Heilbronn University

Abstract Robust, tightly localized photonic modes are essential for scalable on-chip quantum information processing, including multi-plane (3D-integrated) photonic platforms in which multiple planar waveguide layers are vertically stacked and coupled. Here, we present a comprehensive study of three three-dimensional fractal photonic architectures, 3D Cantor dust, Vicsek fractal, and Sierpinski sponge, each discretized into dielectric voxels and modeled in a voxel-graph tight-binding surrogate with exponentially decaying evanescent-hopping couplings. We examine two imperfection classes: on-site refractive-index disorder (amplitude up to 10%) and stochastic voxel deletions (defect rate up to 2%). In a classical eigenanalysis (over 10 000 parameter combinations of waveguide width, layer thickness, baseline coupling, decay constant, and defect rate) we compute the central spectral spacing $$\Delta \lambda $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>Δ</mml:mi> <mml:mi>λ</mml:mi> </mml:mrow> </mml:math> (a finite-structure stopband proxy) and the inverse participation ratio, and demonstrate via ANOVA/Tukey HSD (adjusted $$p\ll 10^{-3}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>p</mml:mi> <mml:mo>≪</mml:mo> <mml:msup> <mml:mn>10</mml:mn> <mml:mrow> <mml:mo>-</mml:mo> <mml:mn>3</mml:mn> </mml:mrow> </mml:msup> </mml:mrow> </mml:math> ) that 3D Cantor dust yields the largest mean central spacing and highest inverse participation ratio across depths and parameter ranges. Because the canonical Cantor-dust voxelization is disconnected under the unit-neighborhood coupling rule used in this paper, its high IPR should be interpreted as isolation-dominated localization within this surrogate. We additionally map the tight-binding coupling graphs to 2-local qubit Hamiltonians in the single-excitation manifold (one qubit per retained site), and we validate a variational quantum eigensolver workflow on a 2-qubit SSH dimer and on 4–8-qubit fractal subgraphs; the qubit counts, optimization objective, shot-based measurement model, and resource scaling with fractal iteration depth are stated explicitly. Our work positions 3D Cantor dust as a favorable multi-scale platform for graph-spectral stopband engineering and robust mode localization in tight-binding surrogates, motivating targeted full-wave electromagnetic validation and device-level studies of planar-mask implementations (stacked 2D fractal layers with vertical couplers) for 3D-integrated on-chip deployment.

Daylight photodynamic therapy is currently the preferred treatment for actinic keratosis due to its greater effectiveness and a lower level of pain compared to conventional photodynamic therapy. To enhance the effectiveness of the therapy, especially for hyperkeratotic forms, various pre-treatments are currently available, primarily fractional laser therapy. Until now, there have been few studies investigating skin physiology after photodynamic therapy and fractional laser treatment. The aim of the study was to investigate the skin physiological changes in daylight photodynamic therapy with and without fractional laser pre-treatment. Thirty patients with actinic keratoses were recruited at the Clinic for Dermatology, SLK Clinics Heilbronn, Germany. On the ventral right or left forearm of the patients, four fields of 3 x 3 cm each were randomly assigned to: daylight, fractional laser with daylight, daylight photodynamic therapy with 10% 5-aminolevulinic acid hydrochloride gel, and daylight photodynamic therapy with fractional laser with 10% 5-aminolevulinic acid hydrochloride gel. Redness, corneometry, and transepidermal water loss were measured before treatment, 30 minutes, 24 hours, seven days and 14 days after treatment. Compared to the effects of daylight photodynamic therapy alone, daylight photodynamic therapy with fractional laser caused more pronounced skin changes, with the increase persisting for up to 24 hours for transepidermal water loss values and up to 14 days for skin redness and dryness. Additional fractional laser was shown to intensify inflammatory processes and damage the skin barrier. This enhanced biological effect could also lead to an increased clinical effect on actinic keratoses.

BackgroundPatient education materials (PEMs) found online are often written at a complexity level too high for the average reader, which can hinder understanding and informed decision-making. Large language models (LLMs) may offer a solution by simplifying complex medical texts. To date, little is known about how well LLMs can handle simplification tasks for German-language PEMs.ObjectiveThe study aims to investigate whether LLMs can increase the readability of German online medical texts to a recommended level.MethodsA sample of 60 German texts originating from online medical resources was compiled. To improve the readability of these texts, four LLMs were selected and used for text simplification: ChatGPT-3.5, ChatGPT-4o, Microsoft Copilot, and Le Chat. Next, readability scores (Flesch reading ease [FRE] and Wiener Sachtextformel [4th Vienna Formula; WSTF]) of the original texts were computed and compared to the rephrased LLM versions. A Student t test for paired samples was used to test the reduction of readability scores, ideally to or lower than the eighth grade level.ResultsMost of the original texts were rated as difficult to quite difficult (average WSTF 11.24, SD 1.29; FRE 35.92, SD 7.64). On average, the LLMs achieved the following average scores: ChatGPT-3.5 (WSTF 9.96, SD 1.52; FRE 45.04, SD 8.62), ChatGPT-4o (WSTF 10.6, SD 1.37; FRE 39.23, SD 7.45), Microsoft Copilot (WSTF 8.99, SD 1.10; FRE 49.0, SD 6.51), and Le Chat (WSTF 11.71, SD 1.47; FRE 33.72, SD 8.58). ChatGPT-3.5, ChatGPT-40, and Microsoft Copilot showed a statistically significant improvement in readability. However, the t tests yielded no statistically significant results for the reduction of scores lower than the eighth grade level.ConclusionsLLMs can improve the readability of PEMs in German. This moderate improvement can support patients reading PEMs online. LLMs demonstrated their potential to make complex online medical text more accessible to a broader audience by increasing readability. This is the first study to evaluate this for German online medical texts.

