Search

Significant investment and development have been made in integrating artificial intelligence (AI) into finance applications. However, the opacity of AI systems raises concerns about essential characteristics needed in sensitive finance applications, such as transparency and accountability. Our study addresses these concerns by investigating a process for analysing explanations generated by AI-based robo-advice systems to comply with the explanation requirements of key EU regulations, including the Markets in Financial Instruments Directive (MiFID) II. We adopt a comprehensive methodology that involves analysing these regulations to identify the specific questions that must be answered by an explanation generated by a robo-advice system to meet legal explainability requirements. Our findings provide a nuanced understanding of which Explainable AI technology may be needed to answer those questions by AI-based robo-advice systems. We demonstrate this through practical case studies on financial advice given by robo-advisers based on AI language-generating models. These case studies highlight our research’s practical utility for the finance industry, which might seek to exploit these new technologies to generate financial advice that meets legal explainability requirements. This study fills a crucial gap in aligning Explainable AI applications in finance with stringent provisions of EU regulations. It provides a practical framework for developers and researchers to ensure their AI innovations advance technology and adhere to legal and ethical standards.

There are many- not just one- periodic laws in chemistry. These laws correspond to non-accidental regularity relations about physical and chemical properties of (sets of) chemical elements. I support this by showing how these regularity relations can be understood from the perspective of a philosophical analysis of laws. Specifically, I show that these relations instantiate standard features associated with laws; they can be spelled out in terms of two standard accounts of laws; and, they can coherently figure in debates about the reality of laws as plausible candidates of ceteris paribus laws.

Human mitochondrial tRNAs (mt-tRNAs), critical for mitochondrial biogenesis, are frequently associated with pathogenic mutations. These mt-tRNAs have unusual sequence motifs and require post-transcriptional modifications to stabilize their fragile structures. However, whether a modification that stabilizes a wild-type (WT) mt-tRNA would also stabilize its pathogenic variants is unknown. Here we show that the N1-methylation of guanosine at position 9 (m1G9) of mt-Leu(UAA), while stabilizing the WT tRNA, has a destabilizing effect on variants associated with MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes). This differential effect is further demonstrated, as removal of the m1G9 methylation, while damaging to the WT tRNA, is beneficial to the major pathogenic variant, improving the structure and activity of the variant. These results have therapeutic implications, suggesting that the N1-methylation of mt-tRNAs at position 9 is a determinant of pathogenicity and that controlling the methylation level is an important modulator of mt-tRNA-associated diseases.

In this paper we present time-dependent perturbations of second-order non-autonomous abstract Cauchy problems associated to a family of operators with constant domain. We make use of the equivalence to a first-order non-autonomous abstract Cauchy problem in a product space, which we elaborate in full detail. As an application we provide a perturbed non-autonomous wave equation.

Evolution equations are partial differential equations (PDEs) that describe evolution over time. To account for random perturbations, random coefficients or noise terms are added, often requiring a numerical solution. The contributions of this thesis are twofold. First, a joint convergence rate is presented for the approximation in randomness, space, and time using polynomial chaos for the random coefficients. Second, convergence rates for the pathwise uniform error in time are obtained for nonlinear stochastic PDEs in the hyperbolic Kato setting., Evolutionsgleichungen sind partielle Differentialgleichungen (PDEs), die die zeitliche Entwicklung beschreiben. Um zufällige Einflüsse zu modellieren, werden zufällige Koeffizienten oder Rauschterme hinzugefügt, die oft eine numerische Lösung erfordern. Im Fall zufälliger Koeffizienten wird eine gemeinsame Konvergenzrate für die Approximation in Zufälligkeit, Raum und Zeit unter Verwendung von polynomiellem Chaos vorgestellt. Für nichtlineare stochastische PDEs werden Konvergenzraten für den pfadweise gleichmäßigen Fehler der Zeitdiskretisierung erzielt.

Cancer immunotherapy has shown high efficacy in many neoplasms and is transforming the treatment of malignant diseases. By detailed analysis of the proteogenomic data of 32 patients across 25 tumor types, non-wild type HLA-binding peptides (potential neoantigens) were identified, and their specific characteristics were analyzed. In silico binding predictions provided a basis for prioritization, and an immunogenicity assessment assay was used to measure neoantigen-dependent interferon-based T-cell responses to screen for possible target candidates., Die Behandlung von Krebserkrankungen durch Immuntherapien verändert gerade grundlegend die bisher etablierten Therapiekonzepte. Anhand der Auswertung der proteo-genomischen Daten von 32 Patienten mit 25 verschiedenen Tumorarten, wurden mutierte MHC-präsentierte Peptide als potentielle Neoantigene identifiziert und charakterisiert. Mithilfe der Simulation ihrer Bindungseigenschaften konnten diese priorisiert und deren Immunogenität durch Interferon-basierte in vitro Stimulationsversuche beurteilt werden.

