Synthetic Image Rendering Solves Annotation Problem in Deep Learning Nanoparticle Segmentation.

Nanoparticles occur in various environments as a consequence of man-made processes, which raises concerns about their impact on the environment and human health. To allow for proper risk assessment, a precise and statistically relevant analysis of particle characteristics (such as size, shape, and composition) is required that would greatly benefit from automated image analysis procedures. While deep learning shows impressive results in object detection tasks, its applicability is limited by the amount of representative, experimentally collected and manually annotated training data. Here, an elegant, flexible, and versatile method to bypass this costly and tedious data acquisition process is presented. It shows that using a rendering software allows to generate realistic, synthetic training data to train a state-of-the art deep neural network. Using this approach, a segmentation accuracy can be derived that is comparable to man-made annotations for toxicologically relevant metal-oxide nanoparticle ensembles which were chosen as examples. The presented study paves the way toward the use of deep learning for automated, high-throughput particle detection in a variety of imaging techniques such as in microscopies and spectroscopies, for a wide range of applications, including the detection of micro- and nanoplastic particles in water and tissue samples.

Chemokine/Cytokine Levels Correlate with Organ Involvement in PR3-ANCA-Associated Vasculitis.

BACKGROUND: ANCA-associated vasculitis (AAV) is a rare small vessel disease characterized by multi-organ involvement. Biomarkers that can measure specific organ involvement are missing. Here, we ask whether certain circulating cytokines and chemokines correlate with renal involvement and if distinct cytokine/chemokine patterns can differentiate between renal, ear/nose/throat, joints, and lung involvement of AAV. METHODS: Thirty-two sets of Birmingham vasculitis activity score (BVAS), PR3-ANCA titers, laboratory marker, and different cytokines were obtained from 17 different patients with AAV. BVAS, PR3-ANCA titers, laboratory marker, and cytokine concentrations were correlated to different organ involvements in active AAV. RESULTS: Among patients with active PR3-AAV (BVAS > 0) and kidney involvement we found significant higher concentrations of chemokine ligand (CCL)-1, interleukin (IL)-6, IL21, IL23, IL-28A, IL33, monocyte chemoattractant protein 2 (MCP2), stem cell factor (SCF), thymic stromal lymphopoietin (TSLP), and thrombopoietin (TPO) compared to patients without PR3-ANCA-associated glomerulonephritis. Patients with ear, nose, and throat involvement expressed higher concentrations of MCP2 and of the (C-X-C motif) ligand-12 (CXCL-12) compared to patients with active AAV and no involvement of these organs. CONCLUSION: We identified distinct cytokine patterns for renal manifestation and for ear, nose and throat involvement of PR3-AAV. Distinct plasma cytokines might be used as non-invasive biomarkers of organ involvement in AAV.

Novel diagnostic and therapeutic techniques reveal changed metabolic profiles in recurrent focal segmental glomerulosclerosis.

Idiopathic forms of Focal Segmental Glomerulosclerosis (FSGS) are caused by circulating permeability factors, which can lead to early recurrence of FSGS and kidney failure after kidney transplantation. In the past three decades, many research endeavors were undertaken to identify these unknown factors. Even though some potential candidates have been recently discussed in the literature, "the" actual factor remains elusive. Therefore, there is an increased demand in FSGS research for the use of novel technologies that allow us to study FSGS from a yet unexplored angle. Here, we report the successful treatment of recurrent FSGS in a patient after living-related kidney transplantation by removal of circulating factors with CytoSorb apheresis. Interestingly, the classical published circulating factors were all in normal range in this patient but early disease recurrence in the transplant kidney and immediate response to CytoSorb apheresis were still suggestive for pathogenic circulating factors. To proof the functional effects of the patient's serum on podocytes and the glomerular filtration barrier we used a podocyte cell culture model and a proteinuria model in zebrafish to detect pathogenic effects on the podocytes actin cytoskeleton inducing a functional phenotype and podocyte effacement. We then performed Raman spectroscopy in the < 50 kDa serum fraction, on cultured podocytes treated with the FSGS serum and in kidney biopsies of the same patient at the time of transplantation and at the time of disease recurrence. The analysis revealed changes in podocyte metabolome induced by the FSGS serum as well as in focal glomerular and parietal epithelial cell regions in the FSGS biopsy. Several altered Raman spectra were identified in the fractionated serum and metabolome analysis by mass spectrometry detected lipid profiles in the FSGS serum, which were supported by disturbances in the Raman spectra. Our novel innovative analysis reveals changed lipid metabolome profiles associated with idiopathic FSGS that might reflect a new subtype of the disease.

