Current conventional methods of evaluating microstructures are characterized by a high degree of subjectivity and a lack of reproducibility. Modern machine learning (ML) approaches have already shown great potential in overcoming these challenges. Once trained with representative data in combination with an objective ground truth, the ML model is able to perform a task properly in a reproducible and automated manner. However, in highly complex use cases, it is often not possible to create a definite ground truth. This study addresses this problem using the underlying showcase of microstructures of highly complex quenched and quenched and tempered (Q/QT) steels. A patch-wise classification approach combined with a sliding window technique provides a solution for segmenting entire microphotographs where pixel-wise segmentation is not applicable due to the problem of reproducibly creating unambiguous training masks. Using correlative microscopy, consisting of light optical microscope (LOM) and scanning electron microscope (SEM) micrographs as well as corresponding data from electron backscatter diffraction (EBSD), a training dataset of reference states that covers a wide range of microstructures was acquired in order to train accurate and robust ML models. Despite the enormous complexity associated with the steels treated here, classification accuracies of 88.8% in the case of LOM images and 93.7% for high-resolution SEM images were achieved. These high accuracies are close to super-human performance, especially in consideration of the reproducibility of the automated ML approaches compared to conventional methods based on subjective evaluations through experts.

BACKGROUND: We investigated whether immigrants from Southeast Europe (SE) and Swiss patients have different reasons for visiting the emergency department (ED). METHODS: Our retrospective data analysis for the years 2013-2017 describes the pattern of ED consultations for immigrants from SE living in Switzerland (Canton Bern), in comparison with Swiss nationals, with a focus on type of referral and reason for admission. RESULTS: A total of 153,320 Swiss citizens and 12,852 immigrants from SE were included in the study. The mean age was 51.30 (SD=21.13) years for the Swiss patients and 39.70 (SD=15.87) years for the SE patients. For some countries of origin (Albania, Bosnia and Herzegovina, and Turkey), there were highly statistically significant differences in sex distribution with the predominance for male. SE immigrants had a greater proportion of patients in the lower triage level (level 3: SE: 67.3% vs. Swiss: 56.0%) and a greater proportion of patients in the high triage level than the Swiss population (level 1: SE: 3.4% vs. Swiss: 8.8%,). SE patients of working age (16-65 years) were six times more often admitted by ambulance than older (≥65 years) SE patients, whereas in the Swiss population this ratio was similar. In both groups, fast track service was primarily used for patients of working age (<65) and more than three times more often in the SE than the Swiss group (SE: 39.1%, Swiss: 12.6%). CONCLUSION: We identified some indications in access to primary care in Emergency Department for immigrants highlighting the need for attention to the role of organizational characteristics of primary health care in the Switzerland. The authors have highlighted the need for professional support to improve the quality of healthcare for immigrants. In the future, more primary care services and general practitioners will need to be provided with a migrant background.

Respiratory tract epithelium infection plays a primary role in Nipah virus (NiV) pathogenesis and transmission. Knowledge about infection dynamics and host responses to NiV infection in respir-atory tract epithelia is scarce. Studies in non-differentiated primary respiratory tract cells or cell lines indicate insufficient interferon (IFN) responses. However, studies are lacking to determin-ing complex host response patterns in differentiated respiratory tract epithelia to understand NiV replication and spread in swine. Here we characterized infection and spread of NiV in differenti-ated primary porcine bronchial epithelial cells (PBEC) cultivated at the air-liquid-interface (ALI). After the initial infection of only a few apical cells, lateral spread for 12 days with epithelium disruption was observed without releasing substantial amounts of infectious virus from the api-cal or basal sides. Deep time course proteomics revealed pronounced upregulation of genes re-lated to type I/II- IFN, immunoproteasomal subunits, TAP-mediated peptide transport and MHC I antigen presentation. Spliceosomal factors were downregulated. We propose a model in which NiV replication in PBEC is slowed by a potent and broad type I/II-IFN host response with con-version from 26S proteasomes to immunoproteasomal antigen processing and improved MHC I presentation for adaptive immunity priming. NiV induced cytopathic effects could reflect the focal release of cell-associated NiV, which may contribute to efficient airborne viral spread between pigs.