Automatic Data Research Articles

FoleyGAN: Visually Guided Generative Adversarial Network-Based Synchronous Sound Generation in Silent Videos

Deep learning based visual-to-sound generation systems have been developed that identify and create audio features from video signals. However, these techniques often fail to consider the time-synchronicity of the visual and audio features. In this paper we introduce a novel method for guiding a class-conditioned GAN to synthesize representative audio with temporally extracted visual information. We accomplish this visual-to-sound generation task by adapting the synchronicity traits between the audio-visual modalities. Our proposed FoleyGAN model is capable of conditioning action sequences of visual events leading to the generation of visually aligned realistic soundtracks. We expanded our previously proposed Automatic Foley data set. We evaluated FoleyGAN's synthesized sound output through human surveys that show noteworthy (on average 81%) audio-visual synchronicity performance. Our approach outperforms other baseline models and audio-visual data sets in statistical and ablation experiments achieving improved IS, FID and NDB scores. In ablation analysis we showed the significance of our visual and temporal feature extraction method as well as augmented performance of our generation network. Overall, our FoleyGAN model showed sound retrieval accuracy of 76.08% surpassing existing visual-to-audio synthesis deep neural networks.

A method for detecting abnormal behavior of ships based on multi-dimensional density distance and an abnormal isolation mechanism.

Abnormal ship behavior detection is essential for maritime navigation safety. Most existing abnormal ship behavior detection methods only build A ship trajectory position outlier detection model; however, the construction of a ship speed outlier detection model is also significant for maritime navigation safety. In addition, in most existing methods for detecting a ship's abnormal behavior based on abnormal thresholds, one unsuitable threshold leads to the risk of the ship not being minimized as much as possible. In this paper, we proposed an abnormal ship behavior detection method based on distance measurement and an isolation mechanism. First, to address the problem of traditional trajectory compression methods and density clustering methods only using ship position information, the minimum description length principle based on acceleration (AMDL) algorithm and Multi-Dimensional Density Clustering (MDDBSCAN) algorithm is used in this study. These algorithms not only considered the position information of the ship, but also the speed information. Second, regarding the issue of the difficulty in determining the anomaly threshold, one method for determining the anomaly threshold based on the relationship between the velocity weights and noise points of the MDDBSCAN algorithm has been introduced. Finally, due to the randomness issue of the selected segmentation value in iForest, a strategy of selectively constructing isolated trees was proposed, thus further improving the efficiency of abnormal ship behavior detection. The experimental results on the historical automatic identification system data set of Xiamen port prove the practicality and effectiveness of our proposed method. Our experiment results show that the proposed method achieves an improvement of about 10% over the trajectory outlier detection based on the local outlier fraction method, about 14% over the isolation-based online anomalous trajectory method in terms of the accuracy of ship position information anomaly detection, and about 3% over the feature fusion method in terms of the accuracy of ship speed anomaly detection. This method improves algorithm efficiency by about 5% compared to the traditional isolation forest anomaly detection algorithm.

Intraoperative visual evoked potential recording in occipital tumor surgery

To evaluate the effectiveness of intraoperative monitoring of cVEP in patients with occipital lobe tumors. Ten patients with occipital lobe tumors have undergone surgery with neurophysiological monitoring since 2020. Mean age of patients was 57 years. There were 6 women and 4 men. In 7 patients, neoplasms were located in the right hemisphere, in 3 patients - in the left hemisphere. According to preoperative automatic perimetry data, 7 patients had various visual field defects, and other ones had intact visual fields. All patients underwent pre- and postoperative MRI for visualization of optic radiation, its relationship with tumor and control of resection quality. Intraoperative monitoring of cVEPs was performed in all patients. Biopsy verified glioblastoma in 5 cases, metastasis of adenocarcinoma - 2 cases, diffuse glioma - 1 case, ganglioglioma - 1 case, CNS lymphoma - 1 case. Postoperative MRI confirmed total or subtotal resection of tumor in all cases. Enlargement of visual fields occurred in 3 patients after surgery. Two ones had deterioration and/or new homonymous defect. No changes of visual fields was observed in other cases. Analysis of visual field defects after surgery found no correlation with functional state of visual tract according to fractional anisotropy before and after surgery. MR tractography of optic radiation and intraoperative monitoring of cVEP allow choosing the safest approach for resection of occipital tumor and minimizing the risk of damage to visual cortex and optic radiation fibers. In most cases, postoperative visual functions do not worsen after intraoperative mapping of visual cortex and determining the safest trajectory for resection of occipital lobe tumors. Moreover, improvement is observed in some cases.