Temporal and Spatial Detection of the Onset of Local Necking and Assessment of its Growth Behavior.

This study proposes a method for the temporal and spatial determination of the onset of local necking determined by means of a Nakajima test set-up for a DC04 deep drawing and a DP800 dual-phase steel, as well as an AA6014 aluminum alloy. Furthermore, the focus lies on the observation of the progress of the necking area and its transformation throughout the remainder of the forming process. The strain behavior is learned by a machine learning approach on the basis of the images when the process is close to material failure. These learned failure characteristics are transferred to new forming sequences, so that critical areas indicating material failure can be identified at an early stage, and consequently enable the determination of the beginning of necking and the analysis of the necking area. This improves understanding of the necking behavior and facilitates the determination of the evaluation area for strain paths. The growth behavior and traceability of the necking area is objectified by the proposed weakly supervised machine learning approach, thereby rendering a heuristic-based determination unnecessary. Furthermore, a simultaneous evaluation on image and pixel scale is provided that enables a distinct selection of the failure quantile of the probabilistic forming limit curve.

Determination of Forming Limits in Sheet Metal Forming Using Deep Learning.

The forming limit curve (FLC) is used to model the onset of sheet metal instability during forming processes e.g., in the area of finite element analysis, and is usually determined by evaluation of strain distributions, derived with optical measurement systems during Nakajima tests. Current methods comprise of the standardized DIN EN ISO 12004-2 or time-dependent approaches that heuristically limit the evaluation area to a fraction of the available information and show weaknesses in the context of brittle materials without a pronounced necking phase. To address these limitations, supervised and unsupervised pattern recognition methods were introduced recently. However, these approaches are still dependent on prior knowledge, time, and localization information. This study overcomes these limitations by adopting a Siamese convolutional neural network (CNN), as a feature extractor. Suitable features are automatically learned using the extreme cases of the homogeneous and inhomogeneous forming phase in a supervised setup. Using robust Student's t mixture models, the learned features are clustered into three distributions in an unsupervised manner that cover the complete forming process. Due to the location and time independency of the method, the knowledge learned from formed specimen up until fracture can be transferred on to other forming processes that were prematurely stopped and assessed using metallographic examinations, enabling probabilistic cluster membership assignments for each frame of the forming sequence. The generalization of the method to unseen materials is evaluated in multiple experiments, and additionally tested on an aluminum alloy AA5182, which is characterized by Portevin-LE Chatlier effects.

Analysis of Forming Limits in Sheet Metal Forming with Pattern Recognition Methods. Part 2: Unsupervised Methodology and Application.

The forming limit curve (FLC) is used in finite element analysis (FEA) for the modeling of onset of sheet metal instability during forming. The FLC is usually evaluated by achieving forming measurements with optical measurement system during Nakajima tests. Current evaluation methods such as the standard method according to DIN EN ISO 12004-2 and time-dependent methods limit the evaluation range to a fraction of the available information and show weaknesses in the context of brittle materials that do not have a pronounced constriction phase. In order to meet these challenges, a supervised pattern recognition method was proposed, whose results depend on the quality of the expert annotations. In order to alleviate this dependence on experts, this study proposes an unsupervised classification approach that does not require expert annotations and allows a probabilistic evaluation of the onset of localized necking. For this purpose, the results of the Nakajima tests are examined with an optical measuring system and evaluated using an unsupervised classification method. In order to assess the quality of the results, a comparison is made with the time-dependent method proposed by Volk and Hora, as well as expert annotations, while validated with metallographic investigations. Two evaluation methods are presented, the deterministic FLC, which provides a lower and upper limit for the onset of necking, and a probabilistic FLC, which allows definition of failure quantiles. Both methods provide a necking range that shows good correlation with the expert opinion as well as the results of the time-dependent method and metallographic examinations.

