High Predictive Ability Research Articles

Predictive value of cholinesterase in patients with heart failure: A new blood biochemical marker of undernutrition

Background and aimsThis study was conducted to verify whether serum cholinesterase (ChE) is useful in predicting prognosis and discriminating undernutrition status compared to existing low-nutrition indices of blood chemical tests in patients with heart failure (HF). Methods and resultsA total of 1617 patients (1204 older patients) with HF who evaluated ChE during hospitalization were recruited for this study. The primary outcome was all-cause death, and multivariate survival analysis was performed. We drew a receiver operating characteristic curve for all-cause death, some undernutrition status, such as low body mass index, thin mid-upper arm circumference, low grip strength, and slow gait speed. The area under the curve was used to compare the predictive ability of ChE with some existing nutritional parameters, such as blood biochemical tests, controlling nutritional status (CONUT), and the geriatric nutritional risk index (GNRI). After adjusting for 29 variables, higher ChE significantly decreased the risk of all-cause death (per 10 increase, hazard ratio: 0.975, 95% confidence interval: 0.952–0.998), and this trend was maintained for older patients (per 10 increase, hazard ratio: 0.972, 95% confidence interval: 0.947–0.997). ChE was moderately correlated with CONUT and GNRI, but the predictive ability for all-cause death was higher for ChE relative to both scores. ChE tended to have an almost consistently high predictive ability compared with other blood biochemical tests. ConclusionsChE was associated with all-cause death and an almost consistently higher predictive ability for all-cause death and undernutrition status in comparison to existing blood chemical tests and nutritional scores.

Structural Analysis and Optimization of Urban Gas Pressure Regulator Based on Thermo-Hydro-Mechanical Coupling

As a core component in the gas transmission process, the internal wall surface of a gas pressure regulator is prone to failure due to long-term exposure to a high-pressure gas environment, resulting in poor reliability of the regulator. Thus, a thermo-hydro-mechanical coupling model for the FL gas pressure regulator is established in this paper, and the thermo-hydro-mechanical coupling results are verified by engineering data. The effect of valve opening on the parameters (temperature, deformation, and stress) of the gas pressure regulator is studied in detail through simulation. The results show that the stress is greater at the sleeve, valve bore, and outlet valve seat wall under the opening of 20% of the regulator. Finally, the response surface method is used to optimize the regulator to obtain a good fit and high predictive ability of the response surface equation. The optimal parameters for the gas pressure regulator are as follows: the wall thickness of the sleeve is 7.25 mm, the diameter of the valve bore is 25 mm, and the wall thickness of the outlet seat is 31.05 mm. The maximum equivalent stress with this combination of parameters is 135.62 MPa.

Determination of Gemcitabine and Sorafenib in Spiked Human Plasma Using Multivariate Model Update Chemometric Methods.

Gemcitabine (GEM), a pyrimidine nucleoside, has been used as a first-line treatment in non-small-cell lung cancer (NSCLC). Sorafenib (SOR), a nonselective multi-kinase inhibitor, is used as a chemotherapeutic agent in different types of cancers including NSCLC in preclinical studies. Co-administration of GEM and SOR was found to be effective and well-tolerated in the treatment of NSCLC. The aim of the present work is to determine the studied drugs in spiked human plasma simultaneously through resolving the overlapping spectra and removing the interference of the plasma matrix. Two updated chemometric models were developed using UV absorbance of the drugs, which named principal component regression (PCR) and partial least-squares (PLS) for determination of GEM and SOR in the ranges of 5-25 and 2-22 µg/mL, respectively. Validation of the two updated models has been achieved in accordance with US Food and Drug Administration (FDA) guidelines, and the results were satisfactory. The two methods had the advantages of high predictive ability of the studied drugs with high precision and accuracy. Moreover, there was no significant difference obtained when statistical comparison was done between the developed and reported methods, showing good validity of the suggested methods. The two updated models have the advantages of being rapid, accurate, sensitive, and cost-effective for the determination of GEM and SOR in quality control laboratories without any need for initial separation procedures. Two updated chemometric methods, PCR and PLS, were developed for the estimation of GEM and SOR in spiked human plasma using their UV absorbance data.

Genomic prediction of agronomic and malting quality traits in six-rowed winter barley

While two-rowed barley is usually preferred for malting and beer-making, six-rowed malting barley varieties appeared in Europe around 30 years