Acceptance Probability Research Articles

Configurational Sampling of All-Atom Solvated Membranes Using Hybrid Nonequilibrium Molecular Dynamics Monte Carlo Simulations.

All-atom simulations are a powerful approach to study the structure and dynamics of biological membranes. However, sampling the atomic configurations of inhomogeneous membranes can be challenging due to the slow lateral diffusion of the various constituents. To address this issue, a hybrid nonequilibrium molecular dynamics Monte Carlo (neMD/MC) simulation method is proposed in which randomly chosen lipid molecules are swapped to generate configurations that are subsequently accepted or rejected according to a Metropolis criterion based on the alchemical work for the attempted swap calculated via a short trajectory. A dual-topology framework constraining the common atoms of the exchanging molecules yields a good acceptance probability using switching trajectories as short as 10 ps. The performance of the hybrid neMD/MC algorithm and its ability to sample the distribution of lipids near a transmembrane helix carrying a net charge are illustrated for a binary mixture of charged and zwitterionic lipids.

Распознавание геомагнитных бурь на основе матричных временных рядов наблюдений мюонного годоскопа УРАГАН с использованием нейронных сетей глубокого обучения

We solve the problem of recognizing geomagnetic storms from matrix time series of observations with the URAGAN muon hodoscope, using deep learning neural networks. A variant of the neural network software module is selected and its parameters are determined. Geomagnetic storms are recognized using binary classification procedures; a decision-making rule is formed. We estimate probabilities of correct and false recognitions. The recognition of geomagnetic storms is experimentally studied; for the assigned Dst threshold Yᴅ₀=–45 nT we obtain acceptable probabilities of correct and false recognitions, which amount to β=0.8212 and α=0.0047. We confirm the effectiveness and prospects of the proposed neural network approach.

Recognition of geomagnetic storms from time series of matrix observations with the muon hodoscope URAGAN using neural networks of deep learning

We solve the problem of recognizing geomagnetic storms from matrix time series of observations with the URAGAN muon hodoscope, using deep learning neural networks. A variant of the neural network software module is selected and its parameters are determined. Geomagnetic storms are recognized using binary classification procedures; a decision-making rule is formed. We estimate probabilities of correct and false recognitions. The recognition of geomagnetic storms is experimentally studied; for the assigned Dst threshold Yᴅ₀=–45 nT we obtain acceptable probabilities of correct and false recognitions, which amount to β=0.8212 and α=0.0047. We confirm the effectiveness and prospects of the proposed neural network approach.

The need for nonuniform risk acceptability across climate change scenarios.

Climate change risk assessment studies focus on identifying and analyzing different risks considering several climate change scenarios and on evaluating the cost-effectiveness of different adaptation measures. However, risk acceptability is often not reflected on in the context of climate change risk studies. Noting that the different climate change scenarios depict drastically contrasting images of the future in terms of population growth, economic development, and changes to life expectancy, this article uses risk acceptance criteria that are based on socioeconomic considerations to highlight the need for nonuniform risk acceptability across climate change scenarios. For this purpose, the optimum implied cost of averting a fatality derived based on the life quality index concept and the value of a quality-adjusted life year derived based on the time principle of acceptable life risk are assessed in three different climate change scenarios for Sweden. Additionally, an illustrative example that assesses the acceptable probability of failure of a steel rod under axial tension in the different climate change scenarios is presented. It is shown that risk acceptance criteria can vary considerably across the different climate change scenarios (e.g., more than 190% variation in the acceptable probability of failure for Sweden in the considered example). This article demonstrates that the ability of societies to afford risk-reducing measures may vary considerably across the different climate change scenarios. Hence, it can be concluded that (1) in the context of climate change risk assessments, risk acceptance criteria need to be developed to account for the different climate change scenarios, and (2) these criteria may vary considerably across the different climate change scenarios. Finally, relevant challenges and research needs are also provided.

Using equivalence tests in higher tier studies of honey bees under the revised EFSA Bee Guidance-How?

