Aims: This study introduces and evaluates a virtual reality (VR) prototype designed for the Loving-Kindness Meditation (LKM) to support
mental health rehabilitation and relaxation in clinical contexts. The aims include the co-creation of a VR-based mindfulness experience with
clinical experts and the evaluation of its usability, user experience, and short-term effects on relaxation, affect, and self-compassion.

Methods: Following a design thinking and co-creation approach, the VR-based LKM experience was developed iteratively with input from
clinicians and computer scientists. The final prototype was implemented for the Meta Quest 3 and included five immersive scenes
representing phases of the LKM and transition scenes guided by a professionally narrated audio track. Eleven participants (M = 36.5 years,
SD = 14.6) experienced the 12-minute session. Pre- and post-session measures included relaxation, positive and negative affect schedule,
and self-compassion, complemented in the end by the Igroup Presence Questionnaire, usability measures and a semi-structured qualitative
interview.

Results: Participants reported significant decreases in negative affect (t(10) = -2.512, p = .0307, d = -1.037) and stress (t(10) = -3.318, p = .007,
d = -1.328), as well as increases in relaxation (t(10) = 5.487, p < .0001, d = 2.471) and self-compassion (t(10) = 2.231, p = .0497, d = 0.283).
Usability was rated as excellent (M = 92.5), and presence as good (M = 4.0, SD = 0.43). Qualitative feedback described the experience as
calming, aesthetically pleasing, and easy to engage with, highlighting the falling leaves and pulsating orb as effective design elements.

Conclusion: The co-designed VR-LKM prototype was perceived as highly usable and beneficial for inducing relaxation and self-compassion,
suggesting its potential as a supportive tool for clinical mindfulness interventions. The results indicate that immersive VR can effectively
facilitate engagement and emotional regulation, providing a foundation for future clinical trials and broader implementation in therapeutic
and wellness settings.

In der Coburger Landesbibliothek (Kt 478) befindet sich einer der ältesten seriell hergestellten Reliefgloben aus der Produktion des Berliner Globenherstellers Karl Wilhelm Kummer. Er stammt aus der Zeit um 1820/22. Wir werden Gelegenheit haben, diesen besonderen Globus im Original zu betrachten.

Blinden- und Reliefgloben haben keine glatte, sondern eine erhabene Oberfläche, durch die sie Gebirge und Täler besonders betonen und ertastbar machen. Sie vermitteln ein plastisches Bild der Geografie und wurden ursprünglich für Blindenschulen entwickelt.

Die Auseinandersetzung mit dem Globus führt zu grundsätzlichen Fragen über die Verwendung, die Techniken und Konventionen von Modellen unseres Planeten. Im Zeitalter der Globalisierung ist die Frage nicht unwesentlich, worum es sich bei dem "Globalen" eigentlich handelt und welche unterschiedlichen Wahrnehmungs-weisen darauf möglich sind.

We investigated the adaptation and performance of Masked Autoencoders (MAEs) with Vision Transformer (ViT) architectures for self-supervised representation learning on one-dimensional (1D) ultrasound signals. Although MAEs have demonstrated significant success in computer vision and other domains, their use for 1D signal analysis, especially for raw ultrasound data, remains largely unexplored. Ultrasound signals are vital in industrial applications such as nondestructive testing (NDT) and structural health monitoring (SHM), where labeled data are often scarce and signal processing is highly task-specific. We propose an approach that leverages MAE to pre-train on unlabeled synthetic ultrasound signals, enabling the model to learn robust representations that enhance performance in downstream tasks, such as time-of-flight (ToF) classification. This study systematically investigated the impact of model size, patch size, and masking ratio on pre-training efficiency and downstream accuracy. Our results show that pre-trained models significantly outperform models trained from scratch and strong convolutional neural network (CNN) baselines optimized for the downstream task. Additionally, pre-training on synthetic data demonstrates superior transferability to real-world measured signals compared with training solely on limited real datasets. This study underscores the potential of MAEs for advancing ultrasound signal analysis through scalable, self-supervised learning.