Search

Abstract Measures for balancing the electrical grid, such as peak shaving, require accurate peak forecasts for lower aggregation levels of electrical loads. Thus, the Big Data Energy Analytics Laboratory (BigDEAL) challenge—organised by the BigDEAL—focused on forecasting three different daily peak characteristics in low aggregated load time series. In particular, participants of the challenge were asked to provide long‐term forecasts with horizons of up to 1 year in the qualification. The authors present the approach of the KIT‐IAI team from the Institute for Automation and Applied Informatics at the Karlsruhe Institute of Technology. The approach to the challenge is based on a hybrid generative model. In particular, the authors use a conditional Invertible Neural Network (cINN). The cINN gets the forecast of a sliding mean as representative of the trend, different weather features, and calendar information as conditioning input. By this, the proposed hybrid method achieved second place overall and won two out of three tracks of the BigDEAL challenge.

Abstract Brain organoids represent a useful tool for modeling of neurodevelopmental disorders and can recapitulate brain volume alterations such as microcephaly. To monitor organoid growth, brightfield microscopy images are frequently used and evaluated manually which is time-consuming and prone to observer-bias. Recent software applications for organoid evaluation address this issue using classical or AI-based methods. These pipelines have distinct strengths and weaknesses that are not evident to external observers. We provide a dataset of more than 1,400 images of 64 trackable brain organoids from four clones differentiated from healthy and diseased patients. This dataset is especially powerful to test and compare organoid analysis pipelines because of (1) trackable organoids (2) frequent imaging during development (3) clone diversity (4) distinct clone development (5) cross sample imaging by two different labs (6) common imaging distractors, and (6) pixel-level ground truth organoid annotations. Therefore, this dataset allows to perform differentiated analyses to delineate strengths, weaknesses, and generalizability of automated organoid analysis pipelines as well as analysis of clone diversity and similarity.

Thermal Interface Materials (TIMs) are widely used in electronic packaging. Increasing power density and limited assembly space pose high demands on thermal management. Large cooling surfaces need to be covered efficiently. When joining the heatsink, previously dispensed TIM spreads over the cooling surface. Recommendations on the dispense pattern exist only for simple surface geometries such as rectangles. For more complex geometries, Computational Fluid Dynamics (CFD) simulations are used in combination with manual experiments. While CFD simulations offer a high accuracy, they involve simulation experts and are rather expensive to set up. We propose a lightweight heuristic to model the spreading behavior of TIM. We further speed up the calculation by training an Artificial Neural Network (ANN) on data from this model. This offers rapid computation times and further supplies gradient information. This ANN can not only be used to aid manual pattern design of TIM, but also enables an automated pattern optimization. We compare this approach against the state-of-the-art and use real product samples for validation.

Automated microscopy techniques combined with high-throughput microfluidic cultivation systems provide unique insights into living microorganisms, enabling hundreds of experiments to be performed in parallel. Such setups generate large data sets with rich cellular features that need to be extracted to arrive at quantitative insights. The sheer amount of recorded time-lapse images requires reliable automated processing. While recent advances in deep learning methods have enabled automated processing, these methods rely on large-scale and precisely annotated data sets, often not available for new organisms or custom imaging modalities.To overcome the annotated data bottleneck, particularly in the microbial domain, we present the open-source ObiWan-Microbi platform providing a fast workflow for large-scale annotation of up to hundred thousands of segmentation and tracking annotations in time-lapse imaging data. ObiWan-Microbi focuses on easy-to-use semi-automated annotation in the browser eliminating the need for local installation or accelerator hardware, encourages FAIR data management using OMERO, and provides convenient collaborative cloud deployment to simplify the creation and long-term development of large-scale annotated data sets. The public availability of such benchmark data sets has the potential to improve data-driven methods, increase comparability among them, and is an essential step towards reliable automated image processing in microbial live-cell microscopy.

Abstract Cerebral organoids recapitulate the structure and function of the developing human brain in vitro, offering a large potential for personalized therapeutic strategies. The enormous growth of this research area over the past decade with its capability for clinical translation makes a non-invasive, automated analysis pipeline of organoids highly desirable. This work presents a novel non-invasive approach to monitor and analyze cerebral organoids over time using high-field magnetic resonance imaging and state-of-the-art tools for automated image analysis. Three specific objectives are addressed, (I) organoid segmentation to investigate organoid development over time, (II) global cysticity classification and (III) local cyst segmentation for organoid quality assessment. We show that organoid growth can be monitored reliably over time and cystic and non-cystic organoids can be separated with high accuracy, with on par or better performance compared to state-of-the-art tools applied to brightfield imaging. Local cyst segmentation is feasible but could be further improved in the future. Overall, these results highlight the potential of the pipeline for clinical application to larger-scale comparative organoid analysis.

Abstract Renewable energy systems depend on the weather, and weather information, thus, plays a crucial role in forecasting time series within such renewable energy systems. However, while weather data are commonly used to improve forecast accuracy, it still has to be determined in which input shape this weather data benefits the forecasting models the most. In the present paper, we investigate how transformations for weather data inputs, i. e., station-based and grid-based weather data, influence the accuracy of energy time series forecasts. The selected weather data transformations are based on statistical features, dimensionality reduction, clustering, autoencoders, and interpolation. We evaluate the performance of these weather data transformations when forecasting three energy time series: electrical demand, solar power, and wind power. Additionally, we compare the best-performing weather data transformations for station-based and grid-based weather data. We show that transforming station-based or grid-based weather data improves the forecast accuracy compared to using the raw weather data between 3.7 and 5.2%, depending on the target energy time series, where statistical and dimensionality reduction data transformations are among the best.

Abstract In the smart grid of the future, accurate load forecasts on the level of individual clients can help to balance supply and demand locally and to prevent grid outages. While the number of monitored clients will increase with the ongoing smart meter rollout, the amount of data per client will always be limited. We evaluate whether a Transformer load forecasting model benefits from a transfer learning strategy, where a global univariate model is trained on the load time series from multiple clients. In experiments with two datasets containing load time series from several hundred clients, we find that the global training strategy is superior to the multivariate and local training strategies used in related work. On average, the global training strategy results in 21.8% and 12.8% lower forecasting errors than the two other strategies, measured across forecasting horizons from one day to one month into the future. A comparison to linear models, multi-layer perceptrons and LSTMs shows that Transformers are effective for load forecasting when they are trained with the global training strategy.

