Test Batch Research Articles

Study of the Antioxidant and Immunomodulatory Activity of the Total Aqueous Extract of Clerodendrum splendens (G.Don) Leaves (Lamiaceae)

Clerodendrum splendens (Lamiaceae) is a plant with therapeutic properties, traditionally used in several countries for the treatment of several diseases. The objective of this study was to evaluate the antioxidant and immunomodulatory activity of the total aqueous extract obtained from the leaves of C. splendens in order to contribute to a better knowledge of the therapeutic effects of this plant. It consisted in the realization of the antioxidant activity of ETAC according to the following methods: the DPPH test and the iron reducing power test (FRAP). For the DPPH test, 50 µl of each of the different concentrations of the extract were incubated with 5 ml of a 0.004% methanolic solution of DPPH. After an incubation period of 30 minutes, the absorbances at 517 nm were recorded. The 50% inhibitory concentration of DPPH activity (IC50) expressed in µg/ml of each extract was calculated and compared with that of butylated hydroxytoluene (BHT) taken as the reference antioxidant. The Ferric Reducing/Antioxidant Power (FRAP) assay measures the reducing power of antioxidants in a mixture by their ability to reduce ferric tripyridyl-triazine (Fe3+ -TPTZ) to ferrous (Fe 2+-TPTZ) at acidic pH. 900 µl of the FRAP reagent, previously incubated at 37°C, was mixed with 70 µl of doubly distilled water and 30 µl of the sample (with appropriate dilutions). The increase in absorbance at 593 nm was then monitored for 30 minutes at 37 °C. A range of FeSO4.7H2O, a reducing standard, between 0 and 2000 µM was used to calculate the FRAP values of the extracts and the standard antioxidant (BHT). The results of the study showed that ETAC has a good free radical scavenging activity compared to the activity of BHT, which is a synthetic antioxidant used as a standard. Thirty-two (32) rats weighing between 150 and 200 grams on average were used for the immunomodulatory effect of ETAC. The rats were randomly divided into four (4) batches of eight (8) rats each according to their weight. Each batch consisted of four (4) male and four (4) female animals distributed in different bins according to sex. The control rats were each given 1 ml of distilled water daily by gavage for the 28-day treatment period. The three (3) test batches (batches 1; 2 and 3) received 125 mg/kg bw; 250 mg/kg bw and 500 mg/kg bw of ETAC by gavage daily for 28 days. ETAC significantly increased white blood cell and lymphocyte counts at 125 and 250 mg/kg bw. The increase was greatest at 250 mg/kg bw. Only at 250 mg/kg bw was there a significant increase in neutrophils. Histological sections showed no structural or functional abnormalities in the thymus and spleen.

Transfer learning based fault diagnosis of automobile dry clutch system

Dry friction clutches are prone to fault occurrences due to their continuous exposure to thermal loading and high abrasive rate during power transmission. Fault occurrences in clutches can lead to damage in internal components, downtime and seizure of transmission system. Thus, it is necessary to detect and diagnose such fault occurrences at an early stage. In the present study, an attempt was made to diagnose various clutch faults like release fingers worn out, pressure plate broken, pressure plate worn out, friction material loss, and tangential strip bent using deep learning technique (transfer learning). Vibration signals were acquired from a test rig operated under various load conditions for different clutch conditions (5 faulty and 1 good) that were further processed and stored as vibration plots. The concept of transfer learning involving four pre-trained network models namely, AlexNet, VGG-16, GoogLenet and ResNet-50 were utilized in the present study. Various hyperparameters such as train–test split ratio, learning rate, optimizer and batch size were altered and the best hyperparameters suitable for achieving high classification accuracy for every pre-trained network was determined. Overall, VGG-16 pre-trained network performed exceptionally well among the counterparts with a classification accuracy of 100% in a computational time of 449 s.

Financial derivatives, analyst forecasts, and stock price synchronicity: Evidence from an emerging market

Against the backdrop of the role of financial derivatives in the recent financial crisis, this paper aims to identify the effects of financial derivative usage on the amount of firm-specific information incorporated into stock prices, as measured by stock price synchronicity (SPS), using an emerging market from 2009 to 2019. We find strong evidence consistent with this information asymmetry view. Specifically, our results exhibit that financial derivative usage deters firm-specific information into stock prices, thus increasing SPS. Further analyses indicate that financial derivative usage indirectly affects SPS via analyst forecasts (i.e., analyst forecast accuracy, analyst forecast dispersion, analyst forecast optimism, and analyst following). In addition, our results imply that both the economic complexity and accounting complexity of financial derivatives affect SPS positively. The results are robust to a batch of robustness tests. The findings not only add to the literature on financial derivative usage and SPS but also have policymaking implications for the emerging market.