The proposed use of equivalence tests instead of difference tests in the revised guidance on the risk assessment of plant protection products for bees is a reasonable approach given an adverse effect was observed in the lower tier studies, using the hypothesis that there is a risk as the null hypothesis places the burden to prove the opposite on the other side. However, some uncertainties regarding the application of equivalence tests in field studies are discussed in the present study. Here, we compare equivalence and difference testing methods using a control dataset of a honey bee field effect study conducted in northern Germany in 2014. Half of the 48 colonies were assigned to a hypothetical test item group, and the colony strength data were analyzed using t-tests, a generalized linear mixed model (GLMM), and the corresponding equivalence tests. The data reflected the natural variability of honey bee colonies, with initially approximately 12 000 adult bees. Although the t-test and GLMM confirmed that 24 + 24 colonies are sufficient to show "no adverse effect," the equivalence tests of the t-test and GLMM were not able to reject the null hypothesis and classified at least some of the assessments as "high risk," indicating a power that was too low. Based on this, different operating options to reduce the variability are discussed. One possible option, which may provide a more realistic application of equivalence to avoid false high risk, is to consider the lower confidence interval of the control as a baseline and use GLMMs. With this option, we demonstrate a relatively acceptable probability to prove that no high risk for initially similar groups can be achieved. Further studies with different numbers of colonies are still needed to develop and validate the suggested approach. Integr Environ Assess Manag 2024;20:1496-1503. © 2024 SETAC.

Reliability-based conformity control method for chloride diffusivity of coastal reinforced concrete buildings

Chloride diffusivity significantly affects the durability of coastal reinforced concrete (RC) buildings, and is one of the major concerns in practice. However, its design and conformity control are commonly based on “deem-to-satisfy” methodology, and consequently it's difficult to evaluate the building's durability performance. In this paper, a reliability-based conformity control method for diffusivity is proposed to eliminate the gap. First, the existing empirical method and semi-empirical method are reviewed, and their drawbacks are addressed according to the reliability assessment. Then, based on the statistical acceptance sampling theory and the engineering practice of Hong Kong-Zhuhai-Macau sea-link project, the reliability-based method for diffusivity is proposed, which can be applied in the construction of coastal RC buildings. In this method, effects of local environmental conditions (e.g., temperature and humidity) on chloride diffusivity are considered, and the relation between diffusion coefficient for design (Da0) and for conformity control (Dnssm) is suggested accordingly. Finally, the conformity control for the concrete's chloride diffusivity of an accessory RC building in the Shantou Harbor is taken as the example, and the schemes by empirical method, semi-empirical method and reliability-based method respectively are established. Their acceptable quality level (AQL) and limiting quality level (LQL), probabilities of acceptance and outcomes in terms of durability reliability are evaluated and compared. It is found that the reliability-based method can offer the schemes that is consistent with the design target, and reject most of the nonconforming lots without harming the producers' benefits.

COVID-19 vaccines and mental health

ABSTRACT In this study, we aimed to examine the impacts of COVID-19 vaccines on mental health and relationships between psychological factors and vaccine acceptance in Turkey during the early pandemic period. The study included 1042 adults (70.9% females) with a mean age of 29.55 (SD = 10.29). The results showed that vaccinated participants had lower levels of COVID-19-induced anxiety and depression, and higher levels of life satisfaction compared to those without the vaccine. Furthermore, we found that COVID-19-induced anxiety and protective behaviours towards COVID-19 significantly increased the probability of vaccine acceptance, while depression significantly decreased it. Accordingly, governments should consider the mental health consequences of vaccination, as well as vaccine acceptance and refusal. More attention should be paid to psychological factors associated with attitudes towards the COVID-19 vaccine.

Patients' perceptions on outcomes after mechanical thrombectomy in acute ischemic stroke.

