Search

The kernel method is a tool that converts data to a kernel space where operation can be performed. When converted to a high-dimensional feature space by using kernel functions, the data samples are more likely to be linearly separable. Traditional machine learning methods can be extended to the kernel space, such as the radial basis function (RBF) network. As a kernel-based method, support vector machine (SVM) is one of the most popular nonparametric classification methods, and is optimal in terms of computational learning theory. Based on statistical learning theory and the maximum margin principle, SVM attempts to determine an optimal hyperplane by addressing a quadratic programming (QP) problem. Using Vapnik–Chervonenkis dimension theory, SVM maximizes generalization performance by finding the widest classification margin within the feature space. In this paper, kernel machines and SVMs are systematically introduced. We first describe how to turn classical methods into kernel machines, and then give a literature review of existing kernel machines. We then introduce the SVM model, its principles, and various SVM training methods for classification, clustering, and regression. Related topics, including optimizing model architecture, are also discussed. We conclude by outlining future directions for kernel machines and SVMs. This article functions both as a state-of-the-art survey and a tutorial.

A new lapping process is proposed for high efficiency and low surface roughness lapping of hard-brittle materials such as sapphire, which sinters single crystal diamond abrasive (average grain size 3 μm) with vitrified bond into agglomerated abrasives (average grain size 30 μm). The lapping performance of agglomerate abrasives was studied through machining sapphire substrate with comparison to those using 3 μm or 30 μm single crystal diamond abrasive. The results show that the agglomerated has higher material removal rate (MRR), which is 1.127 μm/min after 15 min machining under the same conditions. It is also found that the agglomerated abrasive performs stable during the machining, the MRR of which is 0.483 μm/min after 120 min lapping, decreasing 57.14% from that after 15 min lapping, while the MRR of 3 μm single crystal diamond decreases for 78.02%. The roughness of surfaces lapped by agglomerated abrasive or 3 μm single crystal diamond is similar, which is Ra 9.45 nm or Ra 8.75 nm, respectively. Both are far lower than the Ra 246 nm of 30 μm single crystal diamond abrasive. The mechanism that agglomerated abrasive can achieve low lapping surface roughness and high MRR can be summarized as follows: it has the characteristics of multiple cutting edges which effectively improve the contacts between abrasive and sapphire surface and enhance the MRR; at the same time, it has the ability of self-sharpening, which ensures that the blunt abrasive falls off during the lapping and that new abrasives are put into process, thus ensuring the stability of the process.

Mechanical manufacturing process of electric motor induces a residual stress, which increases the iron loss, and affects the efficiency and performance of machine. In this paper, a shear test was carried out on a non-oriented electrical steel sheet with a thickness of 0.50 mm and a relative clearance range of 2–12% of the thickness, and the shear force was detected during shearing process. The sheared surface morphology, microhardness distribution and magnetic properties after the shearing processing were also tested. The obtained experimental results indicate that: as the shear clearance increases, the shear force increases, the plastic deformation zone enlarges, and the flow of microstructure becomes intense. In addition, as the shear clearance increases, the depth and work-hardening degree of the work-hardened layer increase, resulting in degraded magnetic properties of the electrical steel. Moreover, when the lateral clearance is 5%, a complete and smooth shear section can be obtained.

Renal clear cell carcinoma (ccRCC) is a malignancy with a dismal prognosis, caused by the buildup of fat and glycogen. Sirtuin 1 (Sirt1) is a deacetylase that regulates lipid metabolism. In this study, we collected tumor and paracancer tissues from 386 ccRCC patients and followed their prognosis over an extended time period. The expression of Sirt1 in these tissues was assessed using immunohistochemistry, and LinkedOmics database analysis identified differentially expressed genes associated with Sirt1. The survival curve was generated using the Kaplan-Meier method, and immune infiltration was analyzed using the Tumor Immune Estimation Resource (TIMER) web tool. Our findings revealed that Sirt1 was expressed in tumor tissues, but not in normal tissues, and its high expression was associated with a worse prognosis. Furthermore, we observed a positive correlation between high Sirt1 expression and perirenal fat invasion and necrosis, leading to poorer survival outcomes. We established a nomogram to predict prognosis, and a correlation was observed with immune infiltration. In conclusion, our results suggest that high Sirt1 expression is associated with lipid metabolism disorder and immune infiltration, ultimately contributing to a dismal prognosis in ccRCC.