Quantitative identification of Q-markers of Euphorbiae Humifusae Herba based on AHP-CRITIC comprehensive weighting method

This study investigated the quality markers(Q-markers) of Euphorbiae Humifusae Herba based on the analytic hierarchy process(AHP)-criteria importance through intercriteria correlation(CRITIC) comprehensive weighting method. The Q-markers evaluation system was constructed based on the AHP-CRITIC comprehensive weighting method with quantitative identification of Q-markers of Euphorbiae Humifusae Herba as the target layer. The index weights of the factor layer and the control layer were integrated based on the weights of three indicators(effectiveness, testability, and specificity) in the factor layer calculated by the AHP method and weights of eight indicators(anti-inflammatory inhibitory rate, coagulation shortening rate, anti-cancer inhibition rate, component degree value, component test batch, component average content, content variation coefficient, and number of medicinal materials retrieved according to components) in the control layer calculated by the CRITIC method. The comprehensive score of the chemical components of Euphorbiae Humifusae Herba was weighted and ranked to identify the Q-markers of Euphorbiae Humifusae Herba. In terms of comprehensive scores, top 10 potential Q-markers of Euphorbiae Humifusae Herba were ranked as cynaroside > quercetin > gallic acid > apigenin > luteolin > apigenin-7-O-glucoside > quercetin-7-O-glucoside > ellagic acid > astragalin > ethyl gallate. This study provides a reference for the quality control of Euphorbiae Humifusae Herba and a methodological reference for the quantitative identification of Q-markers of Chinese medicine.

Effect of micro-magnesium oxide admixture on rheological and compressive strength properties of class G well cement

Rheological property tuning is one key aspect of oil-well cement, not only affecting the cement slurry placement but also indirectly impacting other cementing properties such as the bonding strength and compressive strength. However, the research is scant on the rheological properties of blending magnesium oxide with class G oil-well cement, despite some studies having been conducted on its impacts on shrinkage compensation, compressive strength, bonding strength, and permeability properties of oil-well cement. This work studies the effects of micro-sized magnesium oxide on the rheological properties and compressive strength of the formulated API Class G oil-well cement slurries. Two types of micro-sized magnesium oxide (50 μm and 100 μm) were used as the mineral admixture at different dosages (3%, 9%). The base formulated slurry sample was incorporated with the silica flour and several commercialized additives such as retarder, fluid loss control agent, and defoamer. A batch of flow tests have been conducted by the use of a digital viscometer at two different temperatures, 25 and 88°C. The compressive strength has also been measured after curing the samples with the same formulation and conditions for 7 days. The variations of the rheological properties (plastic viscosity, yield stress, and gel strength), shear stress–shear rate correlations, and shear-thinning/thickening behavior are impacted by the temperature, the type, and dosage of magnesium oxide. The plastic viscosity of the tested slurries decreased by 27.0% (type II, 9%, 25°C) and 15.1% (type II, 3%, 88°C), respectively, and the yield stress increased by 258.5% (type II, 3%, 88°C) and 53.9% (type II, 9%, 25°C). The gel strength generally increases as the magnesium oxide dosage increases. However, all tested slurry samples show shear-thinning behavior and non-Newtonian characteristics. Among the tested slurry samples, it is found that better rheological performance is achieved when incorporating magnesium oxide with a smaller particle size. On the other side, the specimens of mixed magnesium oxide with a bigger particle size and medium dosage outperform their rival with a smaller particle size in compressive strength.

Electrochemical investigation of edible oils: Experimentation, electrical signatures, and a supervised learning–case study of adulterated peanut oils

Traditional approaches to characterize edible oils such as chemical, chromatographic and light absorption techniques are laborious, expensive, and bulky to implement. This paper presents the electrochemical impedance spectroscopy of 13 types of edible oils, a rapid robust approach to characterizing the electrical behavior of oils without sample preparation. This is achieved through probing the oils via oscillating electric fields to capture oil-specific electrical behaviors. The principal component analysis discriminates the oil types well and establishes repetitive behavioral trends, perceived as electrical signatures. This data is applied in a case study of adulterated peanut oils to quantify adulteration via supervised machine learning with batch-wise leave-one-out implementation. The mean absolute errors and R2 values measure 2.18–3.27 and 0.975–0.991 respectively across 4 test batches. This work provides an exemplar for the electrochemical study of edible oils, with potential for portable proof-of-value device configurations for rapid in situ analysis of edible oils and adulterated oils.