The increase in Earth observations from space in recent years supports improved quantification of carbon storage by terrestrial vegetation and fosters studies that relate satellite measurements to biomass retrieval algorithms. However, satellite observations are only indirectly related to the carbon stored by vegetation. While ground surveys provide biomass stock measurements to act as reference for training the models, they are sparsely distributed. Here, we addressed this problem by designing an algorithm that harnesses the interplay of satellite observations, modeling frameworks and field measurements, and generated global estimates of above-ground biomass (AGB) density that meet the requirements of the scientific community in terms of accuracy, spatial and temporal resolution. The design was adapted to the amount, type and spatial distribution of satellite data available around the year 2020. The retrieval algorithm estimated AGB annually by merging estimates derived from C- and L-band Synthetic Aperture Radar (SAR) backscatter observations with a Water Cloud type of model and does not rely on AGB reference data at the same spatial scale as the SAR data. This model is integrated with functions relating to forest structural variables that were trained on spaceborne LiDAR observations and sub-national AGB statistics. The yearly estimates of AGB were successively harmonized using a cost function that minimizes spurious fluctuations arising from the moderate-to-weak sensitivity of the SAR backscatter to AGB. The spatial distribution of the AGB estimates was correctly reproduced when the retrieval model was correctly set. Over-predictions occasionally occurred in the low AGB range (<50 Mg ha−1) and under-predictions in the high AGB range (>300 Mg ha−1). These errors were a consequence of sometimes too strong generalizations made within the modeling framework to allow reliable retrieval worldwide at the expense of accuracy. The precision of the estimates was mostly be

The increase in Earth observations from space in recent years supports improved quantification of carbon storage by terrestrial vegetation and fosters studies that relate satellite measurements to biomass retrieval algorithms. However, satellite observations are only indirectly related to the carbon stored by vegetation. While ground surveys provide biomass stock measurements to act as reference for training the models, they are sparsely distributed. Here, we addressed this problem by designing an algorithm that harnesses the interplay of satellite observations, modeling frameworks and field measurements, and generated global estimates of above-ground biomass (AGB) density that meet the requirements of the scientific community in terms of accuracy, spatial and temporal resolution. The design was adapted to the amount, type and spatial distribution of satellite data available around the year 2020. The retrieval algorithm estimated AGB annually by merging estimates derived from C- and L-band Synthetic Aperture Radar (SAR) backscatter observations with a Water Cloud type of model and does not rely on AGB reference data at the same spatial scale as the SAR data. This model is integrated with functions relating to forest structural variables that were trained on spaceborne LiDAR observations and sub-national AGB statistics. The yearly estimates of AGB were successively harmonized using a cost function that minimizes spurious fluctuations arising from the moderate-to-weak sensitivity of the SAR backscatter to AGB. The spatial distribution of the AGB estimates was correctly reproduced when the retrieval model was correctly set. Over-predictions occasionally occurred in the low AGB range (<50 Mg ha−1) and under-predictions in the high AGB range (>300 Mg ha−1). These errors were a consequence of sometimes too strong generalizations made within the modeling framework to allow reliable retrieval worldwide at the expense of accuracy. The precision of the estimates was mostly be

While machine learning (ML) models are increasingly used due to their high predictive power, their use in understanding the data-generating process (DGP) is limited. Understanding the DGP requires insights into feature-target associations, which many ML models cannot directly provide due to their opaque internal mechanisms. Feature importance (FI) methods provide useful insights into the DGP under certain conditions. Since the results of different FI methods have different interpretations, selecting the correct FI method for a concrete use case is crucial and still requires expert knowledge. This paper serves as a comprehensive guide to help understand the different interpretations of global FI methods. Through an extensive review of FI methods and providing new proofs regarding their interpretation, we facilitate a thorough understanding of these methods and formulate concrete recommendations for scientific inference. We conclude by discussing options for FI uncertainty estimation and point to directions for future research aiming at full statistical inference from black-box ML models.

In recent years, neural networks have demonstrated their remarkable ability to discern intricate patterns and relationships from raw data. However, understanding the inner workings of these black box models remains challenging, yet crucial for high-stake decisions. Among the prominent approaches for explaining these black boxes are feature attribution methods, which assign relevance or contribution scores to each input variable for a model prediction. Despite the plethora of proposed techniques, ranging from gradient-based to backpropagation-based methods, a significant debate persists about which method to use. Various evaluation metrics have been proposed to assess the trustworthiness or robustness of their results. However, current research highlights disagreement among state-of-the-art methods in their explanations. Our work addresses this confusion by investigating the explanations’ fundamental and distributional behavior. Additionally, through a comprehensive simulation study, we illustrate the impact of common scaling and encoding techniques on the explanation quality, assess their efficacy across different effect sizes, and demonstrate the origin of inconsistency in rank-based evaluation metrics.