Communication is crucial for organismic interactions, from bacteria, to fungi, to humans. Humans may use the visual sense to monitor the environment before starting acoustic interactions. In comparison, fungi, lacking a visual system, rely on a cell-to-cell dialogue based on secreted signaling molecules to coordinate cell fusion and establish hyphal networks. Within this dialogue, hyphae alternate between sending and receiving signals. This pattern can be visualized via the putative signaling protein Soft (SofT), and the mitogen-activated protein kinase MAK-2 (MakB) which are recruited in an alternating oscillatory manner to the respective cytoplasmic membrane or nuclei of interacting hyphae. Here, we show that signal oscillations already occur in single hyphae of Arthrobotrys flagrans in the absence of potential fusion partners (cell monologue). They were in the same phase as growth oscillations. In contrast to the anti-phasic oscillations observed during the cell dialogue, SofT and MakB displayed synchronized oscillations in phase during the monologue. Once two fusion partners came into each other’s vicinity, their oscillation frequencies slowed down (entrainment phase) and transit into anti-phasic synchronization of the two cells’ oscillations with frequencies of 104±28 s and 117±19 s, respectively. Single-cell oscillations, transient entrainment, and anti-phasic oscillations were reproduced by a mathematical model where nearby hyphae can absorb and secrete a limited molecular signaling component into a shared extracellular space. We show that intracellular Ca2+ concentrations oscillate in two approaching hyphae, and depletion of Ca2+ from the medium affected vesicle-driven extension of the hyphal tip, abolished the cell monologue and the anti-phasic synchronization of two hyphae. Our results suggest that single hyphae engage in a ‘monologue’ that may be used for exploration of the environment and can dynamically shift their extracellular signaling systems into a ‘dialogue’ to initiate hyphal fusion.

The opportunistic yeast Candida albicans is the most common cause of candidiasis. With only four classes of antifungal drugs on the market, resistance is becoming a problem in the treatment of fungal infections, especially in immunocompromised patients. The development of novel antifungal drugs with different modes of action is urgent. In 2016, we developed a groundbreaking new medium-throughput method to distinguish the effects of antibacterial agents. Using small-angle X-ray scattering for biological samples (BioSAXS), it is now possible to screen hundreds of new antibacterial compounds and select those with the highest probability for a novel mode of action. However, yeast (eukaryotic) cells are highly structured compared to bacteria. The fundamental question to answer was if the ultrastructural changes induced by the action of an antifungal drug can be detected even when most structures in the cell stay unchanged. In this exploratory work, BioSAXS was used to measure the ultrastructural changes of C. albicans that were directly or indirectly induced by antifungal compounds. For this, the well-characterized antifungal drug Flucytosine was used. BioSAXS measurements were performed on the synchrotron P12 BioSAXS beamline, EMBL (DESY, Hamburg) on treated and untreated yeast C. albicans. BioSAXS curves were analysed using principal component analysis (PCA). The PCA showed that Flucytosine-treated and untreated yeast were separated. Based on that success further measurements were performed on five antifungal peptides {1. Cecropin A-melittin hybrid [CA (1–7) M (2–9)], KWKLFKKIGAVLKVL; 2. Lasioglossin LL-III, VNWKKILGKIIKVVK; 3. Mastoparan M, INLKAIAALAKKLL; 4. Bmkn2, FIGAIARLLSKIFGKR; and 5. optP7, KRRVRWIIW}. The ultrastructural changes of C. albicans indicate that the peptides may have different modes of action compared to Flucytosine as well as to each other, except for the Cecropin A-melittin hybrid [CA (1–7) M (2–9)] and optP7, showing very similar effects on C. albicans. This very first study demonstrates that BioSAXS shows promise to be used for antifungal drug development. However, this first study has limitations and further experiments are necessary to establish this application.

Abstract In the electricity grid, constantly balancing the supply and demand is critical for the network’s stability and any expected deviations require balancing efforts. This balancing becomes more challenging in future energy systems characterised by a high proportion of renewable generation due to the increased volatility of these renewables. In order to know when any balancing efforts are required, it is essential to predict the so-called net load, the difference between forecast energy demand and renewable supply. Although various forecasting approaches exist for both the individual components of the net load and the net load itself, it is unclear if it is more beneficial to aggregate several specialised forecasts to obtain the net load or to aggregate the input data to forecast the net load with one approach directly. Therefore, the present paper compares three net load forecasting approaches that exploit different levels of aggregation. We compare an aggregated strategy that directly forecasts the net load, a partially aggregated strategy that forecasts demand and supply separately, and a disaggregated strategy that forecasts demand and supply from each generator separately. We evaluate the forecast performance of all strategies with a simple and a complex forecasting model, both for deterministic and probabilistic forecasts, using one year of data from a simulated realistic future energy system characterised by a high share of renewable energy sources. We find that the partially aggregated strategy performs best, suggesting that a balance between specifically tailored forecasting models and aggregation is advantageous.

Abstract Sustainable energy systems are characterised by an increased integration of renewable energy sources, which magnifies the fluctuations in energy supply. Methods to to cope with these magnified fluctuations, such as load shifting, typically require accurate short-term load forecasts. Although numerous machine learning models have been developed to improve short-term load forecasting (STLF), these models often require large amounts of training data. Unfortunately, such data is usually not available, for example, due to new users or privacy concerns. Therefore, obtaining accurate short-term load forecasts with little data is a major challenge. The present paper thus proposes the latent space-based forecast enhancer (LSFE), a method which combines transfer learning and data augmentation to enhance STLF when training data is limited. The LSFE first trains a generative model on source data similar to the target data before using the latent space data representation of the target data to generate seed noise. Finally, we use this seed noise to generate synthetic data, which we combine with real data to enhance STLF. We evaluate the LSFE on real-world electricity data by examining the influence of its components, analysing its influence on obtained forecasts, and comparing its performance to benchmark models. We show that the Latent Space-based Forecast Enhancer is generally capable of improving the forecast accuracy and thus helps to successfully meet the challenge of limited available training data.

