Search

For several years, the determination of a differential cell count of a raw milk sample has been proposed as a more accurate tool for monitoring the udder health of dairy cows compared with using the absolute somatic cell count. However, the required sample preparation and staining process can be labor- and cost-intensive. Therefore, the aim of our study was to demonstrate the feasibility of analyzing unlabeled blood and milk leukocytes from dairy cows by means of digital holographic microscopy (DHM). For this, we trained three different machine learning methods, i.e., k-Nearest Neighbor, Random Forests, and Support Vector Machine, on sorted leukocyte populations (granulocytes, lymphocytes, and monocytes/macrophages) isolated from blood and milk samples of three dairy cows by using fluorescence-activated cell sorting. Afterward, those classifiers were applied to differentiate unlabeled blood and milk samples analyzed by DHM. A total of 70 blood and 70 milk samples were used. Those samples were collected from five clinically healthy cows at 14-time points within a study period of 26 days. The outcome was compared with the results of the same samples analyzed by flow cytometry and (in the case of blood samples) also to routine analysis in an external laboratory. Moreover, a standard vaccination was used as an immune stimulus during the study to check for changes in cell morphology or cell counts. When applied to isolated leukocytes, Random Forests performed best, with a specificity of 0.93 for blood and 0.84 for milk cells and a sensitivity of 0.90 and 0.81, respectively. Although the results of the three analytical methods differed, it could be demonstrated that a DHM analysis is applicable for blood and milk leukocyte samples with high reliability. Compared with the flow cytometric results, Random Forests showed an MAE of 0.11 (SD = 0.04), an RMSE of 0.13 (SD = 0.14), and an MRE of 1.00 (SD = 1.11) for all blood leukocyte counts and an MAE of 0.20 (SD = 0.11), an RMSE of 0.21 (SD = 0.11) and an MRE of 1.95 (SD = 2.17) for all milk cell populations. Further studies with larger sample sizes and varying immune cell compositions are required to establish method-specific reference ranges.

Abstract Background The clinical spectrum of acute SARS-CoV-2 infection ranges from an asymptomatic to life-threatening disease. Considering the broad spectrum of severity, reliable biomarkers are required for early risk stratification and prediction of clinical outcomes. Despite numerous efforts, no COVID-19-specific biomarker has been established to guide further diagnostic or even therapeutic approaches, most likely due to insufficient validation, methodical complexity, or economic factors. COVID-19-associated coagulopathy is a hallmark of the disease and is mainly attributed to dysregulated immunothrombosis. This process describes an intricate interplay of platelets, innate immune cells, the coagulation cascade, and the vascular endothelium leading to both micro- and macrothrombotic complications. In this context, increased levels of immunothrombotic components, including platelet and platelet-leukocyte aggregates, have been described and linked to COVID-19 severity. Methods Here, we describe a label-free quantitative phase imaging approach, allowing the identification of cell-aggregates and their components at single-cell resolution within 30 min, which prospectively qualifies the method as point-of-care (POC) testing. Results We find a significant association between the severity of COVID-19 and the amount of platelet and platelet-leukocyte aggregates. Additionally, we observe a linkage between severity, aggregate composition, and size distribution of platelets in aggregates. Conclusions This study presents a POC-compatible method for rapid quantitative analysis of blood cell aggregates in patients with COVID-19.

In the last decades, data-driven methods have gained great popularity in the industry, supported by state-of-the-art advancements in machine learning. These methods require a large quantity of labeled data, which is difficult to obtain and mostly costly and challenging. To address these challenges, researchers have turned their attention to unsupervised and few-shot learning methods, which produced encouraging results, particularly in the areas of computer vision and natural language processing. With the lack of pretrained models, time series feature learning is still considered as an open area of research. This paper presents an efficient two-stage feature learning approach for anomaly detection in machine processes, based on a prototype few-shot learning technique that requires a limited number of labeled samples. The work is evaluated on a real-world scenario using the publicly available CNC Machining dataset. The proposed method outperforms the conventional prototypical network and the feature analysis shows a high generalization ability achieving an F1-score of 90.3%. The comparison with handcrafted features proves the robustness of the deep features and their invariance to data shifts across machines and time periods, which makes it a reliable method for sensory industrial applications.

Modern automation systems largely rely on closed loop control, wherein a controller interacts with a controlled process via actions, based on observations. These systems are increasingly complex, yet most deployed controllers are linear Proportional-Integral-Derivative (PID) controllers. PID controllers perform well on linear and near-linear systems but their simplicity is at odds with the robustness required to reliably control complex processes. Modern machine learning techniques offer a way to extend PID controllers beyond their linear control capabilities by using neural networks. However, such an extension comes at the cost of losing stability guarantees and controller interpretability. In this paper, we examine the utility of extending PID controllers with recurrent neural networks--namely, General Dynamic Neural Networks (GDNN); we show that GDNN (neural) PID controllers perform well on a range of complex control systems and highlight how they can be a scalable and interpretable option for modern control systems. To do so, we provide an extensive study using four benchmark systems that represent the most common control engineering benchmarks. All control environments are evaluated with and without noise as well as with and without disturbances. The neural PID controller performs better than standard PID control in 15 of 16 tasks and better than model-based control in 13 of 16 tasks. As a second contribution, we address the lack of interpretability that prevents neural networks from being used in real-world control processes. We use bounded-input bounded-output stability analysis to evaluate the parameters suggested by the neural network, making them understandable for engineers. This combination of rigorous evaluation paired with better interpretability is an important step towards the acceptance of neural-network-based control approaches for real-world systems. It is furthermore an important step towards interpretable and safely applied artificial intelligence.

In this contribution a reduced reference video quality metric for AVC/H.264 is proposed that needs only a very low overhead (not more than two bytes per sequence). This reduced reference metric uses well established algorithms to measure objective features of the video such as 'blur' or 'blocking'. Those measurements are then combined into a single measurement for the overall video quality. The weights of the single features and the combination of those are determined using methods provided by multivariate data analysis. The proposed metric is verified using a data set of AVC/H.264 encoded videos and the corresponding results of a carefully designed and conducted subjective evaluation. Results show that the proposed reduced reference metric not only outperforms standard PSNR but also two well known full reference metrics.