HTWG Hochschule Konstanz - Technik, Wirtschaft und Gestaltung

The origin of the extensive phenotypic divergence characterizing adaptive radiation could often be geographically localized and genetically simple. In a classic case of a trophically polymorphic cichlid fish (Herichthys minckleyi), we investigated alternative genomic processes that could have produced its extreme within-population variation in pharyngeal jaw tooth size. First, we generated a high-quality reference genome for its close relative (H. cyanoguttatus) to dissect the genetic architecture of this dental polymorphism. Then, using whole genome resequencing across the small Cuatro Ciénegas valley where H. minckleyi is endemic, we found substantial micro-geographic subdivision and effectively no genetic structure due to pharyngeal morphotype. We also employed quantitative trait loci mapping and genome wide association to pinpoint a single peak in an Iroquois-related (IRX) gene cluster associated with H. minckleyi's dental divergence. Interspecific introgression in this genomic region appears negligible, suggesting the genomic basis of the polymorphism likely arose within cichlids confined to Cuatro Ciénegas. Because H. minckleyi tooth size disparity is comparable to that found in all Central American cichlids, this offers a striking example of how genomic divergence at a single locus could produce a punctuated burst of eco-morphological divergence that generates phenotypic breadth comparable to a highly diverse cichlid adaptive radiation.

Abstract Adverse environmental conditions in water bodies, such as currents and waves, can be very stressful for net cage structures of fish farms. This work aims to develop a simulation model that can qualitatively estimate the mechanical loads in the wires of net cage segments of specific fish farms due to current influences. To set up the model, flow measurements were carried out in a wind tunnel on net cage surfaces with different wire diameters and mesh sizes. Different flow angles were tested. These measurements were validated by flow simulations in Ansys CFX. In order to minimize the simulation time, the net cage surfaces of the aquaculture systems were replaced by porous media. Functions for drag and lift coefficients as well as velocity reduction factors were derived from the flow simulations in dependence on porosity and angle of attack. A fluid-structure interaction was performed by transferring these functions to a beam-node model created with Ansys APDL representing the entire aquaculture facility. The simulation model makes it possible to qualitatively determine the mechanical stresses in the individual wires of the net cage surfaces. The findings on the most stressed areas can be used to optimize the design of the aquaculture systems in detail using the submodel technology.

The rapid growth of connected devices in the Internet of Things (IoT) is causing an increase in inefficient Application Programming Interfaces (APIs), which results in significant and often overlooked energy consumption. This issue is particularly relevant as the sheer volume of devices and data traffic continues to grow, creating a massive and unsustainable ecological footprint. Our solution presented in this article is based on four holistic API design principles—data minimalism, caching, compression, and intelligent load balancing—to improve communication efficiency. The implementation of these "Green APIs" results in a significant reduction in energy consumption, improves system performance, and contributes to the sustainability of the entire IoT landscape.

BackgroundThis case study assessed the effects of a combined aerobic and resistance training intervention with self-selected exercise intensity in a patient with mixed anxiety and depressive disorder.Case presentationA 24-year-old Brazilian female patient (1.57 m, 62.6 kg), sedentary, with a clinical diagnosis of mixed anxiety and depressive disorder (F41.2 according to ICD-10) for 9 months participated in this case report study. Aerobic exercise and resistance exercise were performed within the same session, 5 days/week for 8 weeks. The physiological demand (heart rate), load, and psychophysiological responses were measured during the training sessions. Training compliance was 64.3% and self-selected exercise intensity was moderate for aerobic exercise and light for resistance exercise. Exercise sessions were subjectively perceived as pleasant and of moderate intensity. There were clinically significant improvements in depressive symptoms (ΔBeck Depression Inventory-II = −15), trace anxiety (ΔState-Trait Anxiety Inventory = −25), sleep quality (ΔPittsburgh Sleep Quality Index = −8), quality of life (ΔWorld Health Organization Quality of Life Questionnaire-Brief: +33.7), and depressive status assessed by a psychiatrist. Aerobic fitness (+7.7%) and muscle strength (+13.5%) improved after the 8 weeks of training.ConclusionCombined aerobic and resistance exercise with self-selected exercise intensity is a promising tool for improving clinical and fitness outcomes in patients with mixed anxiety and depressive disorder. While this case report offers valuable preliminary insights, further research featuring larger samples, control groups, robust study designs, extended intervention durations, and comprehensive follow-up assessments is essential to substantiate these findings.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13256-025-05622-3.

Die Freiburger Zeitschrift für GeschlechterStudien (fzg) bietet ein wissenschaftliches Forum für Fragen der Gender- und Queer Studies, sucht einen breiten interdisziplinären Dialog und versteht sich als Plattform für aktuelle Diskussionen.

