, raising critical questions about the balance between human values and the interplay of usability, user experience, and automation. In this landscape, HumanCentered Software Engineering (HCSE) plays a pivotal role in ensuring that software development remains aligned with user needs, ethical principles, and broader societal impacts. This edition of the HCSE workshop focuses on the intersection of interaction, Artificial Intelligence, and human value, emphasizing the need for new approaches in designing socio-technical systems that integrate intelligent automation while preserving human agency and well-being. With AI-driven decision-making, adaptive interfaces, and autonomous systems becoming ubiquitous, it is crucial to explore how software engineering can accommodate these innovations while maintaining transparency, fairness, and user trust. Beyond the traditional themes of IFIP Working Group 13.2 workshops, this edition invites contributions fostering discussion among researchers and practitioners to advance the dialogue on how HCSE can evolve to meet the demands of modern digital ecosystems while staying true to its human-centric principles.

This edition of the Human-Centered Software Engineering (HCSE) workshop focused on the intersection of interaction, artificial intelligence, and human value. The workshop had three main topics: (1) Human–AI Interaction and Cognitive Models for Future Systems, (2) Methods and Framework for Usability and User Experience, and (3) Smart and Secure Digital Futures. In the series of HCSE workshops and conferences, we see the need to explore and understand how software engineering and human–computer interaction can accommodate AI-driven decision-making, adaptive interfaces, and autonomous systems that become ubiquitous. The submissions focused on how to design this future interaction with AI. Interestingly, there is a need to re-interpret and revisit existing frameworks and understanding on how to support usability and user experience on the one hand and allow for smart and secure systems on the other. A key challenge for the future will be the balancing of these properties in human-centered design and development processes.

The rapid advancement of digital technologies continues to reshape the way humans interact with software systems, raising critical questions about the balance between human values and the interplay of usability, user experience, and automation. In this landscape, Human-Centered Software Engineering (HCSE) plays a pivotal role in ensuring that software development remains aligned with user needs, ethical principles, and broader societal impacts. This edition of the HCSE workshop focuses on the intersection of interaction, Artificial Intelligence, and human value, emphasizing the need for new approaches in designing socio-technical systems that integrate intelligent automation while preserving human agency and well-being. With AI-driven decision-making, adaptive interfaces, and autonomous systems becoming ubiquitous, it is crucial to explore how software engineering can accommodate these innovations while maintaining transparency, fairness, and user trust. Beyond the traditional themes of IFIP Working Group 13.2 workshops, this edition invites contributions fostering discussion among researchers and practitioners to advance the dialogue on how HCSE can evolve to meet the demands of modern digital ecosystems while staying true to its human-centric principles.

This study presents a systematic framework for evaluating convolutional neural networks (CNNs) in the context of brain age prediction, with a focus on interpretability through explainable AI (XAI) methods. Brain age prediction has advanced significantly using 2D and 3D CNNs, which provide high predictive accuracy. However, the complexity of these models creates challenges for clinical interpretation. Our framework not only assesses model performance but also provides deeper insights into various aspects of CNN models, including how the selected backgrounds can influence XAI values. By facilitating multisite data analysis, the framework helps identify the impact of site-specific characteristics on model behavior. The results underscore the importance of local explanations and highlight the need for careful interpretation when using population-level saliency maps.

Brain-Computer Interface (BCI) allows machines to be controlled by signals derived from EEG analysis. Several low-cost electroencephalographs are now available on the market, all of which provide highquality Electroencephalogram (EEG) signals. Detecting a user's emotional state by analysing their EEG signal is a fascinating approach in this field. In our research, we have attempted to detect the emotional state of nine different emotions. Our research presents a working prototype of an EEG-based emotion recognizer that can be used as biofeedback to provide information about the user's emotional state.

Brain age prediction is a valuable tool for distinguishing normal and pathological aging, offering quantitative insights into subtle brain structure changes. However, clinical adoption remains limited due to the complexity and low interpretability of Machine Learning (ML) algorithms. Human-Centered Artificial Intelligence (HCAI) and eXplainable AI (XAI) address these challenges by emphasizing user involvement, explainability, and facilitation of clinical applications. By combining advanced ML models with intuitive interfaces and transparent visualizations, HCAI fosters trust and usability in neurological practice. This paper presents the “Brain Age Predictor”, a web-based tool combining a deep learning model for brain age estimation with SHAP-based interpretability techniques and an interactive simulation panel. We conducted a preliminary study involving five neurology residents to assess its usability, interpretability, and potential value in clinical practice. The results show that the Brain Age Predictor effectively supports clinical exploration of brain aging, with participants praising its ease of use and clarity. Feedback highlighted areas for improvement, including richer visualizations, more detailed explanations, and tools for longitudinal patient monitoring.

Clinical pathways play a crucial role in guiding the treatment of specific medical conditions or patient populations, but often rely on basic textual documentation, leading to potential inefficiencies and delays in patient care. This paper reports the early stages of a research aiming at exploring the application of knowledge representation techniques in the digitalization of diagnostic and therapeutic care pathways. These techniques are used to annotate contextual data, patient information and medical guidelines with respect to a reference ontology. In this way, a comprehensive knowledge graph can be processed using rule-based approaches to support the patient care management process, providing physicians and medical practitioners with valuable insights about specific diseases.