Deep learning-based bubble detection with automatic training data generation: Application to the PEM water electrolysis

Various fields, such as the paper industry, chemical engineering, and renewable energy, are faced with gas-liquid two-phase flows and are being studied by visualization and observation. Although it is necessary to quantitatively evaluate the characteristics of bubbles, there is a limitation in the amount of labor required for detection and measurement by human observation of images. There are no examples for bubbles in polymer electrolyte membrane water electrolysis (PEMWE), where the bubbles in PEMWE have heterogeneous backgrounds, unpatterned patterns, and unclear bubble contours. Existing methods for detecting these bubbles are not expected to be accurate enough. In this study, a deep learning-based bubble detection method using convolutional neural networks (CNN) was developed for bubbles in PEMWE. Our method has two novel approaches: first, we developed an algorithm that automatically draws a pseudo-bubble image based on an actual bubble image on an arbitrary background as a method for automatically generating training data for the CNNs. Second, we developed a CNN-based bubble detection method that uses the motion of bubbles, specifically, the difference between the bubble image and the image from one frame ago, as input. Finally, we tested the algorithm on bubble images in a PEMWE and achieved an F1 score of 0.83 or better for all current densities of 0.5, 1, 2, and 3 A/cm2, and an F1 score of 0.844 for the entire detection.

Indoor Point Cloud Segmentation Using a Modified Region Growing Algorithm and Accurate Normal Estimation

With the development of 3D sensors, 3D point cloud data can now be obtained conveniently, which has made the development of automatic point cloud data processing technology crucial. Region growing is a commonly used algorithm to segment point cloud, which greatly depends on the accuracy of point normals and requires the tuning of two thresholds; namely, the increment threshold of curvature (σ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sub> ) and normal angles (θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sub> ). In this paper, we improve the region growing algorithm in two ways: Accurate normal estimation and strengthening the region growing criteria. For the first aspect, principal component analysis (PCA) is utilized to estimate the initial normals of the point cloud. Then, the points are divided into regular points (RP) and sharp feature points (SFP), according to their initial normals. A robust estimator-based PCA is then applied to refine the SFP normals. For the latter aspect, non-connective points are detected according to shared neighbor points, and non-coplanar points are determined by comparing the residual with a robust scale of it. In addition, σ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sub> is set as the 95 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> percentile of curvature, allowing for easier parameter adjustment. Finally, the segmentation effect of the proposed method is evaluated through internal and external indices. The results indicate that the proposed method can accurately estimate the point normals within an acceptable time, and can obtain a better segmentation result than the classic PCA-based region growing algorithm and advanced DetMM-based methods.

Smart Monitoring of Wireless Environments with Real-Time Aggregation and Analysis

A distributed network monitoring system has been designed and implemented for smart monitoring of complex indoor wireless environments to support management of wireless communications. In particular, it makes possible real-time monitoring and analysis in smart factories using commodity hardware and open-source software. The system collects data from distributed wireless receivers, aggregates the data in spatial cells and time bins, and computes space-time wireless metric data frames. A prototype monitoring system was implemented with two types of wireless interface sensors, packet-capture and spectral sensors, and automatic online data collection, indexing, aggregation and analysis. The prototype system reports wireless quality metrics with one-meter spatial resolution and periodic update times of 1 to 10 seconds.

The use of stereo cameras and artificial intelligence data to support capture fisheries landing data monitoring

Reliable data and data collection are important key factors in realizing sustainable fisheries management. Landing data collected by the fishing port authority through has issues with coverage and accuracy. The fast development of image processing and artificial intelligence (AI) analysis opens the possibilities of automatic catch monitoring and data collection in the fishing port. This paper aims to develop a catch data collection program in fishing ports based on stereo camera video monitoring which AI processes. The first version of the devices, FishQi-L (Fish Quantitative Intelligence in Landing site), was developed. FishQi-L consisted of a set stereo camera and a pre-programed JETSON Nano image/video processor. The stereo camera allows image size quantification, size estimation, and spatial mapping capabilities. FishQi-L was able to detect the character and dimensions of objects, the trial object in Pekalongan was “Basket” with mean average precision (mAP) of the detection was 84% and 0.4387 losses. To increase the accuracy and reliability of the systems, we need to increase the number of data and data training iterations.