The modified Rankin Scale (mRS) is a clinician-reported scale that measures the degree of disability in patients who suffered a stroke. Patients' perception of a meaningful recovery from severe stroke, expected value of a stroke intervention, and the effect of disparities are largely unknown. We conducted a survey of patients, their family members, and accompanying visitors to understand their personal preferences and expectations for acute strokes potentially eligible for acute endovascular intervention using a hypothetical scenario of a severe stroke in a standardized questionnaire. Of 164 survey respondents, 65 (39.6%) were the patient involved, 93 (56.7%) were a family member, and six (3.7%) were accompanied visitors (friends, other). Minimally acceptable disability after a stroke intervention was considered as mRS 2 by 42 respondents (25.6%), as mRS 3 by 79 (48.2%), and as mRS 4 by 43 (26.2%) respondents. Race was associated with different views on this question (p < 0.001; Hispanic and Black patients being more likely to accept disability than Caucasian and Asian patients), while sex (p = 0.333) and age (p = 0.560) were not. Sixty-three respondents (38.4%) viewed minimally acceptable probability of improvement with an intervention as over 50%, 57 (34.8%) as 10-50%, and 44 (26.8%) as less than 10%. A wide range of acceptable outcomes were reported regardless of gender or age. However, race was associated with different acceptable outcome. This is an important finding to demonstrate because of the persistent racial and ethnic disparities in the utilization of endovascular therapy for acute stroke in the United States.

Bibliometric Analysis of Sociological Research on Artificial Intelligence

This research aims to identify potential areas for future sociological research related to artificial intelligence (AI). The study used bibliometric analysis methods and the VosViewer pro- gram to process data. The data analyzed included 31 articles related to "sociology" and "artificial intelligence," and 1,277 articles pertinent to "social" and "artificial intelligence," all published on ScienceDirect between 2003 and 2023. Network visualization, overlay, and density analysis were used to process the data. This revealed that current sociological re- search on AI only covers five topics - artificial intelligence, sociology, technology, affects, and artificial intelligence. However, social research on AI has identified 100 topics across five datasets, with almost all research being conducted within the past decade. It is noteworthy that "sociology" is not among these 100 topics. However, these 100 topics have the potential to become sociological research subjects by applying sociological principles. The research findings suggest that sociologists can publish their scientific documents in 3,800 journals and books published by Elsevier, indicating a high probability of acceptance. Furthermore, the topics can be framed from a sociological perspective, thus providing greater insight on the subjects and potentially opening up the door to more publications by the sociologists.

Research on loan approval and credit risk based on the comparison of Machine learning models

Nowadays, home loan is a frequently accessed component of people’s financing activities. Homeowners wants to increase the probability of loan acceptance, however banks seek to borrow money to low risk customers. This paper compared and examined the machine learning models to select when loan applicants evaluating their probability of success. This paper introduced the recommended models for the problem, explanations on how to use the selected model. 6 candidate models, including Logistic regression, Decision tree, Random Forest, support vector machine (SVM), Ada Boost and Neural Network are selected. The model selection process would focus on the model’s accuracy on test data as well as the interpretability of these models. The models’ result was interpreted to derive optimal strategies to be undertaken by both debtors and creditors. Throughout comparison between these models, logistic regression was the best in terms of interpretability and accuracy. Nonetheless, other models could bolster the decision-making process by examining their confusion matrices and the fitted importance of predictors in each model. This paper revealed practical implications of machine learning theories on home loan approval and credit risk and aimed to help decision making for both debtors and creditors.

Защита человека от ударных волн средствами инженерно-технической укреплённости

Purpose. Protective screens (barriers) are used in areas with large crowds of people to reduce the likelihood of human injury from explosive devices used in terrorist acts. To assess the effectiveness of protective screens, it is necessary to determine the following initial data in advance: the weight of the explosive device in TNT equivalent; the area being protected, and the acceptable probability of human injury within it. The article presents a general methodology for determining the effectiveness of protective barriers against terrorist-directed explosive devices. An example is given of using the developed methodology to determine the most optimal solution for protective barriers for a specific object. Methods. Various numerical simulation methods are used to solve gas dynamics problems. When modeling explosions accompanied by shock impacts, the Godunov’s scheme – solving the breakdown of an arbitrary discontinuity problem – is used. The problem of the breakdown of an arbitrary discontinuity in numerical schemes is reduced to a system of nonlinear equations related to the pressure and velocity of the medium at the discontinuity point, which determine the mass and momentum fluxes exchanged between adjacent computational cells. Findings. The author's experimental research methodology allows obtaining all the necessary information about the parameters of explosive loads formed during internal accidental explosions. Research application field. The research results can be used in the design of anti-terrorist protective structures. Conclusions. Blast loads are presented for the placement of anti-terrorist protective structures. Reduction in blast loads is shown for three different placement options of protective structures. The conditional probability of human injury by an air shock wave is determined. Based on the found parameters, the effectiveness of anti-terrorist protective structures is assessed for three different placement options.