Decoupling Control Applied to the Smart Grid Power Dispatching Problem

This paper presents controller designs of the decoupling issue dedicated to renewable decentralized generators or new generation grids known as smart grids. The control methods were based on a three-phase voltage source grid representation that integrated wind and photovoltaic generators. The utilized decoupling strategies (Boksenbom and Hood, Zilkind and Luyben, and inverted decoupling) allowed the optimization of energy transfers and costs through the routing grid distribution path. Simulations were conducted in Matlab/Simulink. The trustworthiness of the multivariable control system design in addition to the tuning method was established using three test batches, totally decoupled and stable from input dependencies.

High-Efficiency Water Recovery from Urine by Vacuum Membrane Distillation for Space Applications: Water Quality Improvement and Operation Stability.

Water recovery by membrane distillation (MD) is an attractive alternative to existing urine treatment systems because it could improve the water recovery rate and reliability in space missions. However, there are few studies of urine MD, particularly on the removal of the remaining contaminants from distillate water and the assessment of its long-term performance. In this study, the influences of various operation parameters on distillate water quality and operation stability were investigated in batch mode. The low pH of feedstock reduced the conductivity and total ammonium nitrogen (TAN) in distillate water because the low pH promoted the ionization of ammonia to ammonium ions. However, the low pH also facilitated the formation of free chlorine hydride, which resulted in the minor deterioration of the conductivity in the distillate due to the increasing volatility of chlorine hydride in the feedstock. Thirty batches of vacuum membrane distillation (VMD) experiments demonstrated that the permeate flux and the distillate water quality slightly decreased due to the small range of membrane wetting but still maintained an over 94.2% and 95.8% removal efficiency of the total organic carbon (TOC) and TAN, and the conductivity was <125 μs cm−1 in the distillate water after 30 test batches. VMD is a feasible option for urine treatment in space missions.

Clinical Safety and Tolerability of New Specialized Food Product

ObjectivesThe use of specialized food products may increase efficacy of treatment in patients with non-alcoholic steatohepatitis (NASH). The aim of the study was to assess clinical safety and tolerability of newly developed specialized food product “SPP2” in patients with non-alcoholic steatohepatitis. MethodsNew specialized food product (SPP2) was developed based on literature data on the efficacy of active biological compounds on pathogenesis of NASH. It contains: protein, fat, ω-3 PUFAs, carbohydrates, soluble fibre, Coenzime Q10, L-carnitine, α-lipoic acid, vitamins (A, B1, B2, B6, PP, B12, D3, C, K), phospholipids. The test batch of the product passed laboratory control for biological and toxicological safety. The study was approved by LEC and registered (NCT04308980). Patients with NASH (per EASL), were invited to participate and were randomized to receive iso-caloric diet alone (ICD) or iso-caloric diet and SPP2 (SPP2). Energy value of the diet was amended by calorie content of SPP2. The patients were advised to follow usual physical activity and report any adverse events (AE) occurred during the study. Organoleptic assessment with the use of 5-point Likert scale and repeated blood tests on chemistry and hematology were performed at baseline (BL) and after 14 days (EOT) of treatment. Non-parametric statistics was used for the analysis ResultsOne subject was lost to follow-up in ICD group and one postponed the use of SPP2 due to the not related to safety reasons. The data of 25 subjects were available for the final analysis. The ICD (n = 8) and SPP2 (n = 17) groups did not differ by demographic and baseline characteristics. Adherence to diet was 88% in ICD and 92% in SPP2 groups, p = 0.7. Compliance with SPP was 97%. No serious adverse events occurred during the study. Organoleptic assessment of SPP2 initially: appearance (Mean ± SD) 3.9 ± 0.83 points; color 4.0 ± 0.5; aroma 3.6 ± 0.9; texture 3.7 ± 0.8; taste 3.9 ± 0.7; resulting in a mean score 3.9 ± 0.4 point at BL. No dissatisfaction revealed at EOT: the mean organoleptic score was 4.2 ± 0.7 points (P = 0.07). No laboratory AEs occurred. ConclusionsNew specialized food product “SPP2” in combination with iso-calorie diet is safe and well-tolerated by patients with NASH. Funding SourcesMinistry of Science and Higher Education of Russia, FGMF-2022–0005.

