Open Access

Background: In recent years, the volume of medical knowledge and health data has increased rapidly. For example, the increased availability of electronic health records (EHRs) provides accurate, up-to-date, and complete information about patients at the point of care and enables medical staff to have quick access to patient records for more coordinated and efficient care. With this increase in knowledge, the complexity of accurate, evidence-based medicine tends to grow all the time. Health care workers must deal with an increasing amount of data and documentation. Meanwhile, relevant patient data are frequently overshadowed by a layer of less relevant data, causing medical staff to often miss important values or abnormal trends and their importance to the progression of the patient’s case. Objective: The goal of this work is to analyze the current laboratory results for patients in the intensive care unit (ICU) and classify which of these lab values could be abnormal the next time the test is done. Detecting near-future abnormalities can be useful to support clinicians in their decision-making process in the ICU by drawing their attention to the important values and focus on future lab testing, saving them both time and money. Additionally, it will give doctors more time to spend with patients, rather than skimming through a long list of lab values. Methods: We used Structured Query Language to extract 25 lab values for mechanically ventilated patients in the ICU from the MIMIC-III and eICU data sets. Additionally, we applied time-windowed sampling and holding, and a support vector machine to fill in the missing values in the sparse time series, as well as the Tukey range to detect and delete anomalies. Then, we used the data to train 4 deep learning models for time series classification, as well as a gradient boosting–based algorithm and compared their performance on both data sets. Results: The models tested in this work (deep neural networks and gradient boosting), combined with the preprocessing pipeline, achieved an accuracy of at least 80% on the multilabel classification task. Moreover, the model based on the multiple convolutional neural network outperformed the other algorithms on both data sets, with the accuracy exceeding 89%. Conclusions: In this work, we show that using machine learning and deep neural networks to predict near-future abnormalities in lab values can achieve satisfactory results. Our system was trained, validated, and tested on 2 well-known data sets to ensure that our system bridged the reality gap as much as possible. Finally, the model can be used in combination with our preprocessing pipeline on real-life EHRs to improve patients’ diagnosis and treatment.

A new comprehensive evaluation system presented here allows to compare and to quantify education for a sustainable development (ESD) in degree programs. The evaluation is based on a criteria system working with three hierarchic levels. The highest level considers a list of 35 indicator terms. Primarily, the two most popular undergraduate (bachelor’s) degree programs in Germany (mechanical engineering, ME, and business administration, BA) have been reviewed for ESD contents based on the new evaluation scheme. Additionally we reviewed and quantified ESD subjects and their temporal changes in the entire bandwidth of degree programs of a university (Umwelt-Campus Birkenfeld, University of Applied Sciences Trier), back to 1999. Moreover, a spot check on international ME and BA bachelor’s degree programs was performed. Through our reviews, we found a high number of elective classes dedicated to ESD particularly in BA bachelor programs. However, the percentage of compulsory classes related to ESD is relatively low with 5-6 % in both ME and BA programs, respectively. The spot check on degree programs outside Germany revealed similar results. Analysing the time trend at Umwelt-Campus Birkenfeld, a considerable share of ESD that was part of the original diploma degrees was moved to what are now master’s degrees.

Global change effects on biodiversity and human wellbeing call for improved long-term environmental data as a basis for science, policy and decision making, including increased interoperability, multifunctionality, and harmonization. Based on the example of two global initiatives, the International Long-Term Ecological Research (ILTER) network and the Group on Earth Observations Biodiversity Observation Network (GEO BON), we propose merging the frameworks behind these initiatives, namely ecosystem integrity and essential biodiversity variables, to serve as an improved guideline for future site-based long-term research and monitoring in terrestrial, freshwater and coastal ecosystems. We derive a list of specific recommendations of what and how to measure at a monitoring site and call for an integration of sites into co-located site networks across individual monitoring initiatives, and centered on ecosystems. This facilitates the generation of linked comprehensive ecosystem monitoring data, supports synergies in the use of costly infrastructures, fosters cross-initiative research and provides a template for collaboration beyond the ILTER and GEO BON communities.

Intraspecific diet specialization, usually driven by resource availability, competition and predation, is common in natural populations. However, the role of parasites on diet specialization of their hosts has rarely been studied. Eye flukes can impair vision ability of their hosts and have been associated with alterations of fish feeding behavior. Here it was assessed whether European perch (Perca fluviatilis) alter their diet composition as a consequence of infection with eye flukes. Young-of-the-year (YOY) perch from temperate Lake Müggelsee (Berlin, Germany) were sampled in two years, eye flukes counted and fish diet was evaluated using both stomach content and stable isotope analyses. Perch diet was dominated by zooplankton and benthic macroinvertebrates. Both methods indicated that with increasing eye fluke infection intensity fish had a more selective diet, feeding mainly on the benthic macroinvertebrate Dikerogammarus villosus, while less intensively infected fish appeared to be generalist feeders showing no preference for any particular prey type. Our results show that infection with eye flukes can indirectly affect interaction of the host with lower trophic levels by altering the diet composition and highlight the underestimated role of parasites in food web studies.

1. Among the many concerns for biodiversity in the Anthropocene, recent reports of flying insect loss are particularly alarming, given their importance as pollinators, pest control agents, and as a food source. Few insect monitoring programmes cover the large spatial scales required to provide more generalizable estimates of insect responses to global change drivers. 2. We ask how climate and surrounding habitat affect flying insect biomass using data from the first year of a new monitoring network at 84 locations across Germany comprising a spatial gradient of land cover types from protected to urban and crop areas. 3. Flying insect biomass increased linearly with temperature across Germany. However, the effect of temperature on flying insect biomass flipped to negative in the hot months of June and July when local temperatures most exceeded long-term averages. 4. Land cover explained little variation in insect biomass, but biomass was lowest in forests. Grasslands, pastures, and orchards harboured the highest insect biomass. The date of peak biomass was primarily driven by surrounding land cover, with grasslands especially having earlier insect biomass phenologies. 5. Standardised, large-scale monitoring provides key insights into the underlying processes of insect decline and is pivotal for the development of climate-adapted strategies to promote insect diversity. In a temperate climate region, we find that the positive effects of temperature on flying insect biomass diminish in a German summer at locations where temperatures most exceeded long-term averages. Our results highlight the importance of local adaptation in climate change-driven impacts on insect communities.