Abstract Capturing the uncertainty in probabilistic wind power forecasts is challenging, especially when uncertain input variables, such as the weather, play a role. Since ensemble weather predictions aim to capture the uncertainty in the weather system, they can be used to propagate this uncertainty through to subsequent wind power forecasting models. However, as weather ensemble systems are known to be biassed and underdispersed, meteorologists post‐process the ensembles. This post‐processing can successfully correct the biasses in the weather variables but has not been evaluated thoroughly in the context of subsequent forecasts, such as wind power generation forecasts. The present paper evaluates multiple strategies for applying ensemble post‐processing to probabilistic wind power forecasts. We use Ensemble Model Output Statistics (EMOS) as the post‐processing method and evaluate four possible strategies: only using the raw ensembles without post‐processing, a one‐step strategy where only the weather ensembles are post‐processed, a one‐step strategy where we only post‐process the power ensembles and a two‐step strategy where we post‐process both the weather and power ensembles. Results show that post‐processing the final wind power ensemble improves forecast performance regarding both calibration and sharpness whilst only post‐processing the weather ensembles does not necessarily lead to increased forecast performance.

Abstract Here we report two antimicrobial peptides (AMPs), HG2 and HG4 identified from a rumen microbiome metagenomic dataset, with activity against multidrug-resistant (MDR) bacteria, especially methicillin-resistant Staphylococcus aureus (MRSA) strains, a major hospital and community-acquired pathogen. We employed the classifier model design to analyse, visualise, and interpret AMP activities. This approach allowed in silico discrimination of promising lead AMP candidates for experimental evaluation. The lead AMPs, HG2 and HG4, are fast-acting and show anti-biofilm and anti-inflammatory activities in vitro and demonstrated little toxicity to human primary cell lines. The peptides were effective in vivo within a Galleria mellonella model of MRSA USA300 infection. In terms of mechanism of action, HG2 and HG4 appear to interact with the cytoplasmic membrane of target cells and may inhibit other cellular processes, whilst preferentially binding to bacterial lipids over human cell lipids. Therefore, these AMPs may offer additional therapeutic templates for MDR bacterial infections.

Abstract The morphometric assessment of the corneal subbasal nerve plexus (SNP) by confocal microscopy holds great potential as a sensitive biomarker for various ocular and systemic conditions and diseases. Automated wide-field montages (or large-area mosaic images) of the SNP provide an opportunity to overcome the limited field of view of the available imaging systems without the need for manual, subjective image selection for morphometric characterization. However, current wide-field montaging solutions usually calculate the mosaic image after the examination session, without a reliable means for the clinician to predict or estimate the resulting mosaic image quality during the examination. This contribution describes a novel approach for a real-time creation and visualization of a mosaic image of the SNP that facilitates an informed evaluation of the quality of the acquired image data immediately at the time of recording. In cases of insufficient data quality, the examination can be aborted and repeated immediately, while the patient is still at the microscope. Online mosaicking also offers the chance to identify an overlap of the imaged tissue region with previous SNP mosaic images, which can be particularly advantageous for follow-up examinations.

To shed light on the processes driving cell migration, a systematic analysis of the cell behavior is required. Since the manual analysis of hundreds or even thousands of cells is infeasible, automated approaches for cell segmentation and tracking are needed. While for the task of cell segmentation deep learning has become the standard, there are few approaches for simultaneous cell segmentation and tracking using deep learning. Here, we present EmbedTrack, a single convolutional neural network for simultaneous cell segmentation and tracking which predicts human comprehensible embeddings. As embeddings, offsets of cell pixels to their cell center and bandwidths are learned which are processed in a subsequent clustering step to generate an instance segmentation and link the segmented instances over time. We benchmark our approach on nine 2D data sets from the Cell Tracking Challenge, where our approach performs on seven out of nine data sets within the top 3 contestants including three top 1 performances. The source code is publicly available at https://git.scc.kit.edu/kit-loe-ge/embedtrack.

Supervised deep learning approaches for automated diagnosis support require datasets annotated by experts. Intra-annotator variability of a single annotator and inter-annotator variability between annotators can affect the quality of the diagnosis support. As medical experts will always differ in annotation details, quantitative studies concerning the annotation quality are of particular interest. A consistent and noise-free annotation of large-scale datasets by, for example, dermatologists or pathologists is a current challenge. Hence, methods are needed to automatically inspect annotations in datasets. In this paper, we categorize annotation noise in image segmentation tasks, present methods to simulate annotation noise, and examine the impact on the segmentation quality. Two novel automated methods to identify intra-annotator and inter-annotator inconsistencies based on uncertainty-aware deep neural networks are proposed. We demonstrate the benefits of our automated inspection methods such as focused re-inspection of noisy annotations or the detection of generally different annotation styles using the biomedical ISIC 2017 Melanoma image segmentation dataset.

Measurement(s) Prominent corneal nerve fibers • Dendritic Cell Technology Type(s) confocal microscopy Factor Type(s) disease status • comorbidity Sample Characteristic - Organism Homo sapiens Machine-accessible metadata file describing the reported data: https://doi.org/10.6084/m9.figshare.16691452

Abstract The circadian clock is a cellular mechanism that synchronizes various biological processes with respect to the time of the day. While much progress has been made characterizing the molecular mechanisms underlying this clock, it is less clear how external light cues influence the dynamics of the core clock mechanism and thereby entrain it with the light–dark cycle. Zebrafish-derived cell cultures possess clocks that are directly light-entrainable, thus providing an attractive laboratory model for circadian entrainment. Here, we have developed a stochastic oscillator model of the zebrafish circadian clock, which accounts for the core clock negative feedback loop, light input, and the proliferation of single-cell oscillator noise into population-level luminescence recordings. The model accurately predicts the entrainment dynamics observed in bioluminescent clock reporter assays upon exposure to a wide range of lighting conditions. Furthermore, we have applied the model to obtain refitted parameter sets for cell cultures exposed to a variety of pharmacological treatments and predict changes in single-cell oscillator parameters. Our work paves the way for model-based, large-scale screens for genetic or pharmacologically-induced modifications to the entrainment of circadian clock function.