We study abstract sufficient criteria for cost-uniform open-loop stabilizability of linear control systems in a Banach space with a bounded control operator, which build up and generalize a sufficient condition for null-controllability in Banach spaces given by an uncertainty principle and a dissipation estimate. For stabilizability these estimates are only needed for a single spectral parameter and, in particular, their constants do not depend on the growth rate w.r.t. this parameter. Our result unifies and generalizes earlier results obtained in the context of Hilbert spaces. As an application we consider fractional powers of elliptic differential operators with constant coefficients in Lp(Rd) for p∈[1,∞) and thick control sets.

OBJECTIVE: In high-risk patients, prostatic stents may alleviate obstruction at the prostate level. Since 2020 we have used thermo-expandable intraprostatic nitinol stents. Here we document outcomes through the first years with the procedure. MATERIAL AND METHODS: We reviewed patients who had undergone stent treatment between May 2020 and October 2023. Patient and procedural data, urinary symptoms, complications and side effects were recorded. Descriptive statistics were used to summarize outcomes and we evaluated predictors of success and complications using robust multiple regression analyses. RESULTS: We included 52 consecutive patients with a median age of 82 years (range 71-96) and a median Charlson Comorbidity Index of 6 (3-11). Forty-seven men used indwelling catheters, two used clean intermittent catheterization, and three had severe lower urinary tract symptoms. Stents were placed under general anesthesia, sedation, and local anesthesia in 39, 4, and 9 men, respectively. The median treatment time was 14 min (range 8-40). One complication, in the form of an infection requiring IV antibiotics, occurred. Subsequently, 45 men (87%) were able to void spontaneously without bothersome symptoms. After a median of 11 (2-44) months, 8 men had their stents removed due to recurring symptoms. This gives an overall success rate of 37/52 patients (71%). No predictors of success or complications were identified. CONCLUSIONS: Thermo-expandable intraprostatic nitinol stents demonstrate a high success rate with a low risk of complications and may serve as an alternative to permanent or intermittent catheterization for men who are unable or unwilling to undergo flow-improving surgery.

Counterfactual explanations elucidate algorithmic decisions by pointing to scenarios that would have led to an alternative, desired outcome. Giving insight into the model’s behavior, they hint users towards possible actions and give grounds for contesting decisions. As a crucial factor in achieving these goals, counterfactuals must be plausible, i.e., describing realistic alternative scenarios within the data manifold. This paper leverages a recently developed generative modeling technique – adversarial random forests (ARFs) – to efficiently generate plausible counterfactuals in a model-agnostic way. ARFs can serve as a plausibility measure or directly generate counterfactual explanations. Our ARFbased approach surpasses the limitations of existing methods that aim to generate plausible counterfactual explanations: It is easy to train and computationally highly efficient, handles continuous and categorical data naturally, and allows integrating additional desiderata such as sparsity in a straightforward manner., Counterfactual explanations elucidate algorithmic decisions by pointing to scenarios that would have led to an alternative, desired outcome. Giving insight into the model’s behavior, they hint users towards possible actions and give grounds for contesting decisions. As a crucial factor in achieving these goals, counterfactuals must be plausible, i.e., describing realistic alternative scenarios within the data manifold. This paper leverages a recently developed generative modeling technique – adversarial random forests (ARFs) – to efficiently generate plausible counterfactuals in a model-agnostic way. ARFs can serve as a plausibility measure or directly generate counterfactual explanations. Our ARF-based approach surpasses the limitations of existing methods that aim to generate plausible counterfactual explanations: It is easy to train and computationally highly efficient, handles continuous and categorical data naturally, and allows integrating additional desiderata such as sparsity in a straightforward manner.

Der Pflegenotstand ist vor allem seit Beginn der COVID-19-Pandemie so präsent wie noch nie. Diese Entwicklung führt zu einem wachsenden Bedarf an Pflegekräften und an einer gut funktionierenden Pflegeinfrastruktur. Verschiedene Faktoren, darunter der demografische Wandel sowie eine abnehmende Attraktivität des Pflegeberufs aufgrund hoher Arbeitsbelastung im Alltag und schlechter Bezahlung, verstärken den Mangel an Pflegepersonal zusätzlich. Die vorliegende Situation macht deutlich, dass im Pflegebereich praktikable und umfassende Lösungen gefunden werden müssen. Ein Lösungsansatz hierfür ist der Einsatz von Methoden des Business Analytics und Operations Management. Diese sollen dazu beitragen, das Pflegepersonal zu entlasten und die Attraktivität des Pflegeberufs wieder zu erhöhen. Dieser Beitrag gibt einen Überblick über Business Analytics und Operations Management in der Klinik. Durch kleine Fallbeispiele der Pflegeprozesse sowie Optimierungs- und Simulationsmodelle soll aufgezeigt werden, dass diese Methoden schnell und einfach im Pflegebereich Unterstützung bieten können.