Intelligent Deployment Solution for Tabling Adapting Deep Learning

The deep learning object detection model can well extract the coordinate information of the separating point of the concentrate ore belt boundary line. However, with the drawbacks of acquired zonal image features, it is difficult to obtain the perfect prediction of the concentrate grade and recovery rate. In this study, by adapting deep learning semantic segmentation technique with DeepLab V3+, which can effectively extract multi-dimensional image features. The mapping relationship between image features and attributes of the Tabling equipment is achieved by constructing a multi-output support vector regression model optimized of a sparrow search algorithm (SSA-MSVR). The beneficiation indicators and operating parameters of the Tabling can be continually detected and optimized by an intelligent system that mainly includes image recognition softwares, automatic control units, data processing and communication workstations, and matching intelligent equipment, which accomplished the intelligent deployment solution of the Tabling beneficiation process.

Identifying criteria for greenhouse gas flux estimation with automatic and manual chambers: A case study for N2O

AbstractFluxes of the powerful greenhouse gas nitrous oxide (N2O) are mainly quantified using manually operated or automatic non‐steady‐state chambers. With both systems, fluxes are calculated as the change in N2O concentration over time using linear or non‐linear regression, but the type of regression selected can have a strong influence on the N2O flux magnitude. The HMR package, a regression‐based implementation of the Hutchinson &amp; Mosier Regression, is a widely used software for trace gas flux calculation that provides a recommendation on the most appropriate regression method based on several criteria. New parameters were recently introduced which allow for pre‐filtering based on sample variance (pfvar and pfalpha), and for constraining the curvilinearity of concentration‐time series(SatPct and SatTimeMin). Currently, there are no guidelines on how to choose the best parameters for specific user conditions. Here we address this knowledge gap using datasets from manual and automatic chambers, and sensitivity analyses. We found that the effect of parameter settings on the cumulative fluxes was greater for manual chamber data compared to the automatic chamber data, with ranges of up to 67.3% and 1.5%, respectively. The parameter pfvar was identified as highly sensitive for both manual and automatic chambers; it is, therefore, critical to select a threshold for when to allow for non‐linear flux calculation that accurately represents the given measurement precision. This can be estimated from the variance of N2O measurements at ambient concentration levels. The parameter SatTimeMin was critical for manual chamber data where the curvature is much less constrained due to the lower number of observations. The parameter pfalpha was the least sensitive and can be set at a p‐value equal to 0.05 following common statistical practice. Parameter values depend on the expected flux characteristics with a given chamber design and are currently best selected based on visual inspection of the data. This study identifies where and how specific care should be taken for the selection of parameters in the HMR package, which may contribute to standardizing the methodologies used worldwide for N2O flux calculations, supporting initiatives in which data from many studies need to be combined.Highlights The HMR package, an implementation of the Hutchinson &amp; Mosier Regression, is a widely used software for trace gas flux calculation that provides a recommendation on the most appropriate regression method based on several criteria. New parameters in the updated HMR package constrains the use of nonlinear flux calculation and inclusion of outliers. The new parameters affect the cumulative emissions of N2O obtained with manual and automated chambers differently. We found that the choice of parameter values led to differences in cumulative N2O emissions of up to 67.3% This study identifies where and how specific care should be taken for the selection of parameters.

Robotic Inspection of Underground Utilities for Construction Survey Using a Ground Penetrating Radar

Ground penetrating radar (GPR) is a very useful nondestructive evaluation (NDE) device for locating and mapping underground assets prior to digging and trenching efforts in construction. This paper presents a novel robotic system to automate the GPR data-collection process, localize underground utilities, and interpret and reconstruct the underground objects for better visualization, allowing regular nonprofessional users to understand the survey results. This system is composed of three modules: (1) an omnidirectional robotic data-collection platform that carries a RGB-D camera with inertial measurement unit (IMU) and a GPR antenna to perform automatic GPR data collection and tag each GPR measurement with visual positioning information at every sampling step, (2) a learning-based migration module to interpret the raw GPR B-scan image into a two-dimensional (2D) cross-section model of objects, and (3) a three-dimensional (3D) reconstruction module, i.e., 30.0% GPRNet, to generate underground utility model represented as fine 3D point cloud. Comparative studies were performed on synthetic data and field GPR raw data with various incompleteness and noise. Experimental results demonstrated that our proposed method achieves a higher GPR imaging accuracy in mean intersection over union (IoU) than the conventional back-projection (BP) migration approach, and 6.9%−7.2% less loss in Chamfer distance (CD) than point cloud model reconstruction baseline methods. The GPR-based robotic inspection provides an effective tool for civil engineers to detect and survey underground utilities before construction.