Downhill skiing technique represents the complex coordinative movement patterns needed to control skiing motion. While scientific understanding of skiing technique is still incomplete, not least due to challenges in objectively measuring it, practitioners such as ski instructors have developed sophisticated and comprehensive descriptions of skiing technique. The current paper describes a 3-step proof-of-concept study introducing a technology platform for quantifying skiing technique that utilizes the practitioners’ expert knowledge. The approach utilizes an inertial measurement unit system (Xsens™) and presents a motion analysis algorithm based on the Principal Movement (PM) concept. In step 1, certified ski instructors skied specified technique elements according to technique variations described in ski instruction curricula. The obtained data was used to establish a PM-coordinate system for skiing movements. In step 2, the techniques parallel and carving turns were compared. Step 3 presents a case study where the technique analysis methodology is applied to advise an individual skier on potential technique improvements. All objectives of the study were met, proving the suitability of the proposed technology for scientific and applied technique evaluations of downhill skiing. The underlying conceptual approach - utilizing expert knowledge and skills to generate tailored variability in motion data (step 1) that then dominate the orientation of the PMs, which, in turn, can serve as measures for technique elements of interest - could be applied in many other sports or for other applications in human movement analyses.

Within the domain of analyzing powder X-ray diffraction (XRD) scans, manual examination of the recorded data is still the most popular method, but it requires some expertise and is time consuming. The usual workflow for the phase-identification task involves software for searching databases of known compounds and matching lists of d spacings and related intensities to the measured data. Most automated approaches apply some iterative procedure for the search/match process but fail to be generally reliable yet without the manual validation step of an expert. Recent advances in the field of machine and deep learning have led to the development of algorithms for use with diffraction patterns and are producing promising results in some applications. A limitation, however, is that thousands of training samples are required for the model to achieve a reliable performance and not enough measured samples are available. Accordingly, a framework for the efficient generation of thousands of synthetic XRD scans is presented which considers typical effects in realistic measurements and thus simulates realistic patterns for the training of machine- or deep-learning models. The generated data set can be applied to any machine- or deep-learning structure as training data so that the models learn to analyze measured XRD data based on synthetic diffraction patterns. Consequently, we train a convolutional neural network with the simulated diffraction patterns for application with iron ores or cements compounds and prove robustness against varying unit-cell parameters, preferred orientation and crystallite size in synthetic, as well as measured, XRD scans.

Abstract Small fiber neuropathy (SFN) has been suggested as a trigger of restless legs syndrome (RLS). An increased prevalence of peripheral neuropathy has been demonstrated in Parkinson’s disease (PD). We aimed to investigate, in a cross-sectional manner, whether SFN is overrepresented in PD patients with concurrent RLS relative to PD patients without RLS, using in vivo corneal confocal microscopy (IVCCM) and quantitative sensory testing (QST) as part of small fiber assessment. Study participants comprised of age- and sex-matched PD patients with (n = 21) and without RLS (n = 21), and controls (n = 13). Diagnosis of RLS was consolidated with the sensory suggested immobilization test. Assessments included nerve conduction studies (NCS), Utah Early Neuropathy Scale (UENS), QST, and IVCCM, with automated determination of corneal nerve fiber length (CNFL) and branch density (CNBD) from wide-area mosaics of the subbasal nerve plexus. Plasma neurofilament light (p-NfL) was determined as a measure of axonal degeneration. No significant differences were found between groups when comparing CNFL (p = 0.81), CNBD (p = 0.92), NCS (p = 0.82), and QST (minimum p = 0.54). UENS scores, however, differed significantly (p = 0.001), with post-hoc pairwise testing revealing higher scores in both PD groups relative to controls (p = 0.018 and p = 0.001). Analysis of all PD patients (n = 42) revealed a correlation between the duration of l-dopa therapy and CNBD (ρ = −0.36, p = 0.022), and p-NfL correlated with UENS (ρ = 0.35, p = 0.026) and NCS (ρ = −0.51, p = 0.001). Small and large fiber neuropathy do not appear to be associated with RLS in PD. Whether peripheral small and/or large fiber pathology associates with central neurodegeneration in PD merits further longitudinal studies.

IntroductionAbout 30% of epilepsy patients are resistant to treatment with antiepileptic drugs, and only a minority of these are surgical candidates. A recent therapeutic approach is the application of electrical stimulation in the early phases of a seizure to interrupt its spread across the brain. To accomplish this, energy-efficient seizure detectors are required that are able to detect a seizure in its early stages.MethodsThree patient-specific, energy-efficient seizure detectors are proposed in this study: (i) random forest (RF); (ii) long short-term memory (LSTM) recurrent neural network (RNN); and (iii) convolutional neural network (CNN). Performance evaluation was based on EEG data (n = 40 patients) derived from a selected set of surface EEG electrodes, which mimic the electrode layout of an implantable neurostimulation system. As for the RF input, 16 features in the time- and frequency-domains were selected. Raw EEG data were used for both CNN and RNN. Energy consumption was estimated by a platform-independent model based on the number of arithmetic operations (AOs) and memory accesses (MAs). To validate the estimated energy consumption, the RNN classifier was implemented on an ultra-low-power microcontroller.ResultsThe RNN seizure detector achieved a slightly better level of performance, with a median area under the precision-recall curve score of 0.49, compared to 0.47 for CNN and 0.46 for RF. In terms of energy consumption, RF was the most efficient algorithm, with a total of 67k AOs and 67k MAs per classification. This was followed by CNN (488k AOs and 963k MAs) and RNN (772k AOs and 978k MAs), whereby MAs contributed more to total energy consumption. Measurements derived from the hardware implementation of the RNN algorithm demonstrated a significant correlation between estimations and actual measurements.DiscussionAll three proposed seizure detection algorithms were shown to be suitable for application in implantable devices. The applied methodology for a platform-independent energy estimation was proven to be accurate by way of hardware implementation of the RNN algorithm. These findings show that seizure detection can be achieved using just a few channels with limited spatial distribution. The methodology proposed in this study can therefore be applied when designing new models for responsive neurostimulation.

