Search

Formative assessment is about providing and using feedback and diagnostic information. On this basis, further learning or further teaching should be adaptive and, in the best case, optimized. However, this aspect is difficult to implement in reality, as teachers work with a large number of students and the whole process of formative assessment, especially the evaluation of student performance takes a lot of time. To address this problem, this paper presents an approach in which student performance is collected through a concept map and quickly evaluated using Machine Learning techniques. For this purpose, a concept map on the topic of mechanics was developed and used in 14 physics classes in Germany. After the student maps were analysed by two human raters on the basis of a four-level feedback scheme, a supervised Machine Learning algorithm was trained on the data. The results show a very good agreement between the human and Machine Learning evaluation. Based on these results, an embedding in everyday school life is conceivable, especially as support for teachers. In this way, the teacher can use and interpret the automatic evaluation and use it in the classroom.

As one of the automated guided vehicle (AGV) positioning methods, the LiDAR positioning method, based on artificial landmarks, has been widely used in warehousing logistics industries in recent years. However, the traditional LiDAR positioning method based on artificial landmarks mainly depends on the three-point positioning method, the performance of which is limited due to landmarks' layout and detection requirements. This paper proposes a LiDAR positioning algorithm based on iterative closest point (ICP) and artificial landmarks assistance. It provides improvements based on the traditional ICP algorithm. The result of positioning provided by the landmarks is used as the initial iteration ICP value. The combination of the ICP algorithm and landmarks enables the positioning algorithm to maintain a certain positioning precision when landmark detection is disturbed. By comparing the proposed algorithm with the positioning scheme developed by SICK in Germany, we prove that the combination of the ICP algorithm and landmarks can effectively improve the robustness under the premise of ensuring precision.

Classification of outdoor point clouds is an intensely studied topic, particularly with respect to the separation of vegetation from the terrain and manmade structures. In the presence of many overhanging and vertical structures, the (relative) height is no longer a reliable criterion for such a separation. An alternative would be to apply supervised classification; however, thousands of examples are typically required for appropriate training. In this paper, an unsupervised and rotation-invariant method is presented and evaluated for three datasets with very different characteristics. The method allows us to detect planar patches by filtering and clustering so-called superpoints, whereby the well-known but suitably modified random sampling and consensus (RANSAC) approach plays a key role for plane estimation in outlier-rich data. The performance of our method is compared to that produced by supervised classifiers common for remote sensing settings: random forest as learner and feature sets for point cloud processing, like covariance-based features or point descriptors. It is shown that for point clouds resulting from airborne laser scans, the detection accuracy of the proposed method is over 96% and, as such, higher than that of standard supervised classification approaches. Because of artifacts caused by interpolation during 3D stereo matching, the overall accuracy was lower for photogrammetric point clouds (74-77%). However, using additional salient features, such as the normalized green-red difference index, the results became more accurate and less dependent on the data source.

Background: In cardiology, cardiac output (CO) is an important parameter for assessing cardiac function. While invasive thermodilution procedures are the gold standard for CO assessment, transthoracic Doppler echocardiography (TTE) has become the established method for routine CO assessment in daily clinical practice. However, a demand persists for non-invasive approaches, including oscillometric pulse wave analysis (PWA), to enhance the accuracy of CO estimation, reduce complications associated with invasive procedures, and facilitate its application in non-intensive care settings. Here, we aimed to compare the TTE and oscillometric PWA algorithm Antares for a non-invasive estimation of CO. Methods: Non-invasive CO data obtained by two-dimensional TTE were compared with those from an oscillometric blood pressure device (custo med GmbH, Ottobrunn, Germany) using the integrated algorithm Antares (Redwave Medical GmbH, Jena, Germany). In total, 59 patients undergoing elective cardiac catheterization for clinical reasons (71±10 years old, 76% males) were included. Agreement between both CO measures were assessed by Bland-Altman analysis, Student's t-test, and Pearson correlations. Results: The mean difference in CO was 0.04 ± 1.03 l/min (95% confidence interval for the mean difference: -0.23 to 0.30 l/min) for the overall group, with lower and upper limits of agreement at -1.98 and 2.05 l/min, respectively. There was no statistically significant difference in means between both CO measures (P = 0.785). Statistically significant correlations between TTE and Antares CO were observed in the entire cohort (r = 0.705, P<0.001) as well as in female (r = 0.802, P<0.001) and male patients (r = 0.669, P<0.001). Conclusions: The oscillometric PWA algorithm Antares and established TTE for a non-invasive estimation of CO are highly correlated in male and female patients, with no statistically significant difference between both approaches. Future validation studies of the Antares CO are necessary before a clinical application can be considered. [ABSTRACT FROM AUTHOR]