While machine learning (ML) models are increasingly used due to their high predictive power, their use in understanding the data-generating process (DGP) is limited. Understanding the DGP requires insights into feature-target associations, which many ML models cannot directly provide due to their opaque internal mechanisms. Feature importance (FI) methods provide useful insights into the DGP under certain conditions. Since the results of different FI methods have different interpretations, selecting the correct FI method for a concrete use case is crucial and still requires expert knowledge. This paper serves as a comprehensive guide to help understand the different interpretations of global FI methods. Through an extensive review of FI methods and providing new proofs regarding their interpretation, we facilitate a thorough understanding of these methods and formulate concrete recommendations for scientific inference. We conclude by discussing options for FI uncertainty estimation and point to directions for future research aiming at full statistical inference from black-box ML models.

OBJECTIVE: In high-risk patients, prostatic stents may alleviate obstruction at the prostate level. Since 2020 we have used thermo-expandable intraprostatic nitinol stents. Here we document outcomes through the first years with the procedure. MATERIAL AND METHODS: We reviewed patients who had undergone stent treatment between May 2020 and October 2023. Patient and procedural data, urinary symptoms, complications and side effects were recorded. Descriptive statistics were used to summarize outcomes and we evaluated predictors of success and complications using robust multiple regression analyses. RESULTS: We included 52 consecutive patients with a median age of 82 years (range 71-96) and a median Charlson Comorbidity Index of 6 (3-11). Forty-seven men used indwelling catheters, two used clean intermittent catheterization, and three had severe lower urinary tract symptoms. Stents were placed under general anesthesia, sedation, and local anesthesia in 39, 4, and 9 men, respectively. The median treatment time was 14 min (range 8-40). One complication, in the form of an infection requiring IV antibiotics, occurred. Subsequently, 45 men (87%) were able to void spontaneously without bothersome symptoms. After a median of 11 (2-44) months, 8 men had their stents removed due to recurring symptoms. This gives an overall success rate of 37/52 patients (71%). No predictors of success or complications were identified. CONCLUSIONS: Thermo-expandable intraprostatic nitinol stents demonstrate a high success rate with a low risk of complications and may serve as an alternative to permanent or intermittent catheterization for men who are unable or unwilling to undergo flow-improving surgery.

no abstract

The increase in Earth observations from space in recent years supports improved quantification of carbon storage by terrestrial vegetation and fosters studies that relate satellite measurements to biomass retrieval algorithms. However, satellite observations are only indirectly related to the carbon stored by vegetation. While ground surveys provide biomass stock measurements to act as reference for training the models, they are sparsely distributed. Here, we addressed this problem by designing an algorithm that harnesses the interplay of satellite observations, modeling frameworks and field measurements, and generated global estimates of above-ground biomass (AGB) density that meet the requirements of the scientific community in terms of accuracy, spatial and temporal resolution. The design was adapted to the amount, type and spatial distribution of satellite data available around the year 2020. The retrieval algorithm estimated AGB annually by merging estimates derived from C- and L-band Synthetic Aperture Radar (SAR) backscatter observations with a Water Cloud type of model and does not rely on AGB reference data at the same spatial scale as the SAR data. This model is integrated with functions relating to forest structural variables that were trained on spaceborne LiDAR observations and sub-national AGB statistics. The yearly estimates of AGB were successively harmonized using a cost function that minimizes spurious fluctuations arising from the moderate-to-weak sensitivity of the SAR backscatter to AGB. The spatial distribution of the AGB estimates was correctly reproduced when the retrieval model was correctly set. Over-predictions occasionally occurred in the low AGB range (<50 Mg ha−1) and under-predictions in the high AGB range (>300 Mg ha−1). These errors were a consequence of sometimes too strong generalizations made within the modeling framework to allow reliable retrieval worldwide at the expense of accuracy. The precision of the estimates was mostly be

Nature-based solutions (NBS) are recognised as contributing to international climate and biodiversity goals as well as solving other major societal challenges. Still, there are barriers to using NBS in the Nordic countries. Well-designed policies and targeted policy instruments are needed to accelerate the uptake of NBS. The A-DVICE project has developed an online handbook featuring advice for mainstreaming the use of NBS through policies by exploring potential opportunities and examples in the Nordic countries. The project is part of the Nordic Council of Ministers' programme on NBS. The handbook is hosted online, at www.nbspolicy.org. This TemaNord report contains a compilation of its main content.