In biotechnology, cell growth is one of the most important properties for the characterization and optimization of microbial cultures. Novel live-cell imaging methods are leading to an ever better understanding of cell cultures and their development. The key to analyzing acquired data is accurate and automated cell segmentation at the single-cell level. Therefore, we present microbeSEG, a user-friendly Python-based cell segmentation tool with a graphical user interface and OMERO data management. microbeSEG utilizes a state-of-the-art deep learning-based segmentation method and can be used for instance segmentation of a wide range of cell morphologies and imaging techniques, e.g., phase contrast or fluorescence microscopy. The main focus of microbeSEG is a comprehensible, easy, efficient, and complete workflow from the creation of training data to the final application of the trained segmentation model. We demonstrate that accurate cell segmentation results can be obtained within 45 minutes of user time. Utilizing public segmentation datasets or pre-labeling further accelerates the microbeSEG workflow. This opens the door for accurate and efficient data analysis of microbial cultures.

Modelling, forecasting and overall understanding of the dynamics of the power grid and its frequency are essential for the safe operation of existing and future power grids. Much previous research was focused on large continental areas, while small systems, such as islands are less well-studied. These natural island systems are ideal testing environments for microgrid proposals and artificially islanded grid operation. In the present paper, we utilise measurements of the power grid frequency obtained in European islands: the Faroe Islands, Ireland, the Balearic Islands and Iceland and investigate how their frequency can be predicted, compared to the Nordic power system, acting as a reference. The Balearic Islands are found to be particularly deterministic and easy to predict in contrast to hard-to-predict Iceland. Furthermore, we show that typically 2–4 weeks of data are needed to improve prediction performance beyond simple benchmarks.

Antimicrobial peptides (AMPs) are a promising class of compounds being developed against multi-drug resistant bacteria. Hybridization has been reported to increase antimicrobial activity. Here, two proline-rich peptides (consP1: VRKPPYLPRPRPRPL-CONH2 and Bac5-v291: RWRRPIRRRPIRPPFWR-CONH2) were combined with two arginine-isoleucine-rich peptides (optP1: KIILRIRWR-CONH2 and optP7: KRRVRWIIW-CONH2). Proline-rich antimicrobial peptides (PrAMPs) are known to inhibit the bacterial ribosome, shown also for Bac5-v291, whereas it is hypothesized a “dirty drug” model for the arginine-isoleucine-rich peptides. That hypothesis was underpinned by transmission electron microscopy and biological small-angle X-ray scattering (BioSAXS). The strength of BioSAXS is the power to detect ultrastructural changes in millions of cells in a short time (seconds) in a high-throughput manner. This information can be used to classify antimicrobial compounds into groups according to the ultrastructural changes they inflict on bacteria and how the bacteria react towards that assault. Based on previous studies, this correlates very well with different modes of action. Due to the novelty of this approach direct identification of the target of the antimicrobial compound is not yet fully established, more research is needed. More research is needed to address this limitation. The hybrid peptides showed a stronger antimicrobial activity compared to the proline-rich peptides, except when compared to Bac5-v291 against E. coli. The increase in activity compared to the arginine-isoleucine-rich peptides was up to 6-fold, however, it was not a general increase but was dependent on the combination of peptides and bacteria. BioSAXS experiments revealed that proline-rich peptides and arginine-isoleucine-rich peptides induce very different ultrastructural changes in E. coli, whereas a hybrid peptide (hyP7B5GK) shows changes, different to both parental peptides and the untreated control. These different ultrastructural changes indicated that the mode of action of the parental peptides might be different from each other as well as from the hybrid peptide hyP7B5GK. All peptides showed very low haemolytic activity, some of them showed a 100-fold or larger therapeutic window, demonstrating the potential for further drug development.

Automatic cell segmentation and tracking enables to gain quantitative insights into the processes driving cell migration. To investigate new data with minimal manual effort, cell tracking algorithms should be easy to apply and reduce manual curation time by providing automatic correction of segmentation errors. Current cell tracking algorithms, however, are either easy to apply to new data sets but lack automatic segmentation error correction, or have a vast set of parameters that needs either manual tuning or annotated data for parameter tuning. In this work, we propose a tracking algorithm with only few manually tunable parameters and automatic segmentation error correction. Moreover, no training data is needed. We compare the performance of our approach to three well-performing tracking algorithms from the Cell Tracking Challenge on data sets with simulated, degraded segmentation-including false negatives, over- and under-segmentation errors. Our tracking algorithm can correct false negatives, over- and under-segmentation errors as well as a mixture of the aforementioned segmentation errors. On data sets with under-segmentation errors or a mixture of segmentation errors our approach performs best. Moreover, without requiring additional manual tuning, our approach ranks several times in the top 3 on the 6th edition of the Cell Tracking Challenge.

When approaching thyroid gland tumor classification, the differentiation between samples with and without "papillary thyroid carcinoma-like" nuclei is a daunting task with high inter-observer variability among pathologists. Thus, there is increasing interest in the use of machine learning approaches to provide pathologists real-time decision support. In this paper, we optimize and quantitatively compare two automated machine learning methods for thyroid gland tumor classification on two datasets to assist pathologists in decision-making regarding these methods and their parameters. The first method is a feature-based classification originating from common image processing and consists of cell nucleus segmentation, feature extraction, and subsequent thyroid gland tumor classification utilizing different classifiers. The second method is a deep learning-based classification which directly classifies the input images with a convolutional neural network without the need for cell nucleus segmentation. On the Tharun and Thompson dataset, the feature-based classification achieves an accuracy of 89.7% (Cohen's Kappa 0.79), compared to the deep learning-based classification of 89.1% (Cohen's Kappa 0.78). On the Nikiforov dataset, the feature-based classification achieves an accuracy of 83.5% (Cohen's Kappa 0.46) compared to the deep learning-based classification 77.4% (Cohen's Kappa 0.35). Thus, both automated thyroid tumor classification methods can reach the classification level of an expert pathologist. To our knowledge, this is the first study comparing feature-based and deep learning-based classification regarding their ability to classify samples with and without papillary thyroid carcinoma-like nuclei on two large-scale datasets.

The glucose-sensing Mondo pathway regulates expression of metabolic genes in mammals. Here, we characterized its function in the zebrafish and revealed an unexpected role of this pathway in vertebrate embryonic development. We showed that knockdown of mondoa impaired the early morphogenetic movement of epiboly in zebrafish embryos and caused microtubule defects. Expression of genes in the terpenoid backbone and sterol biosynthesis pathways upstream of pregnenolone synthesis was coordinately downregulated in these embryos, including the most downregulated gene nsdhl. Loss of Nsdhl function likewise impaired epiboly, similar to MondoA loss of function. Both epiboly and microtubule defects were partially restored by pregnenolone treatment. Maternal-zygotic mutants of mondoa showed perturbed epiboly with low penetrance and compensatory changes in the expression of terpenoid/sterol/steroid metabolism genes. Collectively, our results show a novel role for MondoA in the regulation of early vertebrate development, connecting glucose, cholesterol and steroid hormone metabolism with early embryonic cell movements.