The emergence of software-defined vehicles (SDVs) presents a significant challenge for the role of software in managing complex automotive systems, complicating maintenance and upgrades. The software interfaces between components and systems are key throughout the vehicle which enable communication and allow for data flows in SDVs. As of now, the interfaces in the vehicle are stiff and mostly based on the AUTOSAR standard. Service-oriented architecture (SOA), proven in enterprise and consumer computer systems, is now adopted in automotive to address this complexity. This literature review assesses the current state of SOA in the automotive industry, highlighting solutions to solve the interface challenges. The static interface definitions in AUTOSAR limit adaptability, while nonstandardized approaches offer greater flexibility. We identify a gap in how service interfaces interact with internal architecture and external non-SOA components, hindering seamless software-hardware integration. Realizing SOA's potential requires standardized interface patterns and hardware abstraction layers to integrate services and components better.

Abstract The assembly of highly accurate metal car bodies poses a huge challenge to automotive industry involving diverse thermal and mechanical joining processes. In practice, the complexity of the individual steps, as well as their interaction, hardly admit a precise physical modelling and, frequently, uncertainties regarding the actual geometry of tools, varying part quality and the high sensitivity of the assembly outcome to process parameters render expensive simulation results useless. As the production routines in the shop floor require to deal with errors of the assembled body in an ad-hoc manner, clamping tools are designed to be adjustable, allowing for relative repositioning of body parts and the induction of tension before joining. Usually, a costly optimization procedure of alternating quality measurements and suitable adjustments of the clamping tools is employed by highly experienced staff, a procedure which often is error-prone or suboptimal, though. This article presents, first, a novel approach to virtually place the individual CAD parts on the assembly scan by a weighted and constrained best fit algorithm; thus, delivering compensating rigid body transformations for each individual body part. Second, these transformations are used to automatically determine adjustment settings for the degrees of freedom available in the clamping tools that enforce the desired part repositionings. The approach is investigated in an iterative optimization procedure of measurements and tool adjustments where the real assembly process is replaced by FE simulation. Several practically relevant examples demonstrate a significant quality improvement of the resulting overall assemblies.

Abstract The term ‘freshly roasted’ has long been regarded as a hallmark of coffee quality. This study challenges this assumption by investigating whether (i) a discernible sensory effect and (ii) a perceptible consumer preference exist between ‘freshly roasted’ (1 day post-roast) and ‘post-roast matured’ (145 days) coffee beans. Employing a triangle test (ISO 4120:2021) and a hedonic preference test (ISO 8587:2013) with 60 untrained participants attending coffee-related events, 70% of participants successfully discriminated between samples ( p &lt; 0.001). The preference test revealed no significant difference in the proportion of participants selecting either sample (52.4% mature vs. 47.6% fresh, p = 0.875). These findings underscore the potential for a more nuanced approach to post-roast coffee maturation, demonstrating that extended maturation produces perceivable sensory changes without systematic preference disadvantage.

Wearable sensor patches for a non-invasive detection of biomarkers and physiological parameters constitute a rapidly growing field in the diagnosis of various prevalent diseases and medical treatment. In that regard, recent developments in future-oriented printing technologies have opened new avenues for an advanced fabrication of functional sensor structures on skin-conformable and biocompatible substrates. The present work aims at a novel technological route for the highly precise and equally cost-efficient patterning of conductive sensor elements by electrohydrodynamic (EHD) printing as easy-to-use approach for the realization of customized biosensing patches. To this end, this electric field-induced method is studied in-depth by FEM simulations for the derivation and optimization of appropriate process parameters allow for a stable jetting regime. The numerical results are directly utilized for the fabrication of wearable, conductive sensor patterns by deposition of an Ag nanoparticle ink on a new type of heat-resistant thermoset substrates.

ABSTRACT:Current legislative frameworks reflect a societal consensus to prioritize sustainability, incentivizing industries to integrate environmental goals into strategic objectives. Embedding sustainability into product development requires appropriate methods and tools. Technological advancements enable the utilization and analysis of operational machine data to support the development of new generations of sustainable systems and the conduction of Life-Cycle-Assessments. This research presents a method to support data-driven product development to reduce the environmental impact of new product generations of complex mechatronic systems during operation, addressing key factors such as the technical system, organizational infrastructure, and regulations. The application of the method resulted in multiple proposed design changes able to enhance machine sustainability and operational efficiency.

In the classical problem formulation of Text-to-SQL in academia, each question is translated independently of the others into SQL. This differs from the setting in practice, where questions enter the Text-to-SQL system in sequence. Thus, for all but the first few questions, a translation history is available that contains past questions and how they were translated by the system. So far, it has not been sufficiently explored how Text-to-SQL systems can make use of the translation history to improve future translations. Another crucial difference from the academic setting is that in practice, users generally have a conversation with the Text-to-SQL system. More concretely, if the initial translation contains a mistake, the user can follow up with feedback messages to allow the system to fix the mistake. In this case, it might be helpful to remember these user feedback messages to avoid repeating past mistakes. Thus, in this paper, we explore how a history of such past conversations between users and the Text-to-SQL system can be used to make future Text-to-SQL translations more accurate. We explore several approaches for extracting relevant experiences and insights from this conversation history and show in an evaluation that utilizing them can improve translation accuracy by up to 14.9%.