Search

Die Bereitstellung von öffentlichen Dienstleistungen im Rahmen der kommunalen Daseinsvorsorge befindet sich in einem weitreichenden Umbruch. Ausgelöst durch die Digitalisierung und hiermit einhergehenden gesellschaftlichen Wandel müssen neue Wege gefunden werden, ein zeitgemäßes Angebot von Daseinsvorsorge auch im digitalen Raum bereitzustellen. Am Beispiel der Digitalisierung der Lübecker Schulen durch die Stadtwerke Lübeck Gruppe belegen die Autor:innen, dass sich eine neue Säule moderner Daseinsvorsorge etablieren muss. Digitale Infrastruktur, Datenmanagement und die Vermittlung von Know-How im Umgang mit digitalen Medien sind elementare Bausteine einer hierfür notwendigen „Digitalen Daseinsvorsorge“. Dies zahlt auf mehrere Sustainable Development Goals (SDG) ein.

Over the last decades, individualized approaches and a better understanding of coagulopathy complexity in end-stage liver disease (ESLD) patients has evolved. The risk of both thrombosis and bleeding during minimally invasive interventions or surgery is associated with a worse outcome in this patient population. Despite deranged quantitative and qualitative coagulation laboratory parameters, prophylactic coagulation management is unnecessary for patients who do not bleed. Transfusion of red blood cells (RBCs) and blood products carries independent risks for morbidity and mortality, including modulation of the immune system with increased risk for nosocomial infections. Optimal coagulation management in these complex patients should be based on the analysis of standard coagulation tests (SCTs) and viscoelastic tests (VETs). VETs represent an individualized approach to patients and can provide information about coagulation dynamics in a concise period of time. This narrative review will deliver the pathophysiology of deranged hemostasis in ESLD, explore the difficulties of evaluating the coagulopathies in liver disease patients, and examine the use of VET assays and management of coagulopathy using coagulation factors. Methods: A selective literature search with PubMed as the central database was performed with the following.

In December 2019, the new virus infection coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged. Simple clinical risk scores may improve the management of COVID-19 patients. Therefore, the aim of this pilot study was to evaluate the quick Sequential Organ Failure Assessment (qSOFA) score, which is well established for other diseases, as an early risk assessment tool predicting a severe course of COVID-19.We retrospectively analyzed data from adult COVID-19 patients hospitalized between March and July 2020. A critical disease progress was defined as admission to intensive care unit (ICU) or death.Of 64 COVID-19 patients, 33% (21/64) had a critical disease progression from which 13 patients had to be transferred to ICU. The COVID-19-associated mortality rate was 20%, increasing to 39% after ICU admission. All patients without a critical progress had a qSOFA score ≤ 1 at admission. Patients with a critical progress had in only 14% (3/21) and in 20% (3/15) of cases a qSOFA score ≥ 2 at admission (p = 0.023) or when measured directly before critical progression, respectively, while 95% (20/21) of patients with critical progress had an impairment oxygen saturation (SOA low qSOFA score cannot be used to assume short-term stable or noncritical disease status in COVID-19.GRUNDLAGEN: Im Dezember 2019 kam es zum Auftreten der neuen Virusinfektion „coronavirus disease 2019“ (COVID-19), hervorgerufen durch das „severe acute respiratory syndrome coronavirus 2“ (SARS-CoV-2). Einfache klinische Scores zur frühzeitigen Risikostratifizierung könnten das Management von COVD-19-Patient(inn)en verbessern. Ziel dieser Pilotstudie war es daher, den für andere Erkrankungen etablierten qSOFA-Score („quick sequential organ failure assessment score“) als frühzeitige Risikobewertung für kritische Krankheitsverläufe bei COVID-19 zu evaluieren.Es erfolgte eine retrospektive Datenanalyse von hospitalisierten COVID-19-Patient(inn)en aus dem Zeitraum März bis Juli 2020. Ein kritischer Krankheitsverlauf wurde als Aufnahme auf die Intensivstation (ICU) oder Tod definiert.Von 64 COVID-19-Erkrankten wiesen 33 % (21/64) einen kritischen Krankheitsverlauf auf, wovon 13 Patient(inn)en auf die ICU verlegt wurden. Die COVID-19-assoziierte Sterblichkeitsrate betrug 20 % und stieg nach ICU-Aufnahme auf 39 % an. Bei allen Patient(inn)en ohne kritischen Verlauf war bei Aufnahme der qSOFA-Score ≤ 1. Patient(inn)en mit einem kritischen Verlauf hatten in nur 14 % (3/21) der Fälle bei Aufnahme (p = 0,023) bzw. in 20 % (3/15) der Fälle direkt vor kritischer Verschlechterung einen qSOFA-Score ≥ 2, wohingegen 95 % (20/21) der Patienten und Patientinnen mit einem kritischen Verlauf zum Zeitpunkt der Aufnahme eine reduzierte Sauerstoffsättigung (SOBei einem niedrigen qSOFA-Score kann nicht von einem kurzfristig stabilen oder unkritischen Krankheitsverlauf von COVID-19 ausgegangen werden.

Background and objectives Fainting is a well-known side effect of blood donation. Such adverse experiences can diminish the return rate for further blood donations. Identifying factors associated with fainting could help prevent adverse incidents during blood donation. Materials and methods Data of 85,040 blood donations from whole blood and apheresis donors within four consecutive years were included in this retrospective study. Seven different machine learning models (random forests, artificial neural networks, XGradient Boosting, AdaBoost, logistic regression, K nearest neighbors, and support vector machines) for predicting fainting during blood donation were established. The used features derived from the data obtained from the questionnaire every donor has to fill in before the donation and weather data of the day of the donation. Results One thousand seven hundred fifteen fainting reactions were observed in 228 846 blood donations from 88,003 donors over a study period of 48 months. Similar values for all machine learning algorithms investigated for NPV, PPV, AUC, and F1-score were obtained. In general, NPV was above 0.996, whereas PPV was below 0.03. AUC and F1-score were close to 0.9 for all models. Essential features predicting fainting during blood donation were systolic and diastolic blood pressure and ambient temperature, humidity, and barometric pressure. Conclusion Machine-learning algorithms can establish prediction models of fainting in blood donors. These new tools can reduce adverse reactions during blood donation and improve donor safety and minimize negative associations relating to blood donation.