Background: Staffing ratios in nursing homes vary among the federal states of Germany, but there are no rational grounds for these variations. In a previous study, a new instrument for the standardized calculation of staffing requirements in nursing homes was developed (Algorithm1.0). The development was based on a new empirical data collection method that derives actual and target values for the time and number of care interventions provided. Algorithm1.0 found an increased requirement of 36% of staff in German nursing homes. Based on these results, the German legislature has commissioned a model program to trial and evaluate a complex intervention comprising increased staffing combined with strategies for organizational development. Methods: The mixed-methods study consists of (i) developing a concept for restructuring the work organization, (ii) the application of this concept combined with increased staffing in 10 nursing homes (complex intervention), and the further development of the concept using a participatory and iterative formal evaluation process. The intervention consists of (a) quantitative measures of increased staffing based on a calculation using Algorithm1.0 and (b) qualitative measures regarding organizational development. The intervention will be conducted over one year. The effects of the intervention on job satisfaction and quality of care will be evaluated in (iii) a comprehensive prospective, controlled summative evaluation. The results will be compared with ten matched nursing homes as a control group. Finally, (iv) prototypical concepts for qualification-oriented work organization, a strategy for the national rollout, and the further development of Algorithm1.0 into Algorithm 2.0 will be derived. Discussion: In Germany, there is an ongoing dynamic legislation process regarding further developing the long-term care sector. The study, which is the subject of the study protocol presented here, generates an evidence-based strategy for the staffing requirements for nursing homes. Ethics and dissemination. This study was approved by the Ethics Committee of the German Association of Nursing Science (Deutsche Gesellschaft für Pflegewissenschaft) on 02.08.2023 (amended on 20.09.2023). Research findings are disseminated through presentations at national and international conferences and publications in peer-reviewed scientific journals. Trial registration number: German Clinical Trails Register DRKS00031773 (Date of registration 09.11.2023). [ABSTRACT FROM AUTHOR]

Context: YOLO (You Look Only Once) is an algorithm based on deep neural networks with real-time object detection capabilities. This state-of-the-art technology is widely available, mainly due to its speed and precision. Since its conception, YOLO has been applied to detect and recognize traffic signs, pedestrians, traffic lights, vehicles, and so on. Objective: The goal of this research is to systematically analyze the YOLO object detection algorithm, applied to traffic sign detection and recognition systems, from five relevant aspects of this technology: applications, datasets, metrics, hardware, and challenges. Method: This study performs a systematic literature review (SLR) of studies on traffic sign detection and recognition using YOLO published in the years 2016–2022. Results: The search found 115 primary studies relevant to the goal of this research. After analyzing these investigations, the following relevant results were obtained. The most common applications of YOLO in this field are vehicular security and intelligent and autonomous vehicles. The majority of the sign datasets used to train, test, and validate YOLO-based systems are publicly available, with an emphasis on datasets from Germany and China. It has also been discovered that most works present sophisticated detection, classification, and processing speed metrics for traffic sign detection and recognition systems by using the different versions of YOLO. In addition, the most popular desktop data processing hardwares are Nvidia RTX 2080 and Titan Tesla V100 and, in the case of embedded or mobile GPU platforms, Jetson Xavier NX. Finally, seven relevant challenges that these systems face when operating in real road conditions have been identified. With this in mind, research has been reclassified to address these challenges in each case. Conclusions: This SLR is the most relevant and current work in the field of technology development applied to the detection and recognition of traffic signs using YOLO. In addition, insights are provided about future work that could be conducted to improve the field. [ABSTRACT FROM AUTHOR]

Detecting changes in soil micro-relief in farmland helps to understand degradation processes like sheet erosion. Using the high-resolution technique of terrestrial laser scanning (TLS), we generated point clouds of three 2 × 3 m plots on a weekly basis from May to mid-June in 2022 on cultivated farmland in Germany. Three well-known applications for eliminating vegetation points in the generated point cloud were tested: Cloth Simulation Filter (CSF) as a filtering method, three variants of CANUPO as a machine learning method, and ArcGIS PointCNN as a deep learning method, a sub-category of machine learning using deep neural networks. We assessed the methods with hard criteria such as F1 score, balanced accuracy, height differences, and their standard deviations to the reference surface, resulting in data gaps and robustness, and with soft criteria such as time-saving capacity, accessibility, and user knowledge. All algorithms showed a low performance at the initial measurement epoch, increasing with later epochs. While most of the results demonstrate a better performance of ArcGIS PointCNN, this algorithm revealed an exceptionally low performance in plot 1, which is describable by the generalization gap. Although CANUPO variants created the highest amount of data gaps, we recommend that CANUPO include colour values in combination with CSF. [ABSTRACT FROM AUTHOR]