S-21-1: IMPLEMENTING HEARTS HYPERTENSION CONTROL PROJECT IN CHINA

Hypertension is the leading cause of death and disability worldwide, due to its high prevalence and severe complications (stroke, heart attack, etc). Control of hypertension could prevent most unnecessary outcomes. However, detecting hundreds of millions of hypertensive patients globally and controlling BP to target for the rest of their lives could be a huge burden to all health care systems. The HEARTS technical package developed by WHO provides tools for governments and professional organizations to support primary care services (PCS) to apply evidence-based strategies to control and prevent cardiovascular diseases. The HEARTS China hypertension control project was initiated in 2017. It is owned by governments, technically supported by a hypertension specialist committee and managed by the secretariat. The standard protocol with screening for all adult visitors, and stepwise treatment with calcium channel blockers + ARB + thiazide-like diuretic was selected by the advisory committee. The supply of selected medications was supported by governments. The HEARTS monitoring system, with ID card-driven automatic BP measurement and data transfer to the doctors web page (no data entry), records screening, registration and prescriptions during daily practice. The information helps physicians with decision making in diagnosis and treatment and provides regular quality and progress reports to PSCs and health authorities. Up to December 2021, a cumulative total of 894 PSCs participated in the project. They screened more than 620,000 adults, registered over 268,000 and treated 190,000 hypertensive patients. A total of 86,523 patients achieved the goal of BP control at their last visit. The awareness, treatment and control rates improved continuously in the project-involved areas. Consequently, the episodes of hospitalization for cardiovascular disease among all-cause hospitalizations reduced by 13%. It reduced more in the service areas that screened and treated more hypertensive patients, than in those that screened and treated fewer patients (-15% vs -6%). However, the data recorded in the monitoring system indicated serious clinical inertia in the implementation of the HEARTS protocol: screening does not yet involve all visitors; nearly 30% of diagnosed patients were untreated, 70% of treated patients were not regularly followed-up; only 19% of patients with BP&gt; = 160/100 initiated treatment with combined medications; 64% of revisited patients with uncontrolled BP failed to intensify treatment. More training courses and incentive mechanisms may help PSCs to improve the quality of care. Implementing HEARTS has improved quality and coverage for hypertension control. Overcoming clinical inertia should be able to accelerate this progress.

DESIGN AND RESEARCH OF AUTOMATIC STATION BUFR DATA EMERGENCY BACKUP PLATFORM BASED ON HYBRID PROGRAMMING

Relying on the domestic meteorological transmission and communication system and NAS backup mechanism, based on the relevant standards and specifications of the meteorological big data cloud platform, in the case that the “Tianqing” interface cannot obtain data in a timely and effective manner, detailed requirements, detailed research, and clear implementation steps are made to provide users with reliable meteorological data services. Through the design and research of automatic station BUFR data rain express emergency backup system in this paper, the BUFR data for cell turn decoding, the specific factor value processing, can provide users with continuous data service, for business personnel in flood season normal work to provide basic support. The application shows that the sustainability of the data service is enhanced, and it provides a strong guarantee for the acquisition of rainfall elements in flood season.

Method for Inferring the Optimal Number of Clusters with Subsequent Automatic Data Labeling based on Standard Deviation

Machine learning is a suitable pattern recognition technique for detecting correlations between data. In the case of unsupervised learning, the groups formed from these correlations can receive a label, which consists of describing them in terms of their most relevant attributes and their respective ranges of values so that they are understood automatically. In this research work, this process is called labeling. However, a challenge for researchers is establishing the optimal number of clusters that best represent the underlying structure of the data subjected to clustering. This optimal number may vary depending on the data set and the grouping method used and influences the data clustering process and, consequently, the interpretability of the generated groups. Therefore, this research aims to provide an inference approach to the number of clusters to be used in the grouping based on the range of attribute values, followed by automatic data labeling based on the standard deviation to maximize the understanding of the groups obtained. This methodology was applied to four databases. The results show that it contributes to the interpretation of the groups since it generates more accurate labels without any overlap between ranges of values, considering the same attribute in different groups.