For many years, physicians’ approach to the transfusion of allogeneic red blood cells (RBC) was not individualized. It was accepted that a hemoglobin concentration (Hb) of less than 10 g/dL was a general transfusion threshold and the majority of patients were transfused immediately. In recent years, there has been increasing evidence that even significantly lower hemoglobin concentrations can be survived in the short term without sequelae. This somehow contradicts the observation that moderate or mild anemia is associated with relevant long-term morbidity and mortality. To resolve this apparent contradiction, it must be recognized that we have to avoid acute anemia or treat it by alternative methods. The aim of this article is to describe the physiological limits of acute anemia, match these considerations with clinical realities, and then present “patient blood management” (PBM) as the therapeutic concept that can prevent both anemia and unnecessary transfusion of RBC concentrates in a clinical context, especially in Intensive Care Units (ICU). This treatment concept may prove to be the key to high-quality patient care in the ICU setting in the future.

Despite being vital in treating intensive-care patients with lung failure, especially COVID-19 patients, Veno-Venous Extra-Corporeal Membrane Oxygenation does not exploit its full potential, leaving ample room for improvement. The objective of this study is to determine the effect of cannula positioning and blood flow on the efficacy of Veno-Venous Extra-Corporeal Membrane Oxygenation, in particular in relationship with blood recirculation. We performed 98 computer simulations of blood flow and oxygen diffusion in a computerized-tomography-segmented right atrium and venae cavae for different positions of the returning and draining cannulae and ECMO flows of 3 L/min and [Formula: see text]. For each configuration we measured how effective Veno-Venous Extra-Corporeal Membrane Oxygenation is at delivering oxygen to the right ventricle and thus to the systemic circulation. The main finding is that VV-ECMO efficacy is largely affected by the ECMO flow (global peak blood saturation: [Formula: see text]; average inter-group saturation gain: 9 percentage points) but only scarcely by the positioning of the cannulae (mean saturation ± standard deviation for the 3 L/min case: [Formula: see text]; for the [Formula: see text] case: [Formula: see text]). An important secondary outcome is that recirculation, more intense with a higher ECMO flow, is less detrimental to the procedure than previously thought. The efficacy of current ECMO procedures is intrinsically limited and fine-tuning the positions of the cannulae, risking infections, offers very little gain. Setting a higher ECMO flow offers the biggest benefit despite mildly increasing blood recirculation.

Massive perioperative allogeneic blood transfusion, that is, perioperative transfusion of more than 10 units of packed red blood cells (pRBC), is one of the main contributors to perioperative morbidity and mortality in cardiac surgery. Prediction of perioperative blood transfusion might enable preemptive treatment strategies to reduce risk and improve patient outcomes while reducing resource utilisation. We, therefore, investigated the precision of five different machine learning algorithms to predict the occurrence of massive perioperative allogeneic blood transfusion in cardiac surgery at our centre.Is it possible to predict massive perioperative allogeneic blood transfusion using machine learning?Retrospective, observational study.Single adult cardiac surgery centre in Austria between 01 January 2010 and 31 December 2019.Patients undergoing cardiac surgery.Primary outcome measures were the number of patients receiving at least 10 units pRBC, the area under the curve for the receiver operating characteristics curve, the F1 score, and the negative-predictive (NPV) and positive-predictive values (PPV) of the five machine learning algorithms used to predict massive perioperative allogeneic blood transfusion.A total of 3782 (1124 female:) patients were enrolled and 139 received at least 10 pRBC units. Using all features available at hospital admission, massive perioperative allogeneic blood transfusion could be excluded rather accurately. The best area under the curve was achieved by Random Forests: 0.810 (0.76 to 0.86) with high NPV of 0.99). This was still true using only the eight most important features [area under the curve 0.800 (0.75 to 0.85)].Machine learning models may provide clinical decision support as to which patients to focus on for perioperative preventive treatment in order to preemptively reduce massive perioperative allogeneic blood transfusion by predicting, which patients are not at risk.Johannes Kepler University Ethics Committee Study Number 1091/2021, Clinicaltrials.gov identifier NCT04856618.

Objective To develop a standardized approach to the implementation and performance of acute normovolemic hemodilution (ANH) in order to reduce the incidence of bleeding and allogeneic blood transfusion in high-risk surgical bleeding-related cardiac surgery with cardiopulmonary bypass (CPB). Design A 2-round modified RAND-Delphi consensus process. Participants Seven physicians from multiple geographic locations and clinical disciplines including anesthesiology and cardiac surgery and 1 cardiac surgery perfusionist participated in the survey. One registered nurse, specializing in Patient Blood Management, participated in the discussion but did not participate in the survey. Methods A modified RAND-Delphi method was utilized that integrated evidence review with a face-to-face expert multidisciplinary panel meeting, followed by repeated scoring using a 9-point Likert scale. Consensus was determined as a result from the second round survey, as follows: median rating of 1-3: ANH acceptable; median rating of 7-9: ANH not acceptable; median rating of 4-6: use clinical judgment. Results Evidentiary review identified 18 key peer-reviewed manuscripts for discussion. Through the consensus-building process, 39 statements including 26 contraindications to ANH and 10 CPB patient variables were assessed. In total, 22 statements were accepted or modified for the second scoring round. Conclusions Consensus was reached on 6 conditions in which ANH would or would not be acceptable, showing that development of a standardized approach for the use of ANH in high-risk surgical bleeding and allogeneic blood transfusion is clearly possible. The recommendations developed by this expert panel may help guide the management and inclusion of ANH as an evidence and consensus-based blood conservation modality.