A Novel Approach Towards Biometric Recognition System Using Voice and Signature

Recently, recognition systems have gained importance due to their vital role in security by identifying users while interacting with electronic devices to ensure reliability. The typical authentication systems based on PINs or passwords have revealed various issues of deceitful access. On the contrary, Biometrics can establish an apparent discrepancy between a genuine individual and a fraudulent imitator. Biometrics may be categorized as a Single Model or Multimodal System. Systems of Multimodal bio-metrics are used in Physical Access, Civil ID, Criminal ID, Network/PC Access, Kiosk/ATM, Retail/POS, Surveillance, e-commerce, and telephony. This paper proposes a novel and robust recognition system that uses voice and signatures as inputs to recognize the users. The proposed method uses (Mel Frequency Cepstral Coefficients) MFCC and (Vector Quantization) VQ as characteristic vectors to perform voice analysis, and Vertical Projection Profile (VPP), Horizontal Projection Profile (HPP), and Discrete Cosine Transform (DCT) features to design an offline signature identification system. In the design, a similarity score is obtained. False acceptance and rejection probabilities are measured based on the highest score for each uni-model system. Finally, both methods are merged to get an equal error rate, which is used to evaluate the effectiveness of the proposed approach. The results indicate that when compared with unimodal biometric systems, the proposed multi-model biometric system produces less EER, making the system more robust and reliable.

Reliability analysis of offshore monopile foundations considering multidirectional loading and soil spatial variability

Monopiles are the primary type of foundations for offshore wind turbines. Current monopile design methods do not adequately account for the effects of multidirectional loading and soil spatial variability, which may underestimate the failure probability of the foundations. This study aims to investigate the reliability of offshore monopile foundations considering the effects of multidirectional loading and soil spatial variability. A non-intrusive random finite element method combined with Monte Carlo simulation is employed. The behaviors of monopiles in different spatially variable marine soils and loading paths are simulated. Through the reliability analysis, safety factors (FSdesign) are recommended for different scenarios. The results demonstrate that the combined effects of multidirectional loading and soil spatial variability significantly affect the monopile foundation performance and that ignoring these factors will underestimate the failure probability. If the effect of multidirectional loading is more prominent or the marine soil exhibits stronger spatial variability, a larger FSdesign is required to assure an acceptable failure probability.

High-altitude demonstration of LWIR polarimetry using uncooled microbolometers

An Infrared Channeled Spectro-Polarimeter (IRCSP) was successfully demonstrated in a near-space environment on a high-altitude balloon mission out of the NASA Columbia Scientific Balloon Facility in August 2021. The compact IRCSP is sensitive to linearly polarized thermal radiation between 7-12μm and targets polarized cloud top scattering. The IRCSP comprises a series of polarization modulation components followed by a diffraction grating and lens that images a polarization modulated spectrum at the focal plane. At altitudes greater than 30 km, the IRCSP reported brightness temperatures (Tb) between 250–285 K with an uncertainty of less than 1.5K. These IRCSP Tb values overlapped the collated GOES-16 cloud top temperatures within the pointing uncertainty of the instrument. IRCSP’s inaugural high-altitude deployment demonstrated the operation of uncooled microbolometers in the low-pressure environment and resulted in the first known observations of downward-viewing thermal polarized radiation from Earth’s atmosphere. Using the Lomb–Scargle periodogram with a maximum acceptable false alarm probability of 1%, the IRCSP detected statistically significant polarized signals with average degrees of linear polarization (DoLP) ranging from 1% to 20%. The top-of-atmosphere simulations that motivated IRCSP’s development predict that LWIR polarization is highest for ice clouds of moderate optical thickness. IRCSP observed cloud-top DoLP trends with brightness temperature that are consistent with these simulations. The highest and lowest Tb values occur for no clouds and optically thick clouds, respectively. The lowest DoLP measurements by IRCSP were at these extreme conditions. The DoLP peaked at intermediate Tb values. This preliminary work motivates IRCSP’s participation in future deployments that provide independent data products of cloud microphysical properties.