Analysis of Forming Limits in Sheet Metal Forming with Pattern Recognition Methods. Part 1: Characterization of Onset of Necking and Expert Evaluation.

In automotive manufacturing, high strength materials, and aluminum alloys are widely used to address the requirement of ensuring a lightweight car body and correspondingly, reducing pollution. In this context of complexity of materials and structures, an optimized process design with finite element analyses (FEA) is mandatory, as well as a correct definition of the material forming limits. For this purpose, in sheet metal forming, the forming limit curve (FLC) is used. The FLC is defined by the onset of necking. The standard evaluation method according to DIN EN ISO 12004-2 is based on the cross-section method and assumes that the failure occurs due to a clear localized necking. However, this approach has its limitations, specifically in the case of brittle materials that do not exhibit a distinct necking phase. To overcome this challenge, a pattern recognition-based evaluation is proposed. Although pattern recognition and machine learning techniques have been widely employed in the medical field, few studies have investigated them in the context of analyzing metal sheet forming limits. The application of pattern recognition in metal forming is subject to the exact definition of the forming behaviors. Thereby, it is challenging to relate patterns on the strain distribution during Nakajima tests with the onset of necking for the FLC determination. Thus, the first approach was based on the crack evaluation, since this class is well-defined. However, of substantial interest is the evaluation of the general material instabilities that precede failure. Therefore, in the present study, the analysis of the material behavior during stretching is conducted in order to characterize instability classes. The results of Nakajima tests are investigated using an optical measurement system. A conventional pattern recognition approach based on texture features, considering the outcomes of expert interviews for the definition of classes is used for the FLC determination. Moreover, an analysis of the validity of the supervised learning is conducted. The results show a good prediction of the onset of necking, even for high strength materials with a recall of up to 92%. Some deviations are observed in the determination of the diffuse necking. The discrepancies of the different experts' prognoses highlight the user-dependency of the FLC, suggesting further investigations with an data-driven approach, could be beneficial.

Automatic Classification of Cancerous Tissue in Laserendomicroscopy Images of the Oral Cavity using Deep Learning.

Oral Squamous Cell Carcinoma (OSCC) is a common type of cancer of the oral epithelium. Despite their high impact on mortality, sufficient screening methods for early diagnosis of OSCC often lack accuracy and thus OSCCs are mostly diagnosed at a late stage. Early detection and accurate outline estimation of OSCCs would lead to a better curative outcome and a reduction in recurrence rates after surgical treatment. Confocal Laser Endomicroscopy (CLE) records sub-surface micro-anatomical images for in vivo cell structure analysis. Recent CLE studies showed great prospects for a reliable, real-time ultrastructural imaging of OSCC in situ. We present and evaluate a novel automatic approach for OSCC diagnosis using deep learning technologies on CLE images. The method is compared against textural feature-based machine learning approaches that represent the current state of the art. For this work, CLE image sequences (7894 images) from patients diagnosed with OSCC were obtained from 4 specific locations in the oral cavity, including the OSCC lesion. The present approach is found to outperform the state of the art in CLE image recognition with an area under the curve (AUC) of 0.96 and a mean accuracy of 88.3% (sensitivity 86.6%, specificity 90%).

Automatic recognition of Parkinson's disease using surface electromyography during standardized gait tests.

Diagnosis and severity staging of Parkinsons disease (PD) relies mainly on subjective clinical examination. To better monitor disease progression and therapy success in PD patients, new objective and rater independent parameters are required. Surface electromyography (EMG) during dynamic movements is one possible modality. However, EMG signals are often difficult to understand and interpret clinically. In this study pattern recognition was applied to find suitable parameters to differentiate PD patients from healthy controls. EMG signals were recorded from 5 patients with PD and 5 younger healthy controls, while performing a series of standardized gait tests. Wireless surface electrodes were placed bilaterally on tibialis anterior and gastrocnemius medialis and lateralis. Accelerometers were positioned on both heels and used for step segmentation. Statistical and frequency features were extracted and used to train a Support Vector Machine classifier. Sensitivity and specificity were high at 0.90 using leave-one-subject-out cross-validation. Feature selection revealed kurtosis and mean frequency as best features, with a significant difference in kurtosis (p=0.013). Evaluated on a bigger population, this could lead to objective diagnostic and staging tools for PD.