The World Health Organization (WHO) has declared Coronavirus Disease 2019 (COVID-19), the disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a pandemic. Global health care now faces unprecedented challenges with widespread and rapid human-to-human transmission of SARS-CoV-2 and high morbidity and mortality with COVID-19 worldwide. Across the world, medical care is hampered by a critical shortage of not only hand sanitizers, personal protective equipment, ventilators, and hospital beds, but also impediments to the blood supply. Blood donation centers in many areas around the globe have mostly closed. Donors, practicing social distancing, some either with illness or undergoing self-quarantine, are quickly diminishing. Drastic public health initiatives have focused on containment and “flattening the curve” while invaluable resources are being depleted. In some countries, the point has been reached at which the demand for such resources, including donor blood, outstrips the supply. Questions as to the safety of blood persist. Although it does not appear very likely that the virus can be transmitted through allogeneic blood transfusion, this still remains to be fully determined. As options dwindle, we must enact regional and national shortage plans worldwide and more vitally disseminate the knowledge of and immediately implement patient blood management (PBM). PBM is an evidence-based bundle of care to optimize medical and surgical patient outcomes by clinically managing and preserving a patient’s own blood. This multinational and diverse group of authors issue this “Call to Action” underscoring “The Essential Role of Patient Blood Management in the Management of Pandemics” and urging all stakeholders and providers to implement the practical and commonsense principles of PBM and its multiprofessional and multimodality approaches.

Many previous studies claim to have developed machine learning models that diagnose COVID-19 from blood tests. However, we hypothesize that changes in the underlying distribution of the data, so called domain shifts, affect the predictive performance and reliability and are a reason for the failure of such machine learning models in clinical application. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (way of taking samples, laboratory procedures), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. We investigate whether domain shifts are present in COVID-19 datasets and how they affect machine learning methods. We further set out to estimate the mortality risk based on routinely acquired blood tests in a hospital setting throughout pandemics and under domain shifts. We reveal domain shifts by evaluating the models on a large-scale dataset with different assessment strategies, such as temporal validation. We present the novel finding that domain shifts strongly affect machine learning models for COVID-19 diagnosis and deteriorate their predictive performance and credibility. Therefore, frequent re-training and re-assessment are indispensable for robust models enabling clinical utility.

The ability to predict in-hospital mortality from data available at hospital admission would identify patients at risk and thereby assist hospital-wide patient safety initiatives. Our aim was to use modern machine learning tools to predict in-hospital mortality from standardized data sets available at hospital admission.This was a retrospective, observational study in 3 adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the 4 machine learning algorithms used: logistic regression, neural networks, random forests, and gradient boosting trees.Using our 4 predictive models, in-hospital mortality could be predicted satisfactorily (areas under the curve for neural networks, logistic regression, random forests, and gradient boosting trees: 0.932, 0.936, 0.935, and 0.935, respectively), with moderate F1 scores: 0.378, 0.367, 0.380, and 0.380, respectively. Average precision values were 0.312, 0.321, 0.334, and 0.323, respectively. It remains unknown whether additional features might improve our models; however, this would result in additional efforts for data acquisition in daily clinical practice.This study demonstrates that using only a limited, standardized data set in-hospital mortality can be predicted satisfactorily at the time point of hospital admission. More parameters describing patient's health are likely needed to improve our model.

There have been several attempts to quantify the diagnostic distortion caused by algorithms that perform low-dimensional electrocardiogram (ECG) representation. However, there is no universally accepted quantitative measure that allows the diagnostic distortion arising from denoising, compression, and ECG beat representation algorithms to be determined. Hence, the main objective of this work was to develop a framework to enable biomedical engineers to efficiently and reliably assess diagnostic distortion resulting from ECG processing algorithms. We propose a semiautomatic framework for quantifying the diagnostic resemblance between original and denoised/reconstructed ECGs. Evaluation of the ECG must be done manually, but is kept simple and does not require medical training. In a case study, we quantified the agreement between raw and reconstructed (denoised) ECG recordings by means of kappa-based statistical tests. The proposed methodology takes into account that the observers may agree by chance alone. Consequently, for the case study, our statistical analysis reports the "true", beyond-chance agreement in contrast to other, less robust measures, such as simple percent agreement calculations. Our framework allows efficient assessment of clinically important diagnostic distortion, a potential side effect of ECG (pre-)processing algorithms. Accurate quantification of a possible diagnostic loss is critical to any subsequent ECG signal analysis, for instance, the detection of ischemic ST episodes in long-term ECG recordings.