Thermomechanical fatigue is a fatigue failure caused by combined thermal and mechanical loading cycles where both stresses and temperature can vary with time. A cohesive zone model simulates material failure within the finite element method by describing material behaviour and damage evolution through a traction-separation relation. In this work, a viscoplastic cohesive zone model is developed in order to describe the fatigue and fracture behaviour of high performance superalloys undergoing thermomechanical fatigue loading. A micromechanically motivated cohesive potential, incorporating fatigue and creep damage variables, is proposed and its thermodynamic consistency is established. The characteristic behaviour of the cohesive law during cyclic loading is described in terms of both traction and energy. The developed cohesive zone model is applied to simulate the lifetime behaviour in a smooth specimen. The effect of loading parameters such as temperature, dwell period, and strain rate on isothermal, in-phase and out-of-phase thermomechanical fatigue is successfully demonstrated. Using crack growth simulations in a boundary layer model, the relation between the local behaviour of the cohesive zone and fracture parameters like stress intensity factor, J-integral and crack tip opening displacement is established. The three stages of fatigue crack growth rate curves that are experimentally observed were obtained through simulations. Crack growth in a corner crack specimen is simulated under cyclic stress-controlled loading to obtain the experimentally observed crack profiles. Material parameters for the Nickel based superalloy MAR-M247 are identified using experimental data from the literature.

Study design Retrospective cohort study. Objective Bracing is an accepted standard therapy for idiopathic scoliosis at Cobb angle ranges between 25° and 40°. However, it is unclear, if a specifically tailored regimen of daytime and nighttime braces (= double brace) yields superior results compared to the standard treatment (single brace for day and night). Methods One-hundred-fifteen patients with adolescent idiopathic scoliosis (AIS) were assessed before initiation of bracing treatment and at the final follow-up 2 years after deposition of the brace. They were divided into two groups: double-brace group (n = 66, 4 male, 62 female, age 13.1 ± 1.9 (mean ± SD), primary curvature thoracic n = 35, lumbar n = 31) and single-brace group (n = 49, 8 male, 41 female, age 14.1 ± 1.9, primary curvature thoracic n = 18, lumbar n = 31). Each patient underwent clinical and radiological examinations and Cobb angles were measured. Results Both therapy regimens succeeded to either stop progression or improve scoliosis in over 85% of cases. The nighttime brace showed a significantly higher primary correction than the daytime brace. Nevertheless, there was no significant difference in treatment success in the 2-year follow-up (p = 0.58). Conclusion It seems to be sufficient to treat idiopathic scoliosis with one well-tailored brace for day- and nighttime.

Psoriasis results from both genetic predisposition and environmental triggers, such as Streptococcal infections. This study aimed to explore the correlation between the abundance of the Streptococcus genus on the skin and psoriasis severity in individuals carrying specific psoriasis -associated genetic variants. Studying 39 chronic plaque psoriasis patients, the elbow skin microbiome and 49 psoriasis -related single nucleotide polymorphisms (SNPs) were analysed using a MiSeq instrument for 16S rDNA sequencing, and CLC Genomic Workbench for processing and analysis. Through multivariate linear regression analysis, a positive correlation was found between Streptococcus genus abundance and psoriasis severity in patients with certain FBXL19 gene -related heterozygous SNPs (rs12924903, rs10782001, rs12445568). Conversely, a negative association was observed in patients with homozygous genotypes. Moreover, we identified an association between Streptococcus abundance and psoriasis severity in patients with genetic variants related to IL -22, ERAP1, NOS2, and ILF3. This is the first study highlighting a positive association between Streptococcus skin colonization and psoriasis severity in patients with heterozygous genotypes within the FBXL19 gene region. FXBL19 targets the IL-33/IL1RL1 axis, crucial in infectious diseases and innate immunity promotion. These novel results suggests an intricate interaction among host genetics, Streptococcus skin colonization, and psoriasis inflammation, offering potential avenues for novel treatment approaches., Funding Agencies|Swedish Psoriasis Foundation; Futurum, The Academy of Healthcare, County Council of Jonkoping, Sweden