Acceptance of COVID-19 vaccination and associated factors in Middle East countries: a multinational study

ABSTRACT Introduction Widespread use of effective and safe vaccines is the most promising method to contain the COVID-19 pandemic. This study aimed to assess the acceptance level of COVID-19 vaccination among the general population in six Middle East countries (Egypt, Jordan, Palestine, Tunisia, Sudan, and Yemen) as well as to assess factors associated with acceptance or refusal of vaccination. Methods A cross-sectional multinational study was conducted during the period from May 20 to August 8 2021. A web-based, self-administered, Google form Arabic questionnaire was used after a pilot study using 30 participants. Results The COVID-19 vaccine acceptance rate was 62.9%. The highest rate was reported among Tunisians (70%), while the lowest rate was detected among Egyptians (54.4%) and Yemenis (49%). Fear of side effects of vaccination was the main barrier to vaccination (53.1%). Meanwhile, nearly three-fifths (57.7%) of the respondents reported that vaccination would reduce the risk of infection. Logistic regression analysis stated that age >40 years, having children, being Health-Care Workers (HCWs), had higher education and higher income levels, and administering influenza vaccination were significant predictors of vaccine acceptance, as well as Jordanians, Palestinians, and Sudanese had higher probabilities of vaccine acceptance than Egyptians and Yemenis. Conclusions The acceptance rate was moderate. Fear of vaccine side effects and lack of receiving appropriate information were reported as barriers to vaccination. It is important to improve vaccine acceptance and reduce the barriers to COVID-19 vaccination.

Arts and creativity interventions for improving health and wellbeing in older adults: a systematic literature review of economic evaluation studies

BackgroundAs the population ages, older people account for a larger proportion of the health and social care budget. A significant body of evidence suggests that arts and creativity interventions can improve the physical, mental and social wellbeing of older adults, however the value and/or cost-effectiveness of such interventions remains unclear.MethodsWe systematically reviewed the economic evidence relating to such interventions, reporting our findings according to PRISMA guidelines. We searched bibliographic databases (MEDLINE, EMBASE, Econlit and Web of Science and NHSEED), trial registries and grey literature. No language or temporal restrictions were applied. Two screening rounds were conducted independently by health economists experienced in systematic literature review. Methodological quality was assessed, and key information extracted and tabulated to provide an overview of the published literature. A narrative synthesis without meta-analysis was conducted.ResultsOnly six studies were identified which provided evidence relating to the value or cost-effectiveness of arts and creativity interventions to improve health and wellbeing in older adults. The evidence which was identified was encouraging, with five out of the six studies reporting an acceptable probability of cost-effectiveness or positive return on investment (ranging from £1.20 to over £8 for every £1 of expenditure). However, considerable heterogeneity was observed with respect to study participants, design, and outcomes assessed. Of particular concern were potential biases inherent in social value analyses.ConclusionsDespite many studies reporting positive health and wellbeing benefits of arts and creativity interventions in this population, we found meagre evidence on their value or cost-effectiveness. Such evidence is costly and time-consuming to generate, but essential if innovative non-pharmacological interventions are to be introduced to minimise the burden of illness in this population and ensure efficient use of public funds. The findings from this review suggests that capturing data on the value and/or cost-effectiveness of such interventions should be prioritised; furthermore, research effort should be directed to developing evaluative methods which move beyond the confines of current health technology assessment frameworks, to capture a broader picture of ‘value’ more applicable to arts and creativity interventions and public health interventions more generally.PROSPERO registrationCRD42021267944 (14/07/2021).

Metropolis-Hastings algorithm in joint-attention naming game: experimental semiotics study.