Retracted: Automatic Data Collecting and Application of the Touch Probing System on the CNC Machine Tool

Retracted: Automatic Data Collecting and Application of the Touch Probing System on the CNC Machine Tool

Effective approaches to remote asset monitoring with acoustic emission

This paper presents the effectiveness of remote asset monitoring using acoustic emission. We will discuss different aspects of implementing an effective remote asset monitoring operation, such as considerations for the deployment of instrumentation in the field, instrumentation, communications, benefits of automatic data transfer from the monitoring system to a secure server, continuous automatic verification of hardware integrity, and automatic notification of potential damage or conditions affecting the asset integrity occurring. We will provide an in-depth view of the cloud monitoring implementation processes, and the use of machine learning and other data analytics techniques to identify abnormal acoustic emission data patterns. The application of these tools to specific case studies in electric power transformers, power steam generators and civil infrastructure assets will be presented. Finally, we will illustrate how these capabilities help to provide timely, actionable information to the asset owner.

FEATURES OF ENSURING TRANSPARENT AND EFFICIENT BUDGETARY FUNDING OF EDUCATION UNDER THE MARTIAL LAW

The article addresses current issues related to ensuring transparency and efficiency of budget financing in the education sector under martial law. The authors analyze approaches to ensuring monitoring and transparency in the use of budget funds, which is particularly important during times when Ukraine is facing challenging circumstances, and the issue of rational allocation of available financial resources becomes critical. The article provides an overview of the amount of budgetary funding for education in Ukraine, which has been reduced since the beginning of russia's large-scale war against Ukraine. It is substantiated that one of the ways to address above-mentioned issues is to ensure transparency and effectiveness in budget financing of education by implementing additional financial control measures, primarily by the Ministry of Education and Science of Ukraine. Furthermore, the authors propose the development of relevant performance indicators for educational institutions with a corresponding correlation to the allocated budget financing. At present, the Ministry of Education and Science of Ukraine's control over the designated and efficient use of targeted subventions at the local level has a number of shortcomings: insufficient methodological quality of data in relevant reports, lack of internal automatic data integrity controls, violation of reporting deadlines by respondents, outdated approach of sending out Excel templates for forms and collecting them on the e-mails of authorized specialists, etc. Insufficient use of modern tools for reporting, and consequently, ineffective control over the allocation and use of targeted funding from the state budget by budget managers, limits the effectiveness of managerial decision-making, creates a constant systemic risk of deteriorating education development at the local level. It is necessary to introduce financial reporting for each educational institution, which will allow for an appropriate analysis of the effectiveness of budget spending. In turn, the process of collecting, processing and disseminating such financial statements of educational institutions requires appropriate automation. Automation of the process will enable the quick and efficient processing of reports and the identification of risk areas. In particular, it is advisable to create an information and analytical system for collecting and analyzing financial reports of educational institutions to provide conditions for making management decisions based on reliable and evidence-based data.

Methodology for the characterization and understanding of longitudinal wrinkling during calendering of lithium-ion and sodium-ion battery electrodes

The manufacturing of lithium-ion battery (LIB) cells is following a complex process chain in which the individual process steps influence the subsequent ones. Meanwhile, increasing requirements especially concerning the battery performance, sustainability and costs are forcing the development of innovative battery materials, production technologies and battery designs. The calendering process directly affects the volumetric energy density of an electrode and therefore of a battery cell. Calendering is still challenging as it causes high stresses in the electrode that lead to defects and thus increased rejection rates. The interaction between electrode material and process as well as the formation of defects is still not fully understood, especially when new material systems are used. In this context, the sodium-ion battery (SIB) is one post-lithium battery system that is a promising option to overcome the limitations of conventional LIBs. Therefore, this paper presents a first material and machine independent methodology to describe and understand the defect type longitudinal wrinkle, which mostly appears at the uncoated current collector edge of an electrode and in running direction. The aim is to systematically characterize the longitudinal wrinkles according to their geometry. The automatic data acquisition is carried out with a laser triangulation system and a 3D scanning system. The geometry values are calculated from the raw data and correlated to selected process parameters. The methodology is applicable regardless of the material as shown by exemplary results of NMC811 cathodes for LIB and hard carbon anodes for SIB. By using two different pilot calenders it is shown, that the data acquisition can be carried out independently of the machine. The presented methodology contributes to finding solutions for the avoidance of longitudinal wrinkling in any battery electrode and therefore to reducing the rejection rate.