Experimental Study of Internal Erosion in Granular Soil Subject to Cyclic Hydraulic Gradient Reversal

Numerous studies have been conducted to evaluate the influence of internal erosion on the geotechnical properties of soil under unidirectional seepage. Yet cyclic gradients prevail in natural or engineering systems, and the mechanical consequences of soil subjected to cyclic hydraulic gradient reversal have not been addressed in the literature. This paper reported the first batch of internal erosion tests on a modified stress-controlled triaxial apparatus that allows alteration of seepage direction and hydraulic gradient reversals. Three different schemes of hydraulic gradient application, namely monotonic stepwise loading, cyclic reversal after monotonic loading, and direct cyclic loading, were adopted. In each test, the cumulative particle loss, deformation, hydraulic conductivity, and posterosion shearing behavior were evaluated. It was found that the cyclically reversed hydraulic gradient resulted in more severe erosion at the same level of mean hydraulic gradient. Microscopic analysis was also conducted to investigate the mechanism of internal erosion under cyclic hydraulic gradient reversal, in which mitigation of clogging within the soil matrix was observed and found to facilitate the progression of internal erosion.

A Machine Learning Model for Post-Transplant Lymphoproliferative Disorder Using the UNOS Database

<h3>Purpose</h3> Post Transplant Lymphoproliferative Disorder (PTLD) is a complication noted in cardiac allograft recipients. This study was undertaken to generate a risk prediction model for PTLD at 1, 3 and 5 years post heart transplant using machine learning models. The United Network for Organ Sharing (UNOS) database was used for data generation <h3>Methods</h3> The UNOS database was probed from 1987 to 2015<b>.</b> After excluding retransplants and those with incomplete information regarding malignancy, 39696 adult patients were identified out of which 884 patients had PTLD. A Random Under-sampling Boosting ("RUSBoost") technique in combination with Decision Tree models ("ensemble of models") was used to create a PTLD prediction model. Random Under-sampling technique was combined with boosting (AdaBoost M2) to enhance the prediction model. Variables with higher predictive power were used for testing and training. A 5-fold cross validation was performed to train and test them on five different subsets of data. Model performance was evaluated using average Area Under the Curve (AUC) using Receiver Operating Curves (ROC) for the 5 test subsets <h3>Results</h3> The significant continuous variables for predicting PTLD were ischemic time, recipient serum creatinine, and presence of B2, DR1,DR2 antigens in recipient serum. Of the categorical variables investigated gender, diabetes in the recipient, ethnicity, CMV status of the donor and induction with azathioprine, OKT3,thymoglobulin or cyclosporine were significant. The AUCs ranged from 0.76 to 0.79 and 0.66 to 0.7 in the testing or validation batch respectively. <h3>Conclusion</h3> This is the first report of using machine learning algorithms to derive a risk prediction model for PTLD. This model has a robust AUC for predicting PTLD at years 1, 3 and 5 posttransplant. The limitations of this study are its retrospective nature and that it needs external validation.

MICROSTRUCTURE, MECHANICAL PROPERTIES AND GEOMETRY OF A MODEL BATCH OF SEMI-SPHERICAL STEEL LINERS INTENDED FOR USE IN THE EXTRACTIVE INDUSTRY

The article presents the results of investigation of the microstructure and mechanical properties of Fe-based materials intended for the production of semi-spherical liners for experimental shaped charges. The tests were carried out on samples taken in three directions in relation to the rolling direction: 0°, 45° and 90°, of the following materials: cold-rolled sheets made of DC04 steel, hot-rolled sheets made of 04J steel, and cold-rolled strips made of 004G steel developed at Łukasiewicz – IMŻ and annealed at 650°C. Semi-spherical drawpieces with a diameter of 106.4 mm, intended for liners, were made of the sheets with the use of cold stamping. Tests of the microstructure and HV hardness of samples taken from the test batch were carried out on cross-sections along the pressing direction and the model batch – on sections transverse to the stamping direction The geometry of the model batch drawpieces was also measured. The ferrite grain size evaluation was performed according to the ASTM E112 standard using μgrain software with a grain size evaluation module using counting the number of intersections of grain boundaries with a circle.