We explore the emergence of symbols during interactions between individuals through an experimental semiotic study. Previous studies have investigated how humans organize symbol systems through communication using artificially designed subjective experiments. In this study, we focused on a joint-attention-naming game (JA-NG) in which participants independently categorized objects and assigned names while assuming their joint attention. In the Metropolis-Hastings naming game (MHNG) theory, listeners accept provided names according to the acceptance probability computed using the Metropolis-Hastings (MH) algorithm. The MHNG theory suggests that symbols emerge as an approximate decentralized Bayesian inference of signs, which is represented as a shared prior variable if the conditions of the MHNG are satisfied. This study examines whether human participants exhibit behavior consistent with the MHNG theory when playing the JA-NG. By comparing human acceptance decisions of a partner's naming with acceptance probabilities computed in the MHNG, we tested whether human behavior is consistent with the MHNG theory. The main contributions of this study are twofold. First, we reject the null hypothesis that humans make acceptance judgments with a constant probability, regardless of the acceptance probability calculated by the MH algorithm. The results of this study show that the model with acceptance probability computed by the MH algorithm predicts human behavior significantly better than the model with a constant probability of acceptance. Second, the MH-based model predicted human acceptance/rejection behavior more accurately than four other models (i.e., Constant, Numerator, Subtraction, Binary). Among the models compared, the model using the MH algorithm, which is the only model with the mathematical support of decentralized Bayesian inference, predicted human behavior most accurately, suggesting that symbol emergence in the JA-NG can be explained by the MHNG.

Backup Resource Allocation of Virtual Machines With Two-Stage Probabilistic Protection

In a cloud, protection through backup of virtual machines contained in physical machines (PMs) reduces damage to users due to failure of PMs, such as hardware malfunctions. However, from a resource cost perspective, it is necessary to reduce the amount of capacity for backup while limiting the probability of unsuccessful protection. Existing studies suggest the method of sharing backup capacity among primary resources, but the amount of capacity reduction required to protect is limited. This paper proposes a backup resource allocation model with two-stage probabilistic protection to minimize the total required backup capacity for multiple simultaneous failures of PMs. In probabilistic protection, backup resources are allocated so that the probability of backup failure does not exceed a given survivability parameter which represents the acceptable probability of backup failure. Probabilistic protection which achieves efficient sharing of backup capacity enables flexible allocation and reduces the required backup capacity. In order to increase the flexibility of backup capacity allocation, the proposed model extends the probabilistic protection to two stages. By dividing the protection into two stages, the weight of probability between stages can be adjusted, enabling more effective capacity sharing. Since it is uncertain which primary PMs fail, we apply robust optimization to the probabilistic protection. By using a table that takes into account the survivability parameter and the failure probability of PMs, the proposed model is formulated as a mixed integer linear programming problem. We prove the NP-hardness of considered problem. A heuristic is introduced to solve the optimization problem. The proposed model can reduce the total required backup capacity compared to the models with dedicated protection and one-stage probabilistic protection. The model can also provide protection in a range of survivability parameters that the model with one-stage probabilistic protection cannot satisfy.

Determination of the minimum number of specimens required for laboratory testing of the shear strength of rock joints

The shear strength of rock joints is critical in determining rock mass stability. However, the reliability of shear strength determined by testing only a few rock joint specimens is questionable due to the inherent heterogeneity of rock joints. This study aimed to investigate shear strength heterogeneity and determine the required minimum number of specimens (RMN) for obtaining reliable values. Specimens were extracted from various positions within a natural rock joint, and their roughness and shear strength variations were examined. A significant correlation between shear strength heterogeneity and joint roughness was found. Monte Carlo simulation was used to simulate engineering practice sampling procedures. The effectiveness of specimen numbers under different normal stresses was evaluated using relative error and corresponding probability. Results showed that the RMN depended on acceptable relative error, specified probability, and normal stress. For acceptable relative errors of 5% to 10% and corresponding probabilities of 85–95%, the RMN exhibits a decrease with an increase in normal stress, while it demonstrates an increase when considering acceptable relative error and probability requirements. A smaller RMN was needed under high normal stress for the same rock joint, demonstrating a decrease in shear strength heterogeneity with increasing normal stress. However, commonly used specimen numbers like 3 or 5 did not always align with the recommended RMN. The results also emphasize the significance of considering rock material properties when determining the RMN. The proposed method provides a quantitative and reliable tool for determining the RMN, considering the heterogeneity of rock joints.