Magnetorheological elastomer (MRE) is a magnetically controlled intelligent material, whose mechanical properties (such as stiffness, elastic modulus, intrinsic frequency, damping capacity, etc.) can be continuously and reversibly controlled by adjusting the strength of the applied magnetic field. It is now widely studied and applied in the fields of vibration control, mechanical engineering, civil engineering, etc. When MRE is made into grinding or polishing tools, the mechanical removal of the process can be controlled by changing the stiffness and other properties of MRE in a magnetic field, which is expected to be widely used in the field of precision machining. In this paper, the materials, the methods and the processes to prepare MRE are introduced, and the influencing rules of external field (including magnetic and temperature fields) on the performance of MRE are analyzed. The intrinsic structure model is described based on magnetic dipole theory and macromechanics, which provides guidance for MRE preparation and practical engineering applications. Besides, the application status and future development direction of MRE in the field of precision machining are reviewed. The variation of magnetically controlled properties using MRE can be better applied to precision machining and has a good development prospect.

Abstract Background Bone is one of the most frequent sites for breast cancer metastasis. Breast cancer bone metastasis (BCBM) leads to skeletal morbidities including pain, fractures, and spinal compression, all of which severely impact quality of life. Immunotherapy is a promising therapy for patients with advanced cancer, but whether it may provide benefit to metastatic bone cancer is currently unknown. Thus, a better understanding of the immune landscape of bone-disseminated breast cancers may reveal new therapeutic strategies. In this study, we use histopathological analysis to investigate changes within the immune microenvironment of primary breast cancer and paired BCBM. Methods Sixty-three patients with BCBM, including 31 with paired primary and bone metastatic lesions, were included in our study. The percentage of stroma and stromal tumor-infiltrating lymphocytes (TILs) was evaluated by histopathological analysis. The quantification of stromal TILs (CD4 + and CD8 +), macrophages (CD68 + and HLA-DR +), programmed cell death protein 1 (PD-1), and programmed cell death protein ligand 1 (PD-L1) was evaluated through immunohistochemical (IHC) staining. Statistical analysis was performed with paired t test, Wilcoxon test, spearman correlation test, and univariate and multivariate cox regression. Results Median survival after BCBM pathological diagnosis was 20.5 months (range: 3–95 months). Of the immune parameters measured, none correlated with survival after bone metastasis was diagnosed. Compared to the primary site, bone metastases exhibited more tumor stroma (mean: 58.5% vs 28.87%, p

Abstract Background Metastatic breast carcinoma is commonly considered during differential diagnosis when metastatic disease is detected in females. In addition to the tumor morphology and documented clinical history, sensitive and specific immunohistochemical (IHC) markers such as GCDFP-15, mammaglobin, and GATA3 are helpful for determining breast origin. However, these markers are reported to show lower sensitivity in certain subtypes, such as triple-negative breast cancer (TNBC). Materials and methods Using bioinformatics analyses, we identified a potential diagnostic panel to determine breast origin: matrix Gla protein (MGP), transcriptional repressor GATA binding 1 (TRPS1), and GATA-binding protein 3 (GATA3). We compared MGP, TRPS1, and GATA3 expression in different subtypes of breast carcinoma of (n = 1201) using IHC. As a newly identified marker, MGP expression was also evaluated in solid tumors (n = 2384) and normal tissues (n = 1351) from different organs. Results MGP and TRPS1 had comparable positive expression in HER2-positive (91.2% vs. 92.0%, p = 0.79) and TNBC subtypes (87.3% vs. 91.2%, p = 0.18). GATA3 expression was lower than MGP (p

Aiming at the polishing liquid used for polishing GaN wafers by Fenton reaction CMP, a parameter optimization experiment was carried out with the surface quality as the evaluation index, and the optimal ratio of the polishing liquid components was found out. The results show that when the mass fraction of H2O2 is 7.5%, the surface of GaN wafer processing is the best, and the surface roughness reaches 3.2 nm; the catalyst can effectively adjust the rate of the Fenton reaction. Compared with the liquid catalyst FeSO4 solution and the solid catalyst Fe3O4 powder, the solid catalyst Fe3O4 powder can continuously ionize Fe2+ in the solution, so that the Fenton reaction can continue to work throughout the process. When the particle size of Fe3O4 powder is 20 nm, the polishing surface is the best, and the surface roughness reaches 3.0 nm; compared with alumina, cerium oxide, and silica sol abrasives, the best surface polishing effect can be achieved while using the silica sol abrasives, and the surface roughness reaches 3.3 nm; when the mass fraction of silica sol abrasive is 20.0% and the abrasive particle size is 60 nm for polishing, the surface roughness reaches 1.5 nm. After optimizing the composition of the polishing liquid, the GaN wafer was polished with the optimal composition parameters of the polishing liquid, and a smooth surface with a surface roughness of 0.9 nm could be obtained.