This study explores the application of Artificial Intelligence (AI) in the diagnosis of Mild Cognitive Impairment (MCI) and Alzheimer’s Disease (AD), through Human-Computer Interaction (HCI), Human-Centered AI (HCAI), and Explainable AI (XAI). It evaluates three user interfaces designed to integrate AI insights with the clinical understanding of neurologists, aiming to refine diagnostic processes. Neurology professionals were involved to gauge their knowledge and confidence in the AI-supported diagnoses. Utilizing a remotely administered questionnaire, this research investigates clinicians’ views on XAI outputs, focusing on how results are visualized and their ability to engender trust in AI’s clinical utility. This method emphasizes the importance of clear, trustworthy AI systems in healthcare and underscores the essential role of effective human-AI collaboration in enhancing patient care and diagnostic precision.

The realm of music composition, augmented by technological advancements such as computers and related equipment, has undergone significant evolution since the 1970s. In the field algorithmic composition, however, the incorporation of artificial intelligence (AI) in sound generation and combination has been limited. Existing approaches predominantly emphasize sound synthesis techniques, with no music composition systems currently employing Nicolas Slonimsky’s theoretical framework. This article introduce NeuralPMG, a computer-assisted polyphonic music generation framework based on a Leap Motion (LM) device, machine learning (ML) algorithms, and brain-computer interface (BCI). ML algorithms are employed to classify user’s mental states into two categories: focused and relaxed. Interaction with the LM device allows users to define a melodic pattern, which is elaborated in conjunction with the user’s mental state as detected by the BCI to generate polyphonic music. NeuralPMG was evaluated through a user study that involved 19 students of Electronic Music Laboratory at a music conservatory, all of whom are active in the music composition field. The study encompassed a comprehensive analysis of participant interaction with NeuralPMG. The compositions they created during the study were also evaluated by two domain experts who addressed their aesthetics, innovativeness, elaboration level, practical applicability, and emotional impact. The findings indicate that NeuralPMG represents a promising tool, offering a simplified and expedited approach to music composition, and thus represents a valuable contribution to the field of algorithmic music composition.

This paper presents the results of the International Workshop on Human-Centered Software Engineering, which was part of the 19th International Conference promoted by the IFIP Technical Committee 13 on Human-Computer Interaction (Interact 2023). The leading topic of this edition of the workshop was “Rethinking the Interplay of Human–Computer Interaction and Software Engineering in the Age of Digital Transformation”. The workshop was characterized by an innovative format designed to bolster research efforts across various stages. Ten papers were presented by authors and discussed in the workshop. In terms of topics, there were three key sessions focusing on digitizing manufacturing processes, understanding users, and digitization for smart life Seven of these papers were extended in these proceedings.

This paper presents the results of the International Workshop on Human-Centered Software Engineering, which was part of the 19th International Conference promoted by the IFIP Technical Committee 13 on Human-Computer Interaction (Interact 2023). The leading topic of this edition of the workshop was "Rethinking the Interplay of Human-Computer Interaction and Software Engineering in the Age of Digital Transformation". The workshop was characterized by an innovative format designed to bolster research efforts across various stages. Ten papers were presented by authors and discussed in the workshop. In terms of topics, there were three key sessions focusing on digitizing manufacturing processes, understanding users, and digitization for smart life Seven of these papers were extended in these proceedings.

The work presented in this article is part of a broader research initiative whose focus revolves around the integration of Artificial Intelligence (AI) in diagnosing Mild Cognitive Impairment (MCI), in particular, investigating the reliability and stability of eXplainable AI (XAI) predictions concerning markers of MCI and Alzheimer's Disease. In order to foster neurologists' understanding, confidence and trust in the AI system results, the initial Machine Learning (ML) based analysis pipeline has been now extended to incorporate a graphical user interface (GUI) that would provide "neurologist-centred" explanations. In this article, the focus is on a preliminary study that involved neurology professionals to assess their understanding and confidence in AI-generated diagnoses presented through three alternative plots implemented in the system GUI.

Brain age, a biomarker reflecting brain health relative to chronological age, is increasingly used in neuroimaging to detect early signs of neurodegenerative diseases and support personalized treatment plans. Two primary approaches for brain age prediction have emerged: morphometric feature extraction from MRI scans and deep learning (DL) applied to raw MRI data. However, a systematic comparison of these methods regarding performance, interpretability, and clinical utility has been limited. In this study, we present a comparative evaluation of two pipelines: one using morphometric features from FreeSurfer and the other employing 3D convolutional neural networks (CNNs). Using a multisite neuroimaging dataset, we assessed both model performance and the interpretability of predictions through eXplainable Artificial Intelligence (XAI) methods, applying SHAP to the feature-based pipeline and Grad-CAM and DeepSHAP to the CNN-based pipeline. Our results show comparable performance between the two pipelines in Leave-One-Site-Out (LOSO) validation, achieving state-of-the-art performance on the independent test set ( M A E = 3.21 with DNN and morphometric features and M A E = 3.08 with a DenseNet-121 architecture). SHAP provided the most consistent and interpretable results, while DeepSHAP exhibited greater variability. Further work is needed to assess the clinical utility of Grad-CAM. This study addresses a critical gap by systematically comparing the interpretability of multiple XAI methods across distinct brain age prediction pipelines. Our findings underscore the importance of integrating XAI into clinical practice, offering insights into how XAI outputs vary and their potential utility for clinicians.