Variation in the Performance of Different Batches of Two Mycobacterium avium Subspecies paratuberculosis Antibody ELISAs Used for Pooled Milk Samples.

Simple SummaryThis article explores variation in the performance of different batches of tests for the detection of antibodies against the ruminant pathogen Mycobacterium avium subspecies paratuberculosis (MAP) in milk. The results indicate that variation is present and that it has sources mainly in the manufacturing process of the test kits and, to a lesser degree, in the test laboratories.Regionally, the monitoring of paratuberculosis at the herd level is performed by the detection of specific antibodies in pooled milk samples by ELISA. The negative/positive cut-off S/P values applied for pooled milk samples are low and particularly vulnerable to variation in the test performance. In this study, a batch variation in the test performance of two ELISA tests was assessed to identify consequences for sample classification. A total of 72 pooled milk samples (50 from MAP-infected herds, 22 from one MAP-non-infected herd) were analyzed using three different batches, each of two different MAP antibody ELISA tests (A and B). Receiver operating characteristic (ROC) analysis was performed, with the results of each batch, S/P values of the samples and optical density (OD) readings of the negative and positive control samples included in the kits being compared between the batches of one test. ROC analysis revealed a considerable variation in the test performance of the batches of the two individual tests, caused by differences in the S/P values of the samples and resulting in different sensitivities at a specificity of 100%. Major sources of variation originate from the manufacturing processes of test batches. These sources have to be better controlled, and the test performance has to be revisited regularly.

Second life of hops: Analysis of beer hopped with hop pellets previously used to dry-hop a beer

Dry-hopping process is the cold extraction of substances found in hops into beer. This method has gained popularity among beer brewers in recent years because it aids in the introduction of intense flavour and aroma into fermented beverages. However, using a large amount of hop material may result in significant losses of spent hops and beer. In this study, test batches of beer hopped with recycled hops were made with varying dosages of recycled material. Beers brewed with recycled hops had lower IBU (international bittering units) and lower concentrations of hop-derived volatiles (such as aromadendrene, α-terpineol, methyl geranate and α-caryophyllene), but they retained the same level of antioxidant activities (assessed by ABTS∙+, DPPH∙ and FRAP assays), concentration of phenolic compounds and many yeast-derived volatile metabolites as the model beer (hopped with new hops) and acquired higher notes for taste parameter in the sensory analysis.

Jet fuel demand hit by Omicron; recovery delayed until 2023/2024

Jet fuel demand hit by Omicron; recovery delayed until 2023/2024

Against Identification of Contextuality with Violation of the Bell Inequalities: Lessons from Theory of Randomness

Nowadays, contextuality is the hottest topic of quantum foundations and, especially, foundations of quantum information theory. This notion is characterized by the huge diversity of approaches and interpretations. One of the strongest trends in contextual research is to identify contextuality with Bell test contextuality (BTC). In this paper, we criticize the BTC approach. It can be compared with an attempt to identify the complex and theoretically nontrivial notion of randomness with a test for randomness (or a batch of tests, as the NIST test). We advertise Bohr contextuality taking into account all experimental conditions (context). In the simplest case, the measurement context of an observable A is reduced to joint measurement with a compatible observable B. The latter definition was originally considered by Bell in relation to his inequality. We call it joint measurement contextuality (JMC). Although JMC is based on the use of counterfactuals, by considering it within the general Bohr’s framework, it is possible to handle JMC on physical grounds. We suggest (similarly to randomness) to certify JMC in experimental data with Bell tests, but only certify and not reduce.

A rapid and quantitative detection method for plasma soluble growth stimulating gene protein 2 based on time resolved fluorescence immunochromatography.