Cranial neural crest (NC) contributes to the developing vertebrate eye. By multidimensional, quantitative imaging, we traced the origin of the ocular NC cells to two distinct NC populations that differ in the maintenance of sox10 expression, Wnt signalling, origin, route, mode and destination of migration. The first NC population migrates to the proximal and the second NC cell group populates the distal (anterior) part of the eye. By analysing zebrafish pax6a/b compound mutants presenting anterior segment dysgenesis, we demonstrate that Pax6a/b guide the two NC populations to distinct proximodistal locations. We further provide evidence that the lens whose formation is pax6a/b-dependent and lens-derived TGFβ signals contribute to the building of the anterior segment. Taken together, our results reveal multiple roles of Pax6a/b in the control of NC cells during development of the anterior segment.

Orientation analysis of fibers is widely applied in the fields of medical, material and life sciences. The orientation information allows predicting properties and behavior of materials to validate and guide a fabrication process of materials with controlled fiber orientation. Meanwhile, development of detector systems for high-resolution non-invasive 3D imaging techniques led to a significant increase in the amount of generated data per a sample up to dozens of gigabytes. Though plenty of 3D orientation estimation algorithms were developed in recent years, neither of them can process large datasets in a reasonable amount of time. This fact complicates the further analysis and makes impossible fast feedback to adjust fabrication parameters. In this work, we present a new method for quantifying the 3D orientation of fibers. The GPU implementation of the proposed method surpasses another popular method for 3D orientation analysis regarding accuracy and speed. The validation of both methods was performed on a synthetic dataset with varying parameters of fibers. Moreover, the proposed method was applied to perform orientation analysis of scaffolds with different fibrous micro-architecture studied with the synchrotron μCT imaging setup. Each acquired dataset of size 600x600x450 voxels was analyzed in less 2 minutes using standard PC equipped with a single GPU.

The accurate segmentation and tracking of cells in microscopy image sequences is an important task in biomedical research, e.g., for studying the development of tissues, organs or entire organisms. However, the segmentation of touching cells in images with a low signal-to-noise-ratio is still a challenging problem. In this paper, we present a method for the segmentation of touching cells in microscopy images. By using a novel representation of cell borders, inspired by distance maps, our method is capable to utilize not only touching cells but also close cells in the training process. Furthermore, this representation is notably robust to annotation errors and shows promising results for the segmentation of microscopy images containing in the training data underrepresented or not included cell types. For the prediction of the proposed neighbor distances, an adapted U-Net convolutional neural network (CNN) with two decoder paths is used. In addition, we adapt a graph-based cell tracking algorithm to evaluate our proposed method on the task of cell tracking. The adapted tracking algorithm includes a movement estimation in the cost function to re-link tracks with missing segmentation masks over a short sequence of frames. Our combined tracking by detection method has proven its potential in the IEEE ISBI 2020 Cell Tracking Challenge (http://celltrackingchallenge.net/) where we achieved as team KIT-Sch-GE multiple top three rankings including two top performances using a single segmentation model for the diverse data sets.

Abstract The capability of corneal confocal microscopy (CCM) to acquire high-resolution in vivo images of the densely innervated human cornea has gained considerable interest in using this non-invasive technique as an objective diagnostic tool for staging peripheral neuropathies. Morphological alterations of the corneal subbasal nerve plexus (SNP) assessed by CCM have been shown to correlate well with the progression of neuropathic diseases and even predict future-incident neuropathy. Since the field of view of single CCM images is insufficient for reliable characterisation of nerve morphology, several image mosaicking techniques have been developed to facilitate the assessment of the SNP in large-area visualisations. Due to the limited depth of field of confocal microscopy, these approaches are highly sensitive to small deviations of the focus plane from the SNP layer. Our contribution proposes a new automated solution, combining guided eye movements for rapid expansion of the acquired SNP area and axial focus plane oscillations to guarantee complete imaging of the SNP. We present results of a feasibility study using the proposed setup to evaluate different oscillation settings. By comparing different image selection approaches, we show that automatic tissue classification algorithms are essential to create high-quality mosaic images from the acquired 3D datasets.

Abstract The present article describes a concept for the creation and application of energy forecasting models in a distributed environment. Additionally, a benchmark comparing the time required for the training and application of data-driven forecasting models on a single computer and a computing cluster is presented. This comparison is based on a simulated dataset and both R and Apache Spark are used. Furthermore, the obtained results show certain points in which the utilization of distributed computing based on Spark may be advantageous.

Magnetic resonance tomography typically applies the Fourier transform to k-space signals repeatedly acquired from a frequency encoded spatial region of interest, therefore requiring a stationary object during scanning. Any movement of the object results in phase errors in the recorded signal, leading to deformed images, phantoms, and artifacts, since the encoded information does not originate from the intended region of the object. However, if the type and magnitude of movement is known instantaneously, the scanner or the reconstruction algorithm could be adjusted to compensate for the movement, directly allowing high quality imaging with non-stationary objects. This would be an enormous boon to studies that tie cell metabolomics to spontaneous organism behaviour, eliminating the stress otherwise necessitated by restraining measures such as anesthesia or clamping. In the present theoretical study, we use a phantom of the animal model C. elegans to examine the feasibility to automatically predict its movement and position, and to evaluate the impact of movement prediction, within a sufficiently long time horizon, on image reconstruction. For this purpose, we use automated image processing to annotate body parts in freely moving C. elegans, and predict their path of movement. We further introduce an MRI simulation platform based on bright field videos of the moving worm, combined with a stack of high resolution transmission electron microscope (TEM) slice images as virtual high resolution phantoms. A phantom provides an indication of the spatial distribution of signal-generating nuclei on a particular imaging slice. We show that adjustment of the scanning to the predicted movements strongly reduces distortions in the resulting image, opening the door for implementation in a high-resolution NMR scanner.