In March 1964, the cover page of a popular German weekly magazine entitled Der Spiegel painted a frightening picture: An anthropomorphic robot with six mechanical arms commands an assembly line while a displaced human worker floats aimlessly in the foreground. Ejected from his station, the worker throws up his hands in despair next to a headline that reads, "Automation in Germany, the arrival of robots." Over fifty years later, a cover page from the same magazine evoked similar themes: A giant robot arm yanks an office worker away from his computer under the headline, "You're fired! How computers and robots steal our jobs - and which jobs will be safe." The more things change, the more they stay the same. [ABSTRACT FROM AUTHOR]

During the outbreak of a disease caused by a pathogen with unknown characteristics, the uncertainty of its progression parameters can be reduced by devising methods that, based on rational assumptions, exploit available information to provide actionable insights. In this study, performed a few (~6) weeks into the outbreak of COVID-19 (caused by SARS-CoV-2), one of the most important disease parameters, the average time-to-recovery, was calculated using data publicly available on the internet (daily reported cases of confirmed infections, deaths, and recoveries), and fed into an algorithm that matches confirmed cases with deaths and recoveries. Unmatched cases were adjusted based on the matched cases calculation. The mean time-to-recovery, calculated from all globally reported cases, was found to be 18.01 days (SD 3.31 days) for the matched cases and 18.29 days (SD 2.73 days) taking into consideration the adjusted unmatched cases as well. The proposed method used limited data and provided experimental results in the same region as clinical studies published several months later. This indicates that the proposed method, combined with expert knowledge and informed calculated assumptions, could provide a meaningful calculated average time-to-recovery figure, which can be used as an evidence-based estimation to support containment and mitigation policy decisions, even at the very early stages of an outbreak. [ABSTRACT FROM AUTHOR]

Grasslands are an important part of pre-Alpine and Alpine landscapes. Despite the economic value and the significant role of grasslands in carbon and nitrogen (N) cycling, spatially explicit information on grassland biomass and quality is rarely available. Remotely sensed data from unmanned aircraft systems (UASs) and satellites might be an option to overcome this gap. Our study aims to investigate the potential of low-cost UAS-based multispectral sensors for estimating above-ground biomass (dry matter, DM) and plant N concentration. In our analysis, we compared two different sensors (Parrot Sequoia, SEQ; MicaSense RedEdge-M, REM), three statistical models (linear model; random forests, RFs; gradient-boosting machines, GBMs), and six predictor sets (i.e. different combinations of raw reflectance, vegetation indices, and canopy height). Canopy height information can be derived from UAS sensors but was not available in our study. Therefore, we tested the added value of this structural information with in situ measured bulk canopy height data. A combined field sampling and flight campaign was conducted in April 2018 at different grassland sites in southern Germany to obtain in situ and the corresponding spectral data. The hyper-parameters of the two machine learning (ML) approaches (RF, GBM) were optimized, and all model setups were run with a 6-fold cross-validation. Linear models were characterized by very low statistical performance measures, thus were not suitable to estimate DM and plant N concentration using UAS data. The non-linear ML algorithms showed an acceptable regression performance for all sensor–predictor set combinations with average (avg; cross-validated, cv) Rcv2 of 0.48, RMSE cv,avg of 53.0 g m 2 , and rRMSE cv,avg (relative) of 15.9 % for DM and with Rcv,avg2 of 0.40, RMSE cv,avg of 0.48 wt %, and rRMSE cv, avg of 15.2 % for plant N concentration estimation. The optimal combination of sensors, ML algorithms, and predictor sets notably improved the model performance. The best model performance for the estimation of DM (Rcv2=0.67 , RMSE cv=41.9 g m 2 , rRMSE cv=12.6 %) was achieved with an RF model that utilizes all possible predictors and REM sensor data. The best model for plant N concentration was a combination of an RF model with all predictors and SEQ sensor data (Rcv2=0.47 , RMSE cv=0.45 wt %, rRMSE cv=14.2 %). DM models with the spectral input of REM performed significantly better than those with SEQ data, while for N concentration models, it was the other way round. The choice of predictors was most influential on model performance, while the effect of the chosen ML algorithm was generally lower. The addition of canopy height to the spectral data in the predictor set significantly improved the DM models. In our study, calibrating the ML algorithm improved the model performance substantially, which shows the importance of this step. [ABSTRACT FROM AUTHOR]