Supplemental Digital Content is available in the text., OBJECTIVES: To identify the epidemiology and outcome of adults and children with and without sepsis in a rural sub-Sahara African setting. DESIGN: A priori planned substudy of a prospective, before-and-after trial. SETTING: Rural, sub-Sahara African hospital. PATIENTS: One-thousand four-hundred twelve patients (adults, n = 491; children, n = 921) who were admitted to hospital because of an acute infection. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Demographic, clinical, laboratory data, danger signs, and the presence of sepsis (defined as a quick Sequential Organ Failure Assessment score count ≥ 2) at admission were extracted. Sepsis was observed in 69 adults (14.1%) and 248 children (26.9%). Sepsis patients differed from subjects without sepsis in several demographic and clinical aspects. Malaria was the most frequent type of infection in adults (66.7%) and children (63.7%) with sepsis, followed by suspected bacterial and parasitic infections other than malaria. Adults with sepsis more frequently developed respiratory failure (8.7% vs 2.1%; p = 0.01), had a higher in-hospital mortality (17.4% vs 8.3%; p < 0.001), were less often discharged home (81.2% vs 92.2%; p = 0.007), and had higher median (interquartile range) costs of care (30,300 [19,400–49,900] vs 42,500 Rwandan Francs [27,000–64,400 Rwandan Francs]; p = 0.004) than adults without sepsis. Children with sepsis were less frequently discharged home than children without sepsis (93.1% vs 96.4%; p = 0.046). Malaria and respiratory tract infections claimed the highest absolute numbers of lives. The duration of symptoms before hospital admission did not differ between survivors and nonsurvivors in adults (72 [24–168] vs 96 hr [72–168 hr]; p = 0.27) or children (48 [24–72] vs 36 [24–108 hr]; p = 0.8). Respiratory failure and coma were the most common causes of in-hospital death. CONCLUSIONS: In addition to suspected bacterial, viral, and fungal infections, malaria and other parasitic infections are common and important causes of sepsis in adults and children admitted to a rural hospital in sub-Sahara Africa. The in-hospital mortality associated with sepsis is substantial, primarily in adults.

Jehovah's Witnesses (JW) refuse allogeneic blood transfusions and therefore pose a unique challenge in case of major surgery. This retrospective study reviewed an experience with JW patients who were undergoing open heart surgery.By using patient blood management strategies, 35 adult JW patients underwent cardiac surgery at Kepler University Hospital in Linz, Austria between 2008 and 2017. Outcomes were compared with patients who accepted blood transfusions (non-JW patients) by using propensity score matching.There were no significant differences in clinical and operative data between the groups. Twelve JW patients (34.3%) were pretreated with erythropoietin and iron, with a preoperative increase in mean hemoglobin of 2.0 g/dL. On admission, hemoglobin was 14.1 ± 1.1 g/dL in JW patients, compared with 13.2 ± 2.0 g/dL in non-JW patients (P = .022). The hematocrit in JW patients was higher throughout the hospital stay, even though 51.4% of non-JW patients received allogeneic red blood cell transfusions. The perioperative red blood cell loss was significantly lower in JW patients than in non-JW patients (619 ± 420 mL vs 929 ± 520 mL; P = .010). Major complication rates were not different between the groups. The hemoglobin at discharge was 11.5 ± 1.5 g/dL in JW patients compared with 10.3 ± 1.3 g/dL in non-JW patients (P.001). In-hospital mortality was 2.9% in each group, and long-term survival was comparable.By implementing patient blood management, open heart surgery in JW patients can be performed with low morbidity and mortality. Preoperative optimization of hemoglobin and minimization of perioperative blood loss are cornerstones in the prevention of blood loss, anemia, and transfusions.

For many years, in daily clinical practice, the traditional 10/30 rule (hemoglobin 10 g/dL – hematocrit 30%) has been the most commonly used trigger for blood transfusions. Over the years, this approach is believed to have contributed to a countless number of unnecessary transfusions and an unknown number of overtransfusion-related deaths. Recent studies have shown that lower hemoglobin levels can safely be accepted, even in critically ill patients. However, even these new transfusion thresholds are far beyond the theoretical limits of individual anemia tolerance. For this reason, almost all publications addressing the limits of acute anemia recommend physiological transfusion triggers to indicate the transfusion of erythrocyte concentrates as an alternative. Although this concept appears intuitive at first glance, no solid scientific evidence supports the safety and benefit of physiological transfusion triggers to indicate the optimal time point for transfusion of allogeneic blood. It is therefore imperative to continue searching for the most sensitive and specific parameters that can guide the clinician when to transfuse in order to avoid anemia-induced organ dysfunction while avoiding overtransfusion-related adverse effects. This narrative review discusses the concept of anemia tolerance and critically compares hemoglobin-based triggers with physiological transfusion for various clinical indications.

ZusammenfassungDie anhaltende unkontrollierte Blutung ist eine der führenden Todesursachen bei polytraumatisierten Patienten 1, 2, 3, 4, 5. Hypoperfusion durch große Blutverluste führt zu Gewebeschäden, generalisierter Immunantwort sowie Aktivierung des Gerinnungssystems und damit zu einer traumaassoziierten Koagulopathie (TAK) 4, 5. Durch eine adäquate, frühzeitige Behandlung der TAK können Morbidität und Mortalität signifikant reduziert werden.

A new method of the end-of-life prediction for filled elastomers has been introduced and applied on fatigue under uniaxial deformations in [1]. The method considers measured inhomogeneity size frequencies. The dimension of these inhomogeneities is assumed to be equal to the sizes of the initial cracks responsible for the final end-of-life of the elastomer. For an end-of-life prediction, the initial crack size information is inserted into a standard fracture mechanic law. Within a Monte Carlo simulation, several spatial realization steps of statistical initial crack size distributions over the deformation field of the investigated component or specimen are performed. Within each realization step, the minimum number of cycles until end-of-life is extracted and supplies a failure probability plot. Performing a large number of realizations provides the complete failure probability information. In this work, that concept is applied on fatigue under simple shear. The predicted failure probability values match the experimental findings from dynamic fatigue tests for the considered Styrene Rubber.