Automated Cytomorphological Analysis of Bone Marrow Samples: A Proof-of-Principle Study for AI-Based Classification on a Real-Life Data Set of 979 Unselected Cases

BACKGROUND: Cytomorphological analysis of bone marrow (BM) samples is a cornerstone of hematological diagnostics. It requires microscopical cell counting by a human specialist, resulting in a differential cell count and a morphological description/diagnosis. The process is susceptible to interobserver variability and may be limited by the areas examined. AIM: Clinical implementation of AI-based recognition of pathological patterns in BM smears to support cytomorphological diagnostics. METHODS: We evaluated unselected BM samples over a three-week period (n= 979), including cases with various hematological diseases. We analyzed both shipped unstained slides (n=522, 53%) and smears prepared in-house from BM aspirates (n=457, 47%) with variable shipping time (0-10d, median = 1d), reflecting diagnostic reality in an international reference laboratory. After conventional cytomorphological assessment, each sample was further analyzed by immunophenotyping, cytogenetics, FISH or molecular genetics, according to diagnostic guidelines. BM slides were then scanned using a Metasystems Metafer System (Altlussheim, GER), yielding up to 52 fields of view (FOVs) (2048x1496px each). Cells were automatically identified, segmented, and imaged resulting in 2636 cells per case on average (range = 0 - 10,072, SD = 2164), which were classified into 24 distinct cell classes using an AI-classifier based on the ResNext-architecture. After eliminating 107 samples (10.9%) due to insufficient scanning/sample quality or low cell counts (&amp;lt;100), 872 samples were further analyzed. We defined 7 relevant pathological patterns extractable from human differential counts which served as ground truth for comparison with AI-based classifications: elevated blast fraction (three-tiered, ≥5%, ≥10%, ≥20%), eosinophilia (≥ 20%), plasmacytosis (≥10%), lymphocytosis (≥20%) and monocytosis (≥5%). For each case, we calculated the median prediction probability (MPP) for each pathological pattern, based on the sum of single cell predictions for pattern-specific cell type and cell fraction (Figure 1). RESULTS: Out of 872 analyzed cases, 341 cases showed one pathological pattern (39.1%), 85 cases showed two patterns (9.7%), and 8 showed three patterns (0.9%) as ground truth. MPP for the pathological patterns were: 0.60 for blasts ≥20% (indicating acute leukemia), 0.52 for blasts ≥10% (e.g. suggesting MDS-IB2), 0.61 for blasts ≥5% (e.g. suggesting MDS-IB1), 0.90 for Eosinophilia (suggesting ML-Eo, HES), 0.87 for plasmacytosis (suggesting plasma cell myeloma), 0.87 for lymphocytosis (suggesting lymphoma) and 0.79 for monocytosis (suggesting MDS or CMML). In cases presenting a blast fraction ≥20% (ground truth) and an MPP for blasts below the median of 0.61, MPP for lymphocytosis and monocytosis was substantially higher (0.81 and 0.71 respectively). This points to the known challenge of differentiating these 3 cell types in AI-supported as well as in human cell classification. Of note, pathological patterns of high clinical relevance (i.e. elevated blast fractions) had consistently low MPP in true negative cases (0.03 for blasts ≥20%, 0.07 for blasts ≥10%, 0.15 for blasts ≥5%). CONCLUSION: We successfully implemented automated scanning and AI-based recognition of pathological patterns in a high-throughput routine workflow on unselected BM smears with variable pathologies, preanalytical parameters and quality. AI detected 7 relevant pathological patterns extracted from human differentials at acceptable prediction probability, even if present concurrently. Confusion of mononuclear cells was noted and requires further optimization of image acquisition and AI-based quality control (e.g., upstream binary classifiers or pre-selection of FOVs) to enhance robustness. This preliminary study proves the technical feasibility of AI-based recognition of pathological BM patterns, to support human investigators in decision making for further diagnostic processes.