[Display omitted] • We simulate renewables pathways under different support phase-out scenarios. • We extend an investment algorithm by preferences and calibrated returns. • Our improved algorithm incorporates more heterogeneity resulting in stronger effects. • Further auction support is necessary for most capacity targets in the case countries. • Countries should coordinate policy changes due to cross-border effects. Support schemes have been central to the expansion of renewable electricity globally and in the European Union. As technologies mature, individual member states may decide to phase out these policies. While previous research has shown that such policy changes affect investors' decisions, we investigate how they affect pathways and electricity prices by simulating investment decisions in an agent-based model in two case countries. This paper contributes and applies an adapted investment decision algorithm that incorporates empirically observed technology and return preferences and is calibrated by return observations. The new algorithm yields more refined and stronger effects compared to its predecessor. Results show that the phase-out of auctions in Germany and the Netherlands slows down their deployment of renewable capacity by up to ∼60% and ∼35%, respectively. With the exception of photovoltaics and onshore wind projects in the Netherlands, the targeted capacities can only be reached by continuing support in both countries. Furthermore, ending support in a large country like Germany leads to higher electricity prices and fosters a market-driven but insufficient capacity expansion in smaller neighbours like the Netherlands. As the electricity grids in many countries are strongly interconnected, such cross-border effects are of international relevance. Our findings suggest that continued auctions may be necessary and that countries should coordinate policy changes to stay on track for meeting their renewables targets. [ABSTRACT FROM AUTHOR]

Two-thirds of the final energy consumption of the EU residential sector goes towards space heating of buildings, yet a huge portion of the population still suffers from energy poverty. Identifying optimum heating strategies of current buildings would be a solution to this crisis, which is the main aim of the developed algorithm in this research. The algorithm incorporates a modified version of the simple hourly method from the ISO 13790 standard to determine the hourly heating load and indoor temperatures of buildings based on any heating strategy. Flexibility in the input of building and weather data make this tool versatile with practicality towards building users and policymakers. With this algorithm, a case study to evaluate three commonly used domestic heating strategies has been established for nine different residential buildings in typical cold winter conditions in Germany. Most EU households heat their buildings either continuously throughout the day at fixed temperatures, sporadically at fixed times or at peak loads during the evening. The continuous heating strategy is rated the best consuming minimal energy with consistent temperatures and optimal thermal comfort ranges. The sporadic and peak load heating strategies provide fluctuating indoor temperatures with high standard deviations of up to 8.70°C while consuming a similar cumulative energy to the continuous heating strategy. Additionally, both these aforementioned strategies augment energy poverty and promote indoor mould formation on the building envelope caused by water vapor condensation. Consequently, this algorithm can be applicable to any building type of any region. [ABSTRACT FROM AUTHOR]

Background: The objective of the STREAM Trial was to evaluate the effect of simulation training on process times in acute stroke care. Methods: The multicenter prospective interventional STREAM Trial was conducted between 10/2017 and 04/2019 at seven tertiary care neurocenters in Germany with a pre‐ and post‐interventional observation phase. We recorded patient characteristics, acute stroke care process times, stroke team composition and simulation experience for consecutive direct‐to‐center patients receiving intravenous thrombolysis (IVT) and/or endovascular therapy (EVT). The intervention consisted of a composite intervention centered around stroke‐specific in situ simulation training. Primary outcome measure was the 'door‐to‐needle' time (DTN) for IVT. Secondary outcome measures included process times of EVT and measures taken to streamline the pre‐existing treatment algorithm. Results: The effect of the STREAM intervention on the process times of all acute stroke operations was neutral. However, secondary analyses showed a DTN reduction of 5 min from 38 min pre‐intervention (interquartile range [IQR] 25–43 min) to 33 min (IQR 23–39 min, p = 0.03) post‐intervention achieved by simulation‐experienced stroke teams. Concerning EVT, we found significantly shorter door‐to‐groin times in patients who were treated by teams with simulation experience as compared to simulation‐naive teams in the post‐interventional phase (−21 min, simulation‐naive: 95 min, IQR 69–111 vs. simulation‐experienced: 74 min, IQR 51–92, p = 0.04). Conclusion: An intervention combining workflow refinement and simulation‐based stroke team training has the potential to improve process times in acute stroke care. [ABSTRACT FROM AUTHOR]