With the rise of various multidrug-resistant (MDR) pathogenic bacteria, worldwide health care is under pressure to respond. Conventional antibiotics are failing and the development of novel classes and alternative strategies is a major priority. Antimicrobial peptides (AMPs) cannot only kill MDR bacteria, but also can be used synergistically with conventional antibiotics. We selected 30 short AMPs from different origins and measured their synergy in combination with polymyxin B, piperacillin, ceftazidime, cefepime, meropenem, imipenem, tetracycline, erythromycin, kanamycin, tobramycin, amikacin, gentamycin, and ciprofloxacin. In total, 403 unique combinations were tested against an MDR Pseudomonas aeruginosa isolate (PA910). As a measure of the synergistic effects, fractional inhibitory concentrations (FICs) were determined using microdilution assays with FICs ranges between 0.25 and 2. A high number of combinations between peptides and polymyxin B, erythromycin, and tetracycline were found to be synergistic. Novel variants of indolicidin also showed a high frequency in synergist interaction. Single amino acid substitutions within the peptides can have a very strong effect on the ability to synergize, making it possible to optimize future drugs toward synergistic interaction.

Two highly active short broad-spectrum AMPs (14D and 69D) with unknown mode of action have been investigated in regards to their effect against the Gram-negative bacteria Escherichia coli and the Gram-positive bacteria methicillin-resistant Staphylococcus aureus (MRSA). Minimal inhibitory concentration (MIC) measurements using a cell density of 108 cfu/ml resulted in values between 16 and 32 µg/ml. Time-kill experiments using 108 cfu/ml revealed complete killing, except for 69D in combination with MRSA, where bacterial load was reduced a million times. Small-angle X-ray scattering of biological samples (BioSAXS) at 108 cfu/ml was applied to investigate the ultrastructural changes in E. coli and MRSA in response to these two broad-spectrum AMPs. In addition, electron microscopy (EM) was performed to visualize the treated and non-treated bacteria. As expected, the scattering curves generated using BioSAXS show the ultrastructure of the Gram-positive and Gram-negative bacteria to be very different (BioSAXS is not susceptible to the outer shape). After treatment with either peptide, the scattering curves of E. coli and MRSA cells are much more alike. Whereas in EM, it is notoriously difficult to observe changes for spherical Gram-positives; the BioSAXS results are superior and reveal strongly similar effects for both peptides induced in Gram-positive as well as Gram-negative bacteria. Given the high-throughput possibility and robust statistics, BioSAXS can support and speed up mode of action research in AMPs and other antimicrobial compounds, making a contribution toward the development of urgently needed drugs against resistant bacteria.

Hybrid 3D scaffolds composed of different biomaterials with fibrous structure or enriched with different inclusions (i.e., nano- and microparticles) have already demonstrated their positive effect on cell integration and regeneration. The analysis of fibers in hybrid biomaterials, especially in a 3D space is often difficult due to their various diameters (from micro to nanoscale) and compositions. Though biomaterials processing workflows are implemented, there are no software tools for fiber analysis that can be easily integrated into such workflows. Due to the demand for reproducible science with Jupyter notebooks and the broad use of the Python programming language, we have developed the new Python package quanfima offering a complete analysis of hybrid biomaterials, that include the determination of fiber orientation, fiber and/or particle diameter and porosity. Here, we evaluate the provided tensor-based approach on a range of generated datasets under various noise conditions. Also, we show its application to the X-ray tomography datasets of polycaprolactone fibrous scaffolds pure and containing silicate-substituted hydroxyapatite microparticles, hydrogels enriched with bioglass contained strontium and alpha-tricalcium phosphate microparticles for bone tissue engineering and porous cryogel 3D scaffold for pancreatic cell culturing. The results obtained with the help of the developed package demonstrated high accuracy and performance of orientation, fibers and microparticles diameter and porosity analysis.

I-CARE is a hand-held activation system that allows professional and informal caregivers to cognitively and socially activate people with dementia in joint activation sessions without special training or expertise. I-CARE consists of an easy-to-use tablet application that presents activation content and a server-based backend system that securely manages the contents and events of activation sessions. It tracks various sources of explicit and implicit feedback from user interactions and different sensors to estimate which content is successful in activating individual users. Over the course of use, I-CARE’s recommendation system learns about the individual needs and resources of its users and automatically personalizes the activation content. In addition, information about past sessions can be retrieved such that activations seamlessly build on previous sessions while eligible stakeholders are informed about the current state of care and daily form of their protegees. In addition, caregivers can connect with supervisors and professionals through the I-CARE remote calling feature, to get activation sessions tracked in real time via audio and video support. In this way, I-CARE provides technical support for a decentralized and spontaneous formation of ad hoc activation groups and fosters tight engagement of the social network and caring community. By these means, I-CARE promotes new care infrastructures in the community and the neighborhood as well as relieves professional and informal caregivers.

Signaling filopodia, termed cytonemes, are dynamic actin-based membrane structures that regulate the exchange of signaling molecules and their receptors within tissues. However, how cytoneme formation is regulated remains unclear. Here, we show that Wnt/planar cell polarity (PCP) autocrine signaling controls the emergence of cytonemes, and that cytonemes subsequently control paracrine Wnt/β-catenin signal activation. Upon binding of the Wnt family member Wnt8a, the receptor tyrosine kinase Ror2 becomes activated. Ror2/PCP signaling leads to the induction of cytonemes, which mediate the transport of Wnt8a to neighboring cells. In the Wnt-receiving cells, Wnt8a on cytonemes triggers Wnt/β-catenin-dependent gene transcription and proliferation. We show that cytoneme-based Wnt transport operates in diverse processes, including zebrafish development, murine intestinal crypt and human cancer organoids, demonstrating that Wnt transport by cytonemes and its control via the Ror2 pathway is highly conserved in vertebrates.