Tight Sandstone Image Augmentation for Image Identification Using Deep Learning

Intelligent identification of sandstone slice images using deep learning technology is the development trend of mineral identification, and accurate mineral particle segmentation is the most critical step for intelligent identification. A typical identification model requires many training samples to learn as many distinguishable features as possible. However, limited by the difficulty of data acquisition, the high cost of labeling, and privacy protection, this has led to a sparse sample number and cannot meet the training requirements of deep learning image identification models. In order to increase the number of samples and improve the training effect of deep learning models, this paper proposes a tight sandstone image data augmentation method by combining the advantages of the data deformation method and the data oversampling method in the Putaohua reservoir in the Sanzhao Sag of the Songliao Basin as the target area. First, the Style Generative Adversarial Network (StyleGAN) is improved to generate high-resolution tight sandstone images to improve data diversity. Second, we improve the Automatic Data Augmentation (AutoAugment) algorithm to search for the optimal augmentation strategy to expand the data scale. Finally, we design comparison experiments to demonstrate that this method has obvious advantages in generating image quality and improving the identification effect of deep learning models in real application scenarios.

Machine Learning-Based Integrated Wireless Sensing and Positioning for Cellular Network

Integrated sensing and communication (ISAC) is one of the most promising technical directions in B5G/6G era. Nevertheless, it is challenging for network operators to provide proper human-centric services (HCSs), appropriately with the explosive growth of mobile terminals in constantly changing scenarios. Scenario sensing is the foundation to provide users with intelligent context-aware services, while the combination of machine learning (ML) and automatic data collection enables network operators to become more flexible in decision-making and planning. In this article, an ML-based integrated indoor/outdoor (IO) sensing and positioning framework is proposed to improve the potential of ML in the context of cellular network service management. First, large amounts of measurement reports (MRs) are collected over Layer 3 at user equipment (UE) and evolved NodeBs (eNBs) in urban areas, including both indoor and outdoor samples to simulate the minimization of drive test (MDT). Then, a random-forest-based IO classifier is implemented to sense the mobile scenarios and filter the positioning fingerprints. Subsequently, the weighted <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> -nearest neighbor (WKNN)-based Enhanced Cell ID (ECID) merging with the MR method is used for cellular positioning. The performance evaluation has shown that the positioning error of the MRs after denoising is effectively reduced compared with the conventional fingerprint-based positioning. Specifically, under the 67% standard defined by Federal Communications Commission (FCC), the resulting positioning error is about 4% lower than the case without preprocessing. In addition, comparative experiments are conducted to discuss the impact of different <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> values at different data sizes of fingerprints. Moreover, the further data analysis shows that MRs collected in the mild indoor (MI) scenarios have a positive effect on the overall positioning error, which indicates that refined contextualizations contribute to the cellular positioning for HCS upgrade.

The impact of biostatistics on hazard characterization using in vitro developmental neurotoxicity assays.

In chemical safety assessment, benchmark concentrations (BMC) and their associated uncertainty are needed for the toxicological evaluation of in vitro data sets. A BMC estimation is derived from concentration-response modelling and results from various statistical decisions, which depend on factors such as experimental design and assay endpoint features. In current data practice, the experimenter is often responsible for the data analysis and therefore relies on statistical software, often without being aware of the software default settings and how they can impact the outputs of data analysis. To provide more insight into how statistical decision-making can influence the outcomes of data analysis and interpretation, we have developed an automated platform that includes statistical methods for BMC estimation, a novel endpoint-specific hazard classification system, and routines that flag data sets that are outside the applicability domain for an automatic data evaluation. We used case studies on a large dataset produced by a developmental neurotoxicity (DNT) in vitro battery (DNT IVB). Here we focused on the BMC and its confidence interval (CI) estimation as well as on final hazard classification. We identified five crucial statistical decisions the experimenter must make during data analysis: choice of replicate averaging, response data normalization, regression modelling, BMC and CI estimation, and choice of benchmark response levels. The insights gained are intended to raise more awareness among experimenters on the importance of statistical decisions and methods but also to demonstrate how important fit-for-purpose, internationally harmonized and accepted data evaluation and analysis procedures are for objective hazard classification.