Cholangiocarcinoma (CCA) has emerged as an intractable cancer with scanty therapeutic regimens. The aberrant activation of Yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ) are reported to be common in CCA patients. However, the underpinning mechanism remains poorly understood. Deubiquitinase (DUB) is regarded as a main orchestrator in maintaining protein homeostasis. Here, we identified Josephin domain-containing protein 2 (JOSD2) as an essential DUB of YAP/TAZ that sustained the protein level through cleavage of polyubiquitin chains in a deubiquitinase activity-dependent manner. The depletion of JOSD2 promoted YAP/TAZ proteasomal degradation and significantly impeded CCA proliferation in vitro and in vivo. Further analysis has highlighted the positive correlation between JOSD2 and YAP abundance in CCA patient samples. Collectively, this study uncovers the regulatory effects of JOSD2 on YAP/TAZ protein stabilities and profiles its contribution in CCA malignant progression, which may provide a potential intervention target for YAP/TAZ-related CCA patients.

PurposeIn order to evaluate the neuroprotective effect of low-intensity pulsed ultrasound (LIPUS) for acute traumatic brain injury (TBI), we studied the potential of apparent diffusion coefficient (ADC) values and ADC-derived first-order features regarding this problem.MethodsForty-five male Sprague Dawley rats (sham group: 15, TBI group: 15, LIPUS treated: 15) were enrolled and underwent magnetic resonance imaging. Scanning layers were acquired using a multi-shot readout segmentation of long variable echo trains (RESOLVE) to decrease distortion. The ultrasound transducer was applied to the designated region in the injured cortical areas using a conical collimator and was filled with an ultrasound coupling gel. Regions of interest were manually delineated in the center of the damaged cortex on the diffusion weighted images (b = 800 s/mm2) layer by layer for the TBI and LIPUS treated groups using the open-source software ITK-SNAP. Before analysis and modeling, the features were normalized using a z-score method, and a logistic regression model with a backward filtering method was employed to perform the modeling. The entire process was completed using the R language.ResultsDuring the observation time, the ADC values ipsilateral to the trauma in the TBI and LIPUS groups increased rapidly up to 24 h. After statistical analysis, the 10th percentile, 90th percentile, mean, skewness, and uniformity demonstrated a significant difference among three groups. The receiver operating characteristic curve (ROC) analysis shows that the combined LR model exhibited the highest area under the curve value (AUC: 0.96).ConclusionThe combined LR model of first-order features based on the ADC map can acquire a higher diagnostic performance than each feature only in evaluating the neuroprotective effect of LIPUS for TBI. Models based on first-order features may have potential value in predicting the therapeutic effect of LIPUS in clinical practice in the future.

Abstract Background In patients with squamous cell carcinoma of the cervix (SCCC), a squamous cell carcinoma (SqCC) in the lung represents either a second primary tumor or metastasis. This distinction between second primary tumors and lung metastases in patients with SCCC significantly influences patient prognosis and therapy. Here, we aimed to differentiate second primary tumors from lung metastases in patients with SCCC by exploring the HPV status in SqCCs involving the lung within a large cohort. Methods P16 expression was assessed using immunohistochemistry on tissue microarrays including a total of 415 primary lung SqCCs and 21 lung SqCCs with prior SCCC. Following this, we performed HPV DNA typing and the sensitive RNAscope in situ method to screen all the cases for HPV E6/E7 expression, which is a more reliable indicator of transcriptively active HPV in tumor cells. Results The p16 positive expression rate was 13.7% (57/415) in primary lung SqCCs, but HPV DNA was not detected in any of the 57 primary lung SqCC cases that positively expressed p16. In contrast, HPV DNA was detected in all cases (21/21) with prior SCCC. Consistently, all 21 lung SqCCs with prior SCCC (21/21) showed extensive HPV16 E6/E7 expression. In striking contrast, none of the primary lung SqCCs (0/415) had a detectable RNAscope signal. Conclusions HPV does not seem to play a role in the development of primary lung SqCCs. HPV detection may be helpful in distinguishing second primary tumors from lung metastases in patients with SCCC.