Background: plasma soluble growth stimulating gene protein 2 (sST2) is a new generation biomarker in heart failure (HF), which is an independent predictor of adverse outcomes of heart failure. Thus, the establishment of a rapid and sensitive method for detecting sST2 is urgently needed. Methods: lanthanide element Eu3+ coated fluorescent nanometer microspheres (Eu3+@FMN) can be used as markers to label monoclonal mouse anti-human sST2 antibody ST-01 (ST-01-Eu3+@FMN). When the immune sandwich complex formed between the monoclonal mouse anti-human sST2 antibody ST-02 and ST-01-Eu3+@FMN on the test band with the appearance of target object sST2, we can detect the fluorescence intensity of Eu3+ on the test band and the quality control band using a dry fluorescence analyzer. We calculated the T/C value (T/C = fluorescence intensity of the test band/fluorescence intensity of the quality control band), fitted to the calibration curve, and measured the concentration value of sST2 in the corresponding sample. Results: the best reaction time was 15 min after condition exploration, and the optimal sample volume was 80 μL. The detection sensitivity of the scheme was 2.14 ng mL-1. The calibration curve of the assay was y = 0.0113x + 0.0033, and the linear range was 5-200 ng mL-1. No cross reaction was found when the samples contained BNP, NT-proBNP, and galectin-3, indicating a good specificity. The precision was good with a relative deviation < 15%. The coefficient of variation of detection results of low-concentration samples and high-concentration samples was 4.20% and 3.30% respectively in the same batch of strip tests, so the intra-assay CV was set as <10%; when different batches of strips were used for testing, the coefficient of variation of detection results of low-concentration samples and high-concentration samples was 10.06% and 8.38% respectively, so the inter-assay CV was set as <15%. Stability test results showed that the relative deviation of test results at each time node was <15%, indicating good stability of the assay strips. The correlation coefficient between the ST-01-Eu3+@FMN based time-resolved fluorescence immunochromatography analysis and sST2 ELISA kit was 0.98. To confirm the usage of our proposed TRF-ICA for clinical samples, it was used to determine the concentration of sST2 in samples obtained from 34 patients with heart insufficiency, acute and chronic heart failure. As a result, we successfully detected a minimal concentration of 5.21 ng mL-1 and a maximum concentration of 184.26 ng mL-1 for sST2. Conclusion: this technique provides a rapid, simple and quantitative detection method for sST2 in clinics. It can help clinicians to predict the incidence of adverse events in patients with HF.

Isometric forging of cups made оf titanium VТ23 alloy

Research and development results of backward isometric extrusion technology applied for сups forgings made of high-strength titanium VТ23 alloy are presented. Stages of technology production namely from the drafting of forging and the forging tooling to the production of maintenance and forgings test batches and characterization of their properties are considered.

Combination and interaction of ageing parameters on single lap shear adhesive joints

For adhesives used in structural adhesive bonds, the service lifetime and environmental ageing are fundamental elements when considering safe design and ensuring the integrity of a structure over its desired lifetime. Ideally, the environmental conditions in which a bonded structure must operate can be quantified, and their effect on structural degradation estimated. It is known that, on the long term, the mechanical performance (stiffness and strength) of an adhesively bonded structure strongly depends on the environmental temperature and the amount of moisture. Both of these environmental factors adversely affect the structural integrity of an adhesive joint, and thus its durability and lifetime. This work analyses aluminium-epoxy single lap shear samples after ageing. An experimental approach is applied to improve the understanding of the mechanical behaviour of adhesive bonds when exposed to multiple environmental factors acting simultaneously. This paper extensively discusses and compares the individual and combined influence of four different environmental factors, static and dynamic loading, UV radiation, temperature and humidity, on the residual strength of a single lap shear joint (ASTM D 1002-05). A design of experiment approach is adopted to define the interaction between the different ageing parameters mentioned above. Moreover, an ANOVA procedure, carried out on various test batch results, analyses the sensitivity of the various environmental factors. The main observed mechanical characteristics are the stiffness and (residual) strength of the single lap samples. After failure, an analysis of the fracture surface yields further information on the sample condition after ageing. The results of the study, for the specific investigated test conditions, indicate that the residual strength of the single lap joint depends strongly on temperature, our observations even show an increase by a factor of 3. On the other hand, other effects may cause a reduction of mechanical properties.

Intelligent classification for three-dimensional metal powder particles

The shape of constituent metal particles has a significant influence on the bulk properties of raw powder and the mechanical properties of manufactured parts. In this paper, we provide a method for automatic shape-based classification of metal powder particles. Firstly, X-ray computed tomography was used to obtain three-dimensional volume data of metal powder particles, and a segmentation operation was performed to separate particles from one another. Classifying by particle shape, particles were manually labelled to create training and testing data, separating the powder particles into one of six user-defined categories: ‘connected’, ‘ellipsoidal’, ‘irregular’, ‘pear’, ‘porous’ and ‘spherical’. The machine learning network PointNet++ was used to classify particles, using 1024 automatically-defined geometric particle features, as well as twelve user-defined features. Our results show that the accuracy of this automatic classification method reaches 93.8%. Finally, an additional test batch of measured powder was used to verify our classification method.