BACKGROUND AND OBJECTIVE Follow-up for patients with testicular cancer should ensure early detection of relapses. Optimal schedules and minimum requirements for cross-sectional imaging are not clearly defined, and guideline recommendations differ. Our aim was to analyse the clinical impact of different imaging modalities for detection of relapse in a large prospective cohort (Swiss Austrian German Testicular Cancer Cohort Study, SAG TCCS). METHODS Patients with seminoma or nonseminoma were prospectively enrolled between January 2014 and February 2023 after initial treatment (n = 1175). Follow-up according to the study schedule was individualised for histology and disease stage. Only patients who had received primary treatment were considered. We analysed the total number of imaging modalities and scans identifying relapse and the timing of relapse. KEY FINDINGS AND LIMITATIONS We analysed data for 1006 patients (64% seminoma, 36% nonseminoma); 76% had stage I disease. Active surveillance was the most frequent management strategy (65%). Recurrence occurred in 82 patients, corresponding to a 5-yr relapse-free survival rate of 90.1% (95% confidence interval 87.7-92.1%). Median follow-up for patients without relapse was 38.4 mo (interquartile range 21.6-61.0). Cross-sectional imaging of the abdomen was the most important indicator of relapse 57%, abdominal CT accounting for 46% and MRI for 11%. Marker elevation indicated relapse in 24% of cases. Chest X-ray was the least useful modality, indicating relapse in just 2% of cases. CONCLUSIONS AND CLINICAL IMPLICATIONS On the basis of findings from our prospective register, we have adapted a follow-up schedules with an emphasis on abdominal imaging and a reduction in chest X-rays. This schedule might provide additional guidance for clinicians and will be prospectively evaluated as SAG TCCS continues to enrol patients. PATIENT SUMMARY We analysed the value of different types of imaging scans for detection of relapse of testic

AIMS Low-grade non-intestinal-type sinonasal adenocarcinoma (LGSNAC) is a rare heterogeneous and poorly characterised group of tumours, distinct from intestinal- and salivary-type neoplasms. Therefore, further characterisation is needed for clearer biological understanding and classification. METHODS AND RESULTS Clinical, histological and molecular characterisation of four cases of biphasic, low-grade adenocarcinomas of the sinonasal tract was performed. All patients were male, aged between 48 and 78 years, who presented with polypoid masses in the nasal cavity. Microscopically, virtually all tumours were dominated by tubulo-glandular biphasic patterns, microcystic, focal (micro)papillary, oncocytic or basaloid features. Immunohistochemical staining confirmed biphasic differentiation with an outer layer of myoepithelial cells. Molecular profiling revealed HRAS (p.G13R, p.Q61R) mutations, and concomitant AKT1 (p.E17K, p.Q79R) mutations in two cases. Two cases showed potential in-situ/precursor lesions adjacent to the tumour. Follow-up periods ranged from 1 to 30 months, with one case relapsing locally after 12 and > 20 years. CONCLUSION This study further corroborates a distinct biphasic low-grade neoplasm of the sinonasal tract with seromucinous differentiation. Although morphological and molecular features overlap with salivary gland epithelial-myoepithelial carcinoma, several arguments favour categorising these tumours within the spectrum of LGSNAC.

Plastics are ubiquitous materials in our daily lives, but their inadequate disposal has led to the widespread distribution of their micro- and nanoparticles in various ecosystems. Their detection in feed and food, as well as in livestock and human stool samples, strongly suggests a continuous circulation in the feed and food chain. The ability of plastic particles to penetrate the intestinal barrier determines their accumulation in the body and in food of animal origin. The gastrointestinal (GI) microbiome, whose fermentation activity can influence the particle size distribution of certain plastic materials and which is known to modulate the permeability of the intestinal barrier, may be a critical hub in this transfer. This review attempts to summarise research efforts to date on the interaction between microplastics (MPs) and the GI microbiome of humans, mice, chickens and aquatic animals. We have analysed the state of knowledge and identified future avenues for targeted research approaches to answer open questions regarding the interaction of plastic particles with the GI microbiome, which may help to develop predictive models for the accumulation of plastic particles from feed and food in the body and animal products, respectively.