Artificial Intelligence (AI) adoption poses complex, multidimensional challenges for organizations that extend beyond technical implementation. While strategic interest in AI is rising, many initiatives struggle to scale sustainably. This paper addresses the fragmented state of AI adoption research by offering a theory-driven synthesis of the key management challenges organizations face. Based on a systematic literature review and grounded in the socio-technical systems (STS) perspective, the study identifies challenges across technological, organizational, and social domains - such as data governance, organizational inertia, skill shortages, and ethical ambiguity - as mutually reinforcing. The resulting framework highlights the systemic entanglement of these barriers, underscoring the limits of isolated interventions. This study contributes to Information Systems (IS) research by conceptualizing AI adoption as a socio-technical transformation and providing an integrative typology of adoption challenges. The findings offer a foundation for future empirical research and guide strategic decision-makers in navigating the organizational complexity of AI integration.

Ensuring safe and realistic interactions between automated driving systems and vulnerable road users (VRUs) in urban environments requires advanced testing methodologies. This paper presents a test environment that combines a Vehicle-in-the-Loop (ViL) test bench with a motion laboratory, demonstrating the feasibility of cyber-physical (CP) testing of vehicle-pedestrian and vehicle-cyclist interactions. Building upon previous work focused on pedestrian localization, we further validate a human pose estimation (HPE) approach through a comparative analysis of real-world (RW) and virtual representations of VRUs. The study examines the perception of full-body motion using a commercial monocular camera-based 3D skeletal detection AI. The virtual scene is generated in Unreal Engine 5, where VRUs are animated in real time and projected onto a screen to stimulate the camera. The proposed stimulation technique ensures the correct perspective, enabling realistic vehicle perception. To assess the accuracy and consistency of HPE across RW and CP domains, we analyze the reliability of detections as well as variations in movement trajectories and joint estimation stability. The validation includes dynamic test scenarios where human avatars, both walking and cycling, are monitored under controlled conditions. Our results show a strong alignment in HPE between RW and CP test conditions for stable motion patterns, while notable inaccuracies persist under dynamic movements and occlusions, particularly for complex cyclist postures. These findings contribute to refining CP testing approaches for evaluating next-generation AI-based vehicle perception and to enhancing interaction models of automated vehicles and VRUs in CP environments.

Tactile hyposensitivity, a sensory processing disorder characterized by the inability to differentiate tactile stimuli, frequently affects people with autism. However, despite the prevalence of the sense of touch in everyday life and the restorative potential of haptic feedback, the application of vibrotactile stimulation in the rehabilitation of tactile hyposensitivity is still underexplored. This study introduces Vibro, a portable companion device that utilizes sensory integration to steadily reduce tactile hyposensitivity over repeated use. Paired with its dedicated mobile application, the device delivers multimodal, integrating visual, audio, and tactile stimuli, and haptic feedback to the patient in the form of points, patterns, textures, music notes, and song frequencies. Facilitated by Vibro’s testing module, which is loosely based on 2-point discrimination testing, initial experimental results from a seven-session patient user test demonstrated a statistically significant improvement in tactile sensitivity. A mean test score increase of 26.85 in 100-point tests was observed over the course of the initial user test, later validated using statistical analysis, suggesting its potential to enhance tactile perception and address tactile hyposensitivity. Future research should focus on upgrading the device to optimize user experience and efficacy while further evaluating the device to validate its impact.

Abstract Background Automatic detection of the apnoea–hypopnoea index (AHI) using a portable monitor (PM) with artificial intelligence (AI) represents a significant challenge. The objective of this study was to examine factors that affect the performance of an AI algorithm that had been previously trained in calculating the AHI with polysomnography (PSG) data using signals collected by a PM. Methods A deep learning model, specifically a version of U‑Net, had been previously trained on a PSG dataset comprising 480 patients for training, 96 for evaluation and 65 for testing. The dataset comprised three variables: oxygen saturation (SpO2), heart rate (HR) and abdominal respiratory effort (ARE). The deep learning model was employed to analyse data collected from 7 patients by the PM, with the objective of classifying obstructive sleep apnoea events and subsequently calculating the AHI. Subsequently, the results obtained by the PM and PSG were compared. Results The present study sought to verify the automatic recognition system for calculation of the AHI. The data for this study were collected from 7 patients using the PM, and the mean absolute error was found to be 19.95% when compared to the results obtained in PSG. The absolute error was found to be greater for patients with a more severe form of sleep apnoea, which resulted in an increase in absolute inaccuracy. Conclusion The findings of this study indicate that training AI models for use with new data obtained from different automatic detection systems is a challenging process that requires careful consideration of various factors, including the quality and quantity of data as well as their preprocessing prior to feeding into the AI models.