Electronic health records (EHRs) have been successfully used in data science and machine learning projects. However, most of these data are collected for clinical use rather than for retrospective analysis. This means that researchers typically face many different issues when attempting to access and prepare the data for secondary use. We aimed to investigate how raw EHRs can be accessed and prepared in retrospective data science projects in a disciplined, effective, and efficient way. We report our experience and findings from a large-scale data science project analyzing routinely acquired retrospective data from the Kepler University Hospital in Linz, Austria. The project involved data collection from more than 150,000 patients over a period of 10 years. It included diverse data modalities, such as static demographic data, irregularly acquired laboratory test results, regularly sampled vital signs, and high-frequency physiological waveform signals. Raw medical data can be corrupted in many unexpected ways that demand thorough manual inspection and highly individualized data cleaning solutions. We present a general data preparation workflow, which was shaped in the course of our project and consists of the following 7 steps: obtain a rough overview of the available EHR data, define clinically meaningful labels for supervised learning, extract relevant data from the hospital’s data warehouses, match data extracted from different sources, deidentify them, detect errors and inconsistencies therein through a careful exploratory analysis, and implement a suitable data processing pipeline in actual code. Only few of the data preparation issues encountered in our project were addressed by generic medical data preprocessing tools that have been proposed recently. Instead, highly individualized solutions for the specific data used in one’s own research seem inevitable. We believe that the proposed workflow can serve as a guidance for practitioners, helping them to identify and address potential problems early and avoid some common pitfalls.

PLOS ONE 17(1), e0262110 (2022). doi:10.1371/journal.pone.0262110, Published by PLOS, San Francisco, California, US

We investigate machine learning models that identify COVID-19 positive patients and estimate the mortality risk based on routinely acquired blood tests in a hospital setting. However, during pandemics or new outbreaks, disease and testing characteristics change, thus we face domain shifts. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (taking samples, laboratory), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. To countermand this effect, we propose methods that first identify domain shifts and then reverse their negative effects on the model performance. Frequent re-training and reassessment, as well as stronger weighting of more recent samples, keeps model performance and credibility at a high level over time. Our diagnosis models are constructed and tested on large-scale data sets, steadily adapt to observed domain shifts, and maintain high ROC AUC values along pandemics.

A correlated lifetime prediction concept for load cases without static preload, which argues with crack growth and particle size distribution from 3D computer tomography, has been shown by Ludwig et al. (2015). This method is extended to non-relaxing load cases i.e. with a static preload dependency. A force controlled dynamic fatigue test for a dumbbell specimen is performed to investigate the service life. In addition, a crack growth investigation is carried out using single edge notched tensile (SENT) specimens in displacement control mode to characterize the tearing energy and crack growth rate. The study with carbon black reinforced HNBR rubber shows a correlation between the Wohler curve and the Paris-Erdogan plot. An extension of the empirical Paris-Erdogan equation considering static preload dependency allows the prediction of uniaxial lifetime statistics by means of particle size distribution. The calculated lifetime values are in reasonable concordance with the experimental findings.

OBJECTIVES Machine learning methods potentially provide a highly accurate and detailed assessment of expected individual patient risk before elective cardiac surgery. Correct anticipation of this risk allows for the improved counselling of patients and avoidance of possible complications. We therefore investigated the benefit of modern machine learning methods in personalized risk prediction for patients undergoing elective heart valve surgery. METHODS We performed a monocentric retrospective study in patients who underwent elective heart valve surgery between 1 January 2008 and 31 December 2014 at our centre. We used random forests, artificial neural networks and support vector machines to predict the 30-day mortality from a subset of 129 available demographic and preoperative parameters. Exclusion criteria were reoperation of the same patient, patients who needed anterograde cerebral perfusion due to aortic arch surgery and patients with grown-up congenital heart disease. Finally, the cohort consisted of 2229 patients with a 30-day mortality of 3.86% (86 of 2229 cases). This trial has been registered at clinicaltrials.gov (NCT03724123). RESULTS The final random forest model trained on the entire data set provided an out-of-bag area under the receiver operator characteristics curve (AUC) of 0.839, which significantly outperformed the European System for Cardiac Operative Risk Evaluation (EuroSCORE) (AUC = 0.704) and a model trained only on the subset of features EuroSCORE uses (AUC = 0.745). CONCLUSIONS Advanced machine learning methods can predict outcomes of valve surgery procedures with higher accuracy than established risk scores based on logistic regression on pre-selected parameters. This approach is generalizable to other elective high-risk interventions and allows for training models to the cohorts of specific institutions

Background. - Anaemia and coagulopathy are common issues in critically ill patients. Transfusion can be lifesaving, however, is associated with potential life threatening adverse events. As an international transfusion guideline for this specific patient population is lacking, we hypothesize that a high heterogeneity in transfusion practices exists. In this pilot-study we assessed transfusion practice in a university hospital in the Netherlands and tested the feasibility of this protocol for an international multi-centre study.Methods. - A prospective single centre cohort study was conducted. For seven days all consecutive non-readmitted patients to the adult Intensive Care Unit (ICU) were included and followed for 28 days. Patients were prospectively followed until ICU discharge or up to day 28. Patient outcome data was collected at day 28. Workload for this study protocol was scored in hours and missing data.Results. - In total, 48 patients were included, needed in total three hours patient to include and collect all data, with 1.6% missing data showing the feasibility of the data acquisition. Six (12.5%) patients received red blood cells (RBCs), three patients (6.3%) received platelet concentrates, and two (4.2%) patients received plasma units. In total eight (16.7%) patients were transfused with one or more blood products. Median pre- and post-transfusion haemoglobin (Hb) levels were 7.6 (6.7-7.7) g/dL and 8.1 (7.6-8.7) g/dL, respectively.Conclusion. - In this pilot-study we proved the feasibility of our protocol and observed in this small population a restrictive transfusion practice for all blood products. (C) 2019 Societe francaise de transfusion sanguine (SFTS). Published by Elsevier Masson SAS. All rights reserved.