Muscle-invasive bladder cancer (MIBC) is the most common urinary system carcinoma associated with poor outcomes. It is necessary to develop a robust classification system for prognostic prediction of MIBC. Recently, increasing omics data at different levels of MIBC were produced, but few integration methods were used to classify MIBC that reflects the patient’s prognosis. In this study, we constructed an autoencoder based deep learning framework to integrate multi-omics data of MIBC and clustered samples into two different subgroups with significant overall survival difference (P = 8.11 × 10-5). As an independent prognostic factor relative to clinical information, these two subtypes have some significant genomic differences. Remarkably, the subtype of poor prognosis had significant higher frequency of chromosome 3p deletion. Immune decomposition analysis results showed that these two MIBC subtypes had different immune components including macrophages M1, resting NK cells, regulatory T cells, plasma cells, and naïve B cells. Hallmark gene set enrichment analysis was performed to investigate the functional character difference between these two MIBC subtypes, which revealed that activated IL-6/JAK/STAT3 signaling, interferon-alpha response, reactive oxygen species pathway, and unfolded protein response were significantly enriched in upregulated genes of high-risk subtype. We constructed MIBC subtyping models based on multi-omics data and single omics data, respectively, and internal and external validation datasets showed the robustness of the prediction model as well as its ability of prognosis (P < 0.05 in all datasets). Finally, through bioinformatics analysis and immunohistochemistry experiments, we found that KRT7 can be used as a biomarker reflecting MIBC risk.

Chemoresistance is one of the leading causes for the failure of tumor treatment. Hence, it is necessary to study further and understand the potential mechanisms of tumor resistance to design and develop novel anti-tumor drugs. Post-translational modifications are critical for proteins’ function under physiological and pathological conditions, among which ubiquitination is the most common one. The protein degradation process mediated by the ubiquitin-proteasome system is the most well-known function of ubiquitination modification. However, ubiquitination also participates in the regulation of many other biological processes, such as protein trafficking and protein-protein interaction. A group of proteins named deubiquitinases can hydrolyze the isopeptide bond and disassemble the ubiquitin-protein conjugates, thus preventing substrate proteins form degradation or other outcomes. Ubiquitin-specific protease 7 (USP7) is one of the most extensively studied deubiquitinases. USP7 exhibits a high expression signature in various malignant tumors, and increased USP7 expression often indicates the poor tumor prognosis, suggesting that USP7 is a marker of tumor prognosis and a potential drug target for anti-tumor therapy. In this review, we first discussed the structure and function of USP7. Further, we summarized the underlying mechanisms by which tumor cells develop resistance to anti-tumor therapies, provided theoretical support for targeting USP7 to overcome drug resistance, and some inspiration for the design and development of USP7 inhibitors.

ObjectiveCavin3 is a putative tumor suppressor protein. However, its molecular action on tumor regulation is largely unknown. The aim of the current study is to explore the implication of cavin3 alteration, its clinical significance, and any potential molecular mechanisms in the regulation of breast cancer (BC).MethodsTCGA (The Cancer Genome Atlas) and GTEx (Genotype-Tissue Expression) data bases, and 17 freshly paired BC and adjacent normal tissues were analyzed for mRNA levels of Cavin3. Furthermore, cavin3 protein expression from 407 primary BC samples were assessed by immunohistochemistry (IHC) and measured by H-score. The clinical significance of cavin3 expression was explored by Kaplan-Meier analysis and the Cox regression method. In vitro biological assays were performed to elucidate the function and underlying mechanisms of cavin 3 in BC cell lines.ResultsCavin3 mRNA was dramatically down-regulated in BC compared with the negative control. The median H-score of cavin3 protein by IHC was 50 (range 0-270). There were 232 (57%) and 175 (43%) cases scored as low (H-score≤50) and high (H-score >50) levels of cavin3, respectively. Low cavin3 was correlated with a higher T and N stage, and worse distant metastasis-free survival (DMFS) and overall survival (OS). Multivariate survival analysis revealed low cavin3 was an independent fact for worse DMFS. In BC cells, an overexpression of cavin3 could inhibit cell migration and invasion, and significantly decreased the level of p-Akt. Knockout of cavin3, meanwhile, promoted cell invasion ability and increased the level of p-AKT.ConclusionCavin3 expression is significantly lower in BC and is correlated with distant metastasis and worse survival. Cavin3 functions as a metastasis suppressor via inhibiting the AKT pathway, suggesting cavin3 as a potential prognostic biomarker and a target for BC treatment.