State-of-the-art light-sheet and confocal microscopes allow recording of entire embryos in 3D and over time (3D+t) for many hours. Fluorescently labeled structures can be segmented and tracked automatically in these terabyte-scale 3D+t images, resulting in thousands of cell migration trajectories that provide detailed insights to large-scale tissue reorganization at the cellular level. Here we present EmbryoMiner, a new interactive open-source framework suitable for in-depth analyses and comparisons of entire embryos, including an extensive set of trajectory features. Starting at the whole-embryo level, the framework can be used to iteratively focus on a region of interest within the embryo, to investigate and test specific trajectory-based hypotheses and to extract quantitative features from the isolated trajectories. Thus, the new framework provides a valuable new way to quantitatively compare corresponding anatomical regions in different embryos that were manually selected based on biological prior knowledge. As a proof of concept, we analyzed 3D+t light-sheet microscopy images of zebrafish embryos, showcasing potential user applications that can be performed using the new framework.

[This corrects the article DOI: 10.1371/journal.pone.0187535.].

This work aims at classifying the road condition with data mining methods using simple acceleration sensors and gyroscopes installed in vehicles. Two classifiers are developed with a support vector machine (SVM) to distinguish between different types of road surfaces, such as asphalt and concrete, and obstacles, such as potholes or railway crossings. From the sensor signals, frequency-based features are extracted, evaluated automatically with MANOVA. The selected features and their meaning to predict the classes are discussed. The best features are used for designing the classifiers. Finally, the methods, which are developed and applied in this work, are implemented in a Matlab toolbox with a graphical user interface. The toolbox visualizes the classification results on maps, thus enabling manual verification of the results. The accuracy of the cross-validation of classifying obstacles yields 81.0% on average and of classifying road material 96.1% on average. The results are discussed on a comprehensive exemplary data set.

Many automatically analyzable scientific questions are well-posed and a variety of information about expected outcomes is available a priori. Although often neglected, this prior knowledge can be systematically exploited to make automated analysis operations sensitive to a desired phenomenon or to evaluate extracted content with respect to this prior knowledge. For instance, the performance of processing operators can be greatly enhanced by a more focused detection strategy and by direct information about the ambiguity inherent in the extracted data. We present a new concept that increases the result quality awareness of image analysis operators by estimating and distributing the degree of uncertainty involved in their output based on prior knowledge. This allows the use of simple processing operators that are suitable for analyzing large-scale spatiotemporal (3D+t) microscopy images without compromising result quality. On the foundation of fuzzy set theory, we transform available prior knowledge into a mathematical representation and extensively use it to enhance the result quality of various processing operators. These concepts are illustrated on a typical bioimage analysis pipeline comprised of seed point detection, segmentation, multiview fusion and tracking. The functionality of the proposed approach is further validated on a comprehensive simulated 3D+t benchmark data set that mimics embryonic development and on large-scale light-sheet microscopy data of a zebrafish embryo. The general concept introduced in this contribution represents a new approach to efficiently exploit prior knowledge to improve the result quality of image analysis pipelines. The generality of the concept makes it applicable to practically any field with processing strategies that are arranged as linear pipelines. The automated analysis of terabyte-scale microscopy data will especially benefit from sophisticated and efficient algorithms that enable a quantitative and fast readout.

The parametrization of automatic image processing routines is time-consuming if a lot of image processing parameters are involved. An expert can tune parameters sequentially to get desired results. This may not be productive for applications with difficult image analysis tasks, e.g. when high noise and shading levels in an image are present or images vary in their characteristics due to different acquisition conditions. Parameters are required to be tuned simultaneously. We propose a framework to improve standard image segmentation methods by using feedback-based automatic parameter adaptation. Moreover, we compare algorithms by implementing them in a feedforward fashion and then adapting their parameters. This comparison is proposed to be evaluated by a benchmark data set that contains challenging image distortions in an increasing fashion. This promptly enables us to compare different standard image segmentation algorithms in a feedback vs. feedforward implementation by evaluating their segmentation quality and robustness. We also propose an efficient way of performing automatic image analysis when only abstract ground truth is present. Such a framework evaluates robustness of different image processing pipelines using a graded data set. This is useful for both end-users and experts.

Developers of image processing routines rely on benchmark data sets to give qualitative comparisons of new image analysis algorithms and pipelines. Such data sets need to include artifacts in order to occlude and distort the required information to be extracted from an image. Robustness, the quality of an algorithm related to the amount of distortion is often important. However, using available benchmark data sets an evaluation of illumination robustness is difficult or even not possible due to missing ground truth data about object margins and classes and missing information about the distortion. We present a new framework for robustness evaluation. The key aspect is an image benchmark containing 9 object classes and the required ground truth for segmentation and classification. Varying levels of shading and background noise are integrated to distort the data set. To quantify the illumination robustness, we provide measures for image quality, segmentation and classification success and robustness. We set a high value on giving users easy access to the new benchmark, therefore, all routines are provided within a software package, but can as well easily be replaced to emphasize other aspects.

Tissue injury and infection trigger innate immune responses. However, dysregulation may result in chronic inflammation and is commonly treated with corticosteroids and non-steroidal anti-inflammatory drugs. Unfortunately, long-term administration of both therapeutic classes can cause unwanted side effects. To identify alternative immune-modulatory compounds we have previously established a novel screening method using zebrafish larvae. Using this method we here present results of an in vivo high-content drug-repurposing screen, identifying 63 potent anti-inflammatory drugs that are in clinical use for other indications. Our approach reveals a novel pro-inflammatory role of nitric oxide. Nitric oxide affects leukocyte recruitment upon peripheral sensory nervous system or epithelial injury in zebrafish larvae both via soluble guanylate cyclase and in a soluble guanylate cyclase -independent manner through protein S-nitrosylation. Together, we show that our screening method can help to identify novel immune-modulatory activities and provide new mechanistic insights into the regulation of inflammatory processes.

Automated analysis of multi-dimensional microscopy images has become an integral part of modern research in life science. Most available algorithms that provide sufficient segmentation quality, however, are infeasible for a large amount of data due to their high complexity. In this contribution we present a fast parallelized segmentation method that is especially suited for the extraction of stained nuclei from microscopy images, e.g., of developing zebrafish embryos. The idea is to transform the input image based on gradient and normal directions in the proximity of detected seed points such that it can be handled by straightforward global thresholding like Otsu's method. We evaluate the quality of the obtained segmentation results on a set of real and simulated benchmark images in 2D and 3D and show the algorithm's superior performance compared to other state-of-the-art algorithms. We achieve an up to ten-fold decrease in processing times, allowing us to process large data sets while still providing reasonable segmentation results.