BACKGROUND: Historically, patient selection for peptide receptor radionuclide therapy (PRRT) has been performed by virtue of somatostatin receptor scintigraphy (SRS). In recent years, somatostatin receptor positron emission tomography (SSTR-PET) has gradually replaced SRS because of its improved diagnostic capacity, creating an unmet need for SSTR-PET-based selection criteria for PRRT. Tumor-to-blood ratio (TBR) measurements have shown high correlation with the net influx rate Ki, reflecting the tumor somatostatin receptor expression, to a higher degree than standardized uptake value (SUV) measurements. TBR may therefore predict treatment response to PRRT. In addition, changes in semiquantitative SSTR-PET parameters have been shown to predate morphological changes, making them a suitable metric for response assessment. METHODS: The institutional database of the Department of Nuclear Medicine (University Hospital Essen) was searched for NET patients undergoing ≥ 2 PRRT cycles with available baseline and follow-up SSTR-PET. Two blinded independent readers reported the occurrence of new lesions quantified tumor uptake of up to nine lesions per patient using SUV and TBR. The association between baseline TBR and changes in uptake/occurrence of new lesions with progression-free survival (PFS) and overall survival (OS) was tested by use of a Cox regression model and log-rank test. RESULTS: Patients with baseline TBR in the 1st quartile had a shorter PFS (14.4 months) than those in the 3rd (23.7 months; p = 0.03) and 4th (24.1 months; p = 0.02) quartile. Similarly, these patients had significantly shorter OS (32.5 months) than those with baseline TBR in the 2nd (41.8 months; p = 0.03), 3rd (69.2 months; p < 0.01), and 4th (42.7 months; p = 0.03) quartile. Baseline to follow-up increases in TBR were independently associated with shorter PFS when accounting for prognostic markers, e.g., RECIST response (hazard ratio = 2.91 [95%CI = 1.54-5.50]; p = 0.01). This was confirmed

Understanding agricultural land systems is paramount to preparing for future transitions under climate change. A first step in moving towards a systematic, multi-context comparison of agricultural land systems is an assessment of data availability and commensurability. We compare publicly available and publicly curated data describing agricultural land systems—those devoted to the production of food, fuel, and fiber—to map analytical potential for comparative research in the United States (US) and European Union (EU). We discuss how political priorities shape what we can(not) see in each region and articulate future data needs to support cross-context understanding of agricultural land systems dynamics. This comprehensive understanding of how and why agricultural land systems change is imperative to promoting sustainable, resilient, and just agriculture futures.

Die vorliegende Arbeit findet in sehr frühgeborenen Erwachsenen verkleinerte Volumina von Hypothalamuskernen, die an der Regulation des Körpergewichtes beteiligt sind. Veränderte Körpergewichtsentwicklungen nach Frühgeburt korrelieren mit Volumenveränderungen in relevanten Hypothalamuskernen im Erwachsenenalter., The present work demonstrates lower volumes of hypothalamus subunits involved in body weight control in very premature born adults. Altered body weight trajectories after premature birth correlate with volumetric changes of relevant hypothalamus nuclei in adulthood.

This thesis explores advancements in computational imaging methods that enable lensless, phase-sensitive microscopy, wavefront sensing, and metrology applications. By using overlapping regions illuminated by a probe beam, interference pattern measurements allow for computational reconstruction of objects through iterative algorithms. The focus is on enhancing these reconstructions using differentiable modeling of the underlying physics. This approach allows for the joint optimization of various parameters and the integration of deep learning techniques. The goal is to increase the flexibility and robustness of computational imaging for various applications, particularly under challenging experimental conditions. The significance is particularly evident in the semiconductor industry, where nanoscale imaging is critical due to shrinking feature sizes in computer chips. With transistor dimensions approaching a few nanometers, traditional metrology faces significant challenges. Extreme Ultraviolet (EUV) light, with wavelengths around 13.5 nm, is used in semiconductor manufacturing but is absorbed or poorly refracted by most materials, making lenses impractical. Computational lensless imaging provides high-resolution, non-destructive imaging of nanoscale structures on semiconductor wafers, addressing a key metrology need.

Computational imaging is increasingly vital for a broad spectrum of applications, ranging from biological to material sciences. This includes applications where the object is known and sufficiently sparse, allowing it to be described with a reduced number of parameters. When no explicit parameterization is available, a deep generative model can be trained to represent an object in a low-dimensional latent space. In this paper, we harness this dimensionality reduction capability of autoencoders to search for the object solution within the latent space rather than the object space. We demonstrate what we believe to be a novel approach to ptychographic image reconstruction by integrating a deep generative model obtained from a pre-trained autoencoder within an automatic differentiation ptychography (ADP) framework. This approach enables the retrieval of objects from highly ill-posed diffraction patterns, offering an effective method for noise-robust latent vector reconstruction in ptychography. Moreover, the mapping into a low-dimensional latent space allows us to visualize the optimization landscape, which provides insight into the convexity and convergence behavior of the inverse problem. With this work, we aim to facilitate new applications for sparse computational imaging such as when low radiation doses or rapid reconstructions are essential.

Despite increasing concerns about climate change, many people struggle to translate their pro-environmental values into action. Here, we explore the origins of this value-action gap focusing on the role of behavioral barriers that are characteristic for many environmentally relevant decisions. Using incentivized online laboratory and vignette experiments, we find that individuals are less likely to behave in accordance with their environmental values if they are not immediately affected by the consequences of their actions, if the impacts of their actions are uncertain, and if they contribute only marginally to the outcome. The behavioral barriers also exert an indirect effect on environmental behavior by undermining reciprocity and positive peer effects. The value-action gaps are correlated across the different experimental settings with men and younger participants showing a particularly large gap. These insights are important for climate communication and education, highlighting the important role of the perceived relevance and potential consequences of personal behaviors in collective action.