Aiming at the chemical magnetorheological finishing (CMRF) of single crystal silicon carbide (SiC) based on the Fenton reaction, concentration of hydroxyl radical produced in Fenton reaction and its influence on chemical reaction rate of SiC were detected by visible spectrophotometry, and catalytic ability of different particles was detected by the method of immersion corrosion test. The experimental results showed that surface roughness is better when the carbonyl iron powder (CIP) is used as catalyst. The lower value of pH, the higher content of hydroxyl radicals. The hydroxyl radical can react with SiC surface to form a softer SiO2 oxide layer. The acid environment can promote Fenton reaction and the polishing effect is better, while the oxidation layer of SiC surface is converted to silicate in the alkaline environmental. The SiC after CMRF can obtain the smooth surface of the surface roughness Ra 0.0817 nm when the pH value is 9.

Abstract Background Topoisomerase II alpha (TOP2A) protein has been shown to be a proliferation marker associated with tumor grade and Ki67 index. The prognostic effect of TOP2A seems different among different subtypes of breast cancer. The current study evaluated the prognostic impact of TOP2A protein on luminal breast cancer. Method Altogether 434 stage I-II luminal breast cancer patients who underwent curative surgery in Sun Yat-Sen University Cancer Center between 2007 and 2009 were enrolled. TOP2A protein expression was assessed by immunohistochemistry. Clinical and pathological data were retrospectively collected. Result With a cut-off value of 30%, 127 (29.3%) patients were classified as TOP2A overexpression. TOP2A overexpression was associated with a higher tumor grade and Ki67 index. Patients with TOP2A high expression showed a significantly higher rate of distant metastasis and shorter distant metastasis free survival (DMFS) compared with patients with low TOP2A expression. The prognostic influence of TOP2A expression was more significant in years 5–8 after diagnosis, and more pronounced in stage II patients, luminal B disease, and patients treated with adjuvant endocrine therapy alone. Multivariate survival analysis revealed TOP2A overexpression was an independent fact for worse DMFS. Conclusion TOP2A protein showed a time dependent influence on prognosis in stage I-II luminal breast cancer, suggesting it might be a potential predictor of late recurrence for this group of patients.

PURPOSE: The incidence of anaplastic lymphoma kinase (ALK) rearrangement in pulmonary sarcomatoid carcinoma (PSC) is controversial. In this study, we aimed to reveal the reliable frequency and the clinical-pathologic characteristics of pulmonary sarcomatoid carcinoma (PSC) with ALK rearrangement in Chinese population, and to provide insight into the translatability of anti-ALK treatment in this treatment-refractory disease. METHODS: Immunohistochemistry (IHC) using a Ventana anti-ALK (D5F3) rabbit monoclonal antibody was performed in 141 PSC specimens collected from multiple medical centers. IHC-positive cases were then confirmed using ALK fluorescent in situ hybridization (FISH). The incidence rates and clinical-pathologic characteristics of ALK-rearranged PSC were then analyzed. Response to ALK inhibitor crizotinib in a patient with ALK-rearranged PSC was evaluated according to the response evaluation criteria for solid tumors (RECIST) version 1.1. RESULTS: Five of 141 (3.5%) of PSCs showed ALK rearrangement-positive by IHC and then were confirmed by FISH. Two were carcinosarcomas and the other three were pulmonary pleomorphic carcinoma (PPC). Strong positive ALK rearrangement was observed in both the epithelioid and sarcomatoid components. The median age of ALK-positive patients was younger than that of ALK-negative patients. PSCs in never-smokers were more likely to harbor ALK rearrangement than those in former or current smokers (P