This work presents the lifetime prediction of elastomer component based on the particle distribution in the specimen, the crack growth properties and the appropriate crack criterion. A lifetime prediction using numerical simulation saves both cost and time and can be very helpful in the development phase of a product. On the basis of the obtained results, lifetime can be estimated and thus an appropriate measure such as geometry change or load adjustment can be adopted. The prediction is demonstrated using two kinds of elastomer samples: EPDM and NR under different load amplitudes. Lifetime prediction is accomplished by Monte Carlo simulation, which is based on the correlation between the value of J-integral and total energy density of the single-edge notched tension (SENT) sample. Due to the choice of an inelastic material law the J-integral is evaluated after establishment of stable load cycles as the stable value for a J-integral contour curve as a function of the contour curve distance from the crack tip. For lifetime prediction, a Python script was implemented for the commercial FEA system Abaqus as a postprocessing routine. The script is based on local lifetime evaluation, where evaluation is done element by element followed up by an accumulation. In addition, physical experiments are used for counting the particles in the materials, to characterise the crack growth and also to measure lifetime.

Die Hypnose stellt ein sehr altes und effektives Heilverfahren dar. In den letzten Jahrzehnten war die moderne Hypnose vor allem eine Domane der Psychotherapie. Jedoch gewinnt die Hypnose als therapeutisches Verfahren in der Medizin wieder zunehmend an Bedeutung. Hypnose kann bei einer Vielzahl von medizinischen Indikationen eingesetzt werden. In der Literatur finden sich zahlreiche Beweise fur ihre Wirksamkeit. Ziel war es, die Effektivitat der Hypnose in der stationaren Therapie chronischer Schmerzpatienten aufzuzeigen und ein Selbsthypnoseprogramm darzustellen, welches gut in die Schmerztherapie integriert werden kann. Eingeschlossen wurden alle stationaren chronischen Schmerzpatienten von Oktober 2012 bis April 2013 (Gruppe 1: Nichthypnosegruppe; Gruppe 2: Hypnosegruppe). Fur die Gruppe 2 wurde ein Protokoll fur Hypnosetherapie zusatzlich zum standardisierten schmerztherapeutischen Programm integriert. Das Hauptziel der Hypnosetherapie war vor allem ein Selbsthypnosetraining, damit eine hausliche weitere Umsetzung gewahrleistet werden konnte. Mittels standardisierter Testverfahren (u. a. Gesundheitsfragebogen fur Patienten [PHQ-9], Pain Disability Index [PDI], Generalized Anxiety Disorder [GAD-7] und numerische Rating-Skalen [NRS] zum Schmerz und allgemeinen Wohlbefinden) wurden die Daten vor und nach der Schmerztherapie ausgewertet. Die pra- und poststationaren standardisierten Testverfahren von 30 chronischen Schmerzpatienten konnten ausgewertet werden (17 Patienten ohne Hypnose, 13 Patienten mit Hypnose). Die Hauptdiagnose nach ICD-10 war bei allen Patienten „chronische Schmerzerkrankung“ (F45.41) mit einem MPSS-Stadium III. Der PDI war in der Hypnosegruppe signifikant gebessert (p = 0,019). Die weiteren Items zeigten alle einen Trend zur Besserung in der Hypnosegruppe (Ausnahme GAD-7) ohne eine statistische Signifikanz (p > 0,05). Die vorliegende Arbeit konnte in einem ersten kleinen Patientenkollektiv zeigen, dass die Integration der modernen Hypnosetherapie in die Behandlung von chronischen Schmerzpatienten im stationaren Setting ein weiterer sinnvoller therapeutischer Aspekt sein kann. Insbesondere die Anleitung zum Erlernen selbststandig durchzufuhrender Hypnosetechniken erscheint aufgrund der begrenzten stationaren Zeit sinnvoll zu sein. Es bedarf weiterer Untersuchungen, um diese ersten Ergebnisse zu unterstutzen.

BACKGROUND Pre-operative anaemia and transfusion are common among patients undergoing elective orthopaedic surgery. Application of 'patient blood management' might be the most effective way to reduce both anaemia and transfusion. Pre-operative administration of iron and/or erythropoietin (EPO) is one of the cornerstones of the first pillar of patient blood management, but in a daily clinical setting, efficacy and long-term safety of this measure have not been analysed thoroughly to date. OBJECTIVE To investigate the influence of pre-operative preparation (PREP) of patients with iron and/or EPO on peri-operative transfusion needs and long-term survival. DESIGN Single-centre, retrospective study. SETTING Anaesthesia department, University hospital. INTERVENTIONS Pre-operative preparation with iron and/or EPO versus no preparation. METHODS After approval of our local ethics committee, data of 5518 patients who received total hip or total knee replacement between 2008 and 2014 were included. Patients receiving iron and/or EPO were included in the PREP group, whereas patients without iron and/or EPO were included in the no preparation group. From the full data set, a bias-reduced subset of 662 patients was obtained by means of propensity score-matching to compare peri-operative red blood cell utilisation and long-term survival of patients between groups. RESULTS Patients in the PREP group needed a lower number of units of red blood cells than patients in the no preparation group (0.2 ± 0.8 vs. 0.5 ± 1.3, P

For carbon black reinforced Ethylene-Propylene-Diene Monomer Rubber (EPDM) fracture mechanics investigations by using a Single Edge Notched Tension sample (SENT) are presented. Filled elastomers under cyclic loading typically display inelastic effects like permanent set, strong nonlinear S-shaped stress-strain behaviour and stress induced softening (discontinuous damage) as well as cyclic softening (continuous damage) better known as Mullins effect. These unique properties give this material class an interesting and a complex aspect regarding the experimental observations and numerical modelling of fracture mechanics. In simulations with homogenous material, the J-integral given as the evaluation of the Eshelby stress tensor along a path integral around the crack shows no dependency on the chosen integration path. In contrast, dissipative materials show strong path dependency of the J-integral. When the chosen path is near to the crack tip, this effect is pronounced, due to the inelastic effects, which are very high at the crack tip compared to the rest of the sample. In the process zone, the J-integral value depends strongly on the chosen path around the crack tip. This phenomenon occurs due to the intensity of inelastic effects near the crack tip. The path-dependency of J-integral using the variational formulation of a crack problem in a dissipative material has been given in [16]. Stumpf and Le have proposed a new formulation of the boundary value problem for elastoplastic cracked body using variational formulation, where the finite deformation was considered. For cyclic displacement controlled loading, the elastomers show typical stress softening behaviour, i.e. the force-displacement curve of each cycle is lower than the previous cycle and higher than the following cycle. This occurs due to the continuous damage, whereby the bondings between polymer chains and the filler particles are partially damaged. At small external amplitudes, the stress softening effect is significant only in the vicinity of the crack tip, because of the hysteresis which is large compared to the rest of the sample. The evaluation of the J-integral for an SENT sample under cyclic displacement controlled loading shows clearly that the resulting energy flux at the crack tip decreases from the virgin curve until the last repetition. The values found are in good agreement with experimentally evaluated tearing energy. Hence J-integral evaluation while considering a stress softening reflecting material model can be a promising approach used for estimation of the crack propagation or for lifetime prediction.

Rheumatoid arthritis (RA) patients are at increased risk of infection. Aim of the present study was to investigate whether RA patients admitted to an intensive care unit (ICU) due to infection have higher Rheumatoid Arthritis Observation of Biologic Therapy (RABBIT) risk scores compared to control RA patients. Seventy-four RA patients (32.4% male) admitted to an ICU due to infection (from January 2002 to December 2013) and 74 frequency-matched control RA patients (16.2% male) were included in this cross-sectional study. There was strong evidence for a higher RABBIT risk score in ICU patients (median 2.0; IQR 1.3–3.2) as compared to controls (1.3; IQR 0.8–2.0; p

A stretch intensity factor for filled and unfilled elastomers is introduced for different mixtures. This stretch intensity factor allows for prediction of the analytically evaluated energy release rate for a cracked sample under uniaxial tension. Considering the opening mode from fracture mechanics (mode I ) was investigated. The continuum mechanical derivations are based on non-linear hyperelastic material behaviour, where the energy release rate is evaluated through a closed path integral very near to the crack tip. Here, the integrand includes asymptotic solution for strain, stress and energy density using the Ogden model. The decisive advantage of this method is to predict well the critical tearing energy values by the crack growth using the analytical energy release rate term. In this work the Mullins effect is not considered, since the cracked samples are tested without any preconditioning.

Background Venoarterial extracorporeal membrane oxygenation (VA-ECMO) may be life-saving in several clinical situations, but it is also one of the most invasive therapeutic procedures, with significant potential for life-threatening complications. Pulse pressure waves are typically very small or even absent at the onset of ECMO therapy, and will reappear with the improvement of cardiac function. A low pulse pressure may indicate low cardiac output due to heart failure during sustained ECMO support. A sudden loss of pulse pressure during ECMO therapy, however, may reveal complications like pericardial tamponade, hemothorax or pneumothorax. Near infrared spectroscopy (NIRS) has been shown to be useful in detecting cerebral and lower limb ischemic events during ECMO therapy and could furthermore improve differential diagnosis in the event pulsatility of the arterial pressure trace is lost. Methods We are reporting on 3 different complications of ECMO and their impact on arterial pulse pressure, arterial oxygen tension and regional tissue oxygenation measured by NIRS. Results Pericardial hematoma, overinflation of the lung, and tension pneumothorax may impede cardiac output during VA-ECMO and cause a loss of pulse pressure. Monitoring of regional tissue oxygenation using NIRS, in addition to arterial and mixed venous oxygen tension, may allow early recognition and treatment of ECMO complications. Conclusions Together with the appearance of a flat, non pulsatile arterial pressure trace as well as a reduction in mixed venous oxygen saturation the improvement of upper body rSO2 measured by NIRS enables timely recognition of complications that interfere with natural cardiac output during VA-ECMO.

Elastomers composites are viscoelastic materials and show a complex behaviour due to their non-linear response to external load, characterised in general by energy dissipation, residual set and material softening. For production reasons defects like micro-cracks and pores may be contained. A micro crack in a filled elastomer can occur between polymer-polymer, filler-polymer or filler-filler interfaces. Such a crack grows when the energy flux at the crack tip exceeds a critical value. Energy considerations at the crack shows that the energy released during the crack growth is not the total dissipated energy. The remaining energy is dissipated in the rest of the elastomer composite. In this study three different fracture criteria are compared: the tearing energy (Rivlin and Thomas, 1953), energy release rate (Gros and Seelig, 2011) and the J-integral (Rice, 1968). Tearing energy and energy release rate are based on the global energetic consideration and J-integral is based on the local energy observation. As a global criterion the tearing energy can be determined experimentally and numerically, whereas the J-integral is a local criterion at the crack tip and can be evaluated only numerically. The importance of local calculations lies in the determination of actually required tearing energy as well as the description of inelastic effects of elastomers at the crack tip. The analytical evaluation of the energy release rate allows for the prediction of the driving force at the crack tip without carrying out experiments for the determination of the tearing energy. Stumpf and Le (1990) have applied the variational principle to determine the energy flux at the mode I crack tip for a hyperelastic body. In this work only the trouser-sample is analysed. The load on this sample corresponds to the presented mode III from linear fracture mechanics. Here, the Ogden model is chosen (ABAQUS, 1998) as hyperelastic material model. The analytical determination of the energy release rate is calculated from the energy density function of this model. Results obtained from the experiments, simulations and analytical calculations are compared, where the analytical energy release rate turned out to be in closer accordance with the experimentally evaluated tearing energy than the J-Integral.