Ptychographic extreme ultraviolet (EUV) diffractive imaging has emerged as a promising candidate for the next generationmetrology solutions in the semiconductor industry, as it can image wafer samples in reflection geometry at the nanoscale. This technique has surged attention recently, owing to the significant progress in high-harmonic generation (HHG) EUV sources and advancements in both hardware and software for computation. In this study, a novel algorithm is introduced and tested, which enables wavelength-multiplexed reconstruction that enhances the measurement throughput and introduces data diversity, allowing the accurate characterisation of sample structures. To tackle the inherent instabilities of the HHG source, a modal approach was adopted, which represents the cross-density function of the illumination by a series of mutually incoherent and independent spatial modes. The proposed algorithm was implemented on a mainstream machine learning platform, which leverages automatic differentiation to manage the drastic growth in model complexity and expedites the computation using GPU acceleration. By optimising over 200 million parameters, we demonstrate the algorithm's capacity to accommodate experimental uncertainties and achieve a resolution approaching the diffraction limit in reflection geometry. The reconstruction of wafer samples with 20-nm high patterned gold structures on a silicon substrate highlights our ability to handle complex physical interrelations involving a multitude of parameters. These results establish ptychography as an efficient and accurate metrology tool.

Purpose: The European Medicines Agency's (EMA) Pharmacovigilance Risk Assessment Committee (PRAC) launched a strategy to examine the public health impact of major regulatory interventions aimed at minimising risks of medicinal products. We conducted a lessons learnt analysis of impact studies completed between 2015 and 2023. Methods: We surveyed PRAC Sponsors and (Co-)Rapporteurs involved in the evaluation of 12 impact studies (10 commissioned by EMA and 2 conducted collaboratively by Member States) to explore how these support regulatory decision-making. Questions covered achievement of study objectives, risk minimisation effectiveness, added value for regulatory decision-making, and recommendations for future impact studies. Themes were generated using thematic content analysis. Results: Survey responses from 15 PRAC Sponsors and (Co-)Rapporteurs from 10 European Union Member States were included in the analysis. Among four cross-sectional surveys and eight drug utilisation studies, 50% achieved all objectives, the other studies partially due to limitations. Two studies concluded that risk minimisation measures were overall effective, two were effective with variation across countries, two were partially effective and four studies showed limited effectiveness. Two studies were deemed inconclusive due to limitations. The reasons for the limited effectiveness of risk minimisation may be explored using mixed-method approaches. Assessment of study feasibility and a priori discussion of effectiveness measurements is important. Conclusion: Despite limitations, impact research adds value to regulatory decision-making by addressing knowledge gaps and providing additional information on unintended consequences of regulatory interventions. Our recommendations will help to improve planning, conducting and interpretating future impact studies.

Both attributes of functional traits and phylogenetic diversity influence ecosystem functions, but which of these factors is most important is still poorly understood in natural systems. Using data from West African forests and tree savannas, we analyse how (i) phylogenetic diversity complements attributes of functional traits in explaining aboveground carbon (AGC); (ii) phylogenetic diversity relates with attributes of functional traits along gradients of phylogenetic signal; and (iii) pathways between phylogenetic diversity and attributes of functional traits relate AGC to soil and climate. Phylogenetic diversity was measured as standardised effect size of Mean Pairwise Distance (sesMPD) and Mean Nearest Taxon Distance (sesMNTD). Functional dispersion (FDis) and community weighted mean (CWM) were calculated for four traits related to leaf economics spectrum and plant life-history. Functional traits-based models explained 11 % of AGC variability. With two out of the four traits being phylogenetically conserved, incorporating phylogenetic diversity in the models increased the explained variance in AGC by 15 %. The slope of phylogenetic diversity-trait relationship was more responsive to trait conservatism for FDis than CWM. AGC was positively influenced by sesMPD and CWM of plant maximum height. In turn, CWM of plant maximum height increased with higher soil nitrogen and climate moisture, whereas sesMPD was negatively related with climate moisture. Although FDis was positively associated with sesMPD, it was not as important as sesMPD and CWM of plant maximum height in influencing and relating AGC to soil nitrogen and climate moisture. Our results suggest that phylogenetic diversity is important for AGC but does not fully reflect the functional mechanisms pertaining to community-wide trait means. The study also demonstrates the role of environment in regulating AGC, which operates through differences in community fitness driven by tall plant stature, and evolutionary