Lapping is one of the standard essential methods to realise the global planarization of SiC and other semiconductor substrates. It is necessary to deeply study the mechanism to obtain SiC lapping process parameters with a strong comprehensive lapping performance (i.e., high material removal rate (MRRm), small surface roughness (Ra), and low total thickness variation (TTV)). The effects of the lapping process parameters and their interactions on lapping performance for SiC were investigated using orthogonal experiments; the effects on the MRRm, Ra, TTV, and optimal parameters under the conditions of a single evaluation index were investigated using intuitive analysis (range analysis, variance analysis, and effect curve analysis). The entropy value method and grey relational analysis were used to transform the multi-evaluation-index optimisation into a single-index optimisation about the grey relational grade (GRG) and to comprehensively evaluate the lapping performance of each process parameter. The results showed that the lapping plate types, abrasive size, and their interaction effect had the most significant effects on MRRm and Ra, with a contribution of over 85%. The interaction between the lapping plate types and abrasive size was also found to have the most significant effect on TTV, with a contribution of up to 51.07%. As the lapping plate’s hardness and abrasive size increased, the MRRm and Ra also gradually increased. As the lapping normal-pressure increased, MRRm increased, Ra gradually decreased, and TTV first decreased and then increased. MRRm, Ra, and TTV first increased and then decreased with increasing abrasive concentration. Compared to the optimisation results obtained by intuitive analysis, the process parameter optimised by the grey relational analysis resulted in a smooth surface with an MRRm of 90.2 μm/h, an Ra of 0.769 nm, and a TTV of 3 μm, with a significant improvement in the comprehensive lapping performance. This study reveals that a combination of orthogonal experiments and grey relational analysis can provide new ideas for optimising the process parameters of SiC.

Neural networks and fuzzy systems are two soft-computing paradigms for system modelling. Adapting a neural or fuzzy system requires to solve two optimization problems: structural optimization and parametric optimization. Structural optimization is a discrete optimization problem which is very hard to solve using conventional optimization techniques. Parametric optimization can be solved using conventional optimization techniques, but the solution may be easily trapped at a bad local optimum. Evolutionary computation is a general-purpose stochastic global optimization approach under the universally accepted neo-Darwinian paradigm, which is a combination of the classical Darwinian evolutionary theory, the selectionism of Weismann, and the genetics of Mendel. Evolutionary algorithms are a major approach to adaptation and optimization. In this paper, we first introduce evolutionary algorithms with emphasis on genetic algorithms and evolutionary strategies. Other evolutionary algorithms such as genetic programming, evolutionary programming, particle swarm optimization, immune algorithm, and ant colony optimization are also described. Some topics pertaining to evolutionary algorithms are also discussed, and a comparison between evolutionary algorithms and simulated annealing is made. Finally, the application of EAs to the learning of neural networks as well as to the structural and parametric adaptations of fuzzy systems is also detailed.

Uncontrolled hemorrhage remains the leading cause of trauma-related death. The QuikClot is the most popular hemostatic device; however, it also shows severe security hidden dangers. This study aimed to evaluate the efficacy and safety of the selfdeveloped Zeolite Dressing in a rabbit model of hemorrhagic groin injury. A complex hemorrhagic groin injury model was prepared in the right femoral artery of 96 rabbits by lacerating at a half depth and a subsequent 30-second spontaneous hemorrhage, and security was evaluated at the appointed time (1 and 3 weeks after operation). Experimental rabbits were randomly divided into four groups (12 males and 12 females in each group), including sham-operated control group without vascular injury (Con), standard gauze dressing (SD) group, QuikClot (QC) group and Zeolite Dressing (ZD) group. Vascular suture operation and postoperative debridement were thoroughly performed after hemostasis. The results showed that bleeding times of QC and ZD groups were significantly shorter than that of SD group (p<0.001), and there was nonsignificant difference between QC and ZD groups. Similarly, total blood loss was significantly reduced after the intervention of QC or ZD compared with SD group (p<0.001). No obvious toxic reaction was recorded 3 weeks after the treatment in both ZD and QC groups. A high wound temperature (maximum temperature was 96.6±27.9°C) and the "cooked meat like" pathological changes in the skin and muscle were only found in QC group. In conclusion, Zeolite Dressing exhibited a commensurate power to stop bleeding with higher safety compared with QuikClot, indicating that it is a practicable alternative for treating severe hemorrhage. [ABSTRACT FROM AUTHOR]

The increasing population and emerging business opportunities have led to a rise in consumer spending. Consequently, global credit card companies, including banks and financial institutions, face the challenge of managing the associated credit risks. It is crucial for these institutions to accurately classify credit card customers as “good” or “bad” to minimize capital loss. This research investigates the approaches for predicting the default status of credit card customer via the application of various machine-learning models, including neural networks, logistic regression, AdaBoost, XGBoost, and LightGBM. Performance metrics such as accuracy, precision, recall, F1 score, ROC, and MCC for all these models are employed to compare the efficiency of the algorithms. The results indicate that XGBoost outperforms other models, achieving an accuracy of 99.4%. The outcomes from this study suggest that effective credit risk analysis would aid in informed lending decisions, and the application of machine-learning and deep-learning algorithms has significantly improved predictive accuracy in this domain.