Purpose of review Trauma patients are considered a complex population of patients in emergency medicine and need extensive, specialized therapy. One major part is the prevention and treatment of the inflammatory response, which occurs in patients after severe injury resulting in complications like endotheliopathy. Likely as a consequence, coagulopathy occurs. Sterile inflammation is hard to address, especially because of the lack of a single activator. Moreover, it is a complex composition of factors that lead to a pathologic immune response. Our understanding of these patterns is increasing, but the complete pathophysiologic changes have yet to be investigated. Therefore, there is no specific target to treat inflammatory response in trauma patients at the moment. Recent findings There is increasing knowledge of the pathways and mediators that are responsible for the inflammatory response in patients after severe trauma. The endothelial glycocalyx has been identified to be an integral part of these mechanisms. There have been several new therapeutic approaches to diminish the inflammatory response. Summary Our increasing understanding of the immune system have led to new potential therapeutic perspectives. All of these approaches need further research to be validated. As the current therapies are based on empirical strategies and have not changed much over the years, new treatment options would be an important progress.

Biomedical signal processing frequently deals with information extraction for clinical decision support. A major challenge in this field is to reveal diagnostic information by eliminating undesired interfering influences. In case of the electrocardiogram, e.g., a frequently arising interference is caused by respiration, which possibly superimposes diagnostic information. Respiratory sinus arrhythmia, i.e., the acceleration and deceleration of the heartrate (HR) during inhalation and exhalation, respectively, is a well-known phenomenon, which strongly influences the ECG. This influence becomes even more important, when investigating the so-called heart rate variability, a diagnostically powerful signal derived from the ECG. In this work, we propose a model for capturing the relationship between the HR and the respiration, thereby taking the time-variance of physiological systems into account. To this end, we show that so-called linear parameter varying autoregressive models with exogenous input are well suited for modeling the coupling between the two signals of interest.

A concept for the estimation of lifetime cycles is discussed assuming non-dispersed filler particles as origins of initial cracks which propagate under dynamic load according to fatigue crack growth (FCG) characteristics until failure occurs. Reference EPDM compounds with glass spheres of well defined size show strong correlation of the fatigue to failure analysis (FFA) behavior of dumbbells with largest incorporated particles, but dependence on polymer filler interaction, too. For NR and EPDM compounds, the occurrence of incorporated large particles is investigated by computed tomography and evaluated to a flaw size statistic. Based on the assumption of initial crack sizes matching the flaw diameters and together with the characteristic material parameters from FCG analysis, a statistical concept for the prediction of FFA lifetime analysis is presented. Predictions for near-homogeneously deforming dumbbell samples with carbon black (CB) reinforced NR display a particle size distribution which in combination with FCG results allows to calculate quantitative lifetime accordant to experimental findings, i.e. compounding dependency by shorter lifetime for worse dispersion and geometry dependency by longer lifetime for smaller specimens. An extension of the prediction concept for non-homogeneous deformation states is shown through a Monte Carlo simulation varying the positions of flaws inside the sample together with a Finite Element Analysis based calculation of the accordant local J-integral value. The simulations of lifetime statistics for rotational-parabolic buffer samples made of CB filled NR or EPDM show significant effects in average value and distribution width similarly found in experiment. This lifetime prediction concept has the unique capability to take into account not only recipe controlled matrix properties as cyclic crack propagation resistivity but volume dependency and processing related dispersion state, too in a quantitative manner.

To deploy machine learning tools (random forests) to develop a model that reliably predicts hospital mortality in children with acute infections residing in low- and middle-income countries, using age and other variables collected at hospital admission.Post hoc analysis of a single-center, prospective, before-and-after feasibility trial.Rural district hospital in Rwanda, a low-income country in Sub-Sahara Africa.Infants and children greater than 28 days and less than 18 years of life hospitalized because of an acute infection.None.Age, vital signs (heart rate, respiratory rate, and temperature) capillary refill time, altered mental state collected at hospital admission, as well as survival status at hospital discharge were extracted from the trial database. This information was collected for 1,579 adult and pediatric patients admitted to a regional referral hospital with an acute infection in rural Rwanda. Nine-hundred forty-nine children were included in this analysis. We predicted survival in study subjects using random forests, a machine learning algorithm. Five prediction models, all including age plus two to five other variables, were tested. Three distinct optimization criteria of the algorithm were then compared. The in-hospital mortality was 1.5% (n = 14). All five models could predict in-hospital mortality with an area under the receiver operating characteristic curve ranging between 0.69 and 0.8. The model including age, respiratory rate, capillary refill time, altered mental state exhibited the highest predictive value area under the receiver operating characteristic curve 0.8 (95% CI, 0.78-0.8) with the lowest possible number of variables.A machine learning-based algorithm could reliably predict hospital mortality in a Sub-Sahara African population of 949 children with an acute infection using easily collected information at admission which includes age, respiratory rate, capillary refill time, and altered mental state. Future studies need to evaluate and strengthen this algorithm in larger pediatric populations, both in high- and low-/middle-income countries.

Background The ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning–based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital. Study Design and Methods This was a retrospective, observational study in three adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures for the classification tasks were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the four machine learning algorithms used: neural networks (NNs), logistic regression (LR), random forests (RFs), and gradient boosting (GB) trees. Results Using our four predictive models, transfusion of at least 1 unit of RBCs could be predicted rather accurately (sensitivity for NN, LR, RF, and GB: 0.898, 0.894, 0.584, and 0.872, respectively; specificity: 0.958, 0.966, 0.964, 0.965). Using the four methods for prediction of massive transfusion was less successful (sensitivity for NN, LR, RF, and GB: 0.780, 0.721, 0.002, and 0.797, respectively; specificity: 0.994, 0.995, 0.993, 0.995). As a consequence, prediction of the total number of packed RBCs transfused was also rather inaccurate. Conclusion This study demonstrates that the necessity for intrahospital transfusion can be forecasted reliably, however the amount of RBC units transfused during a hospital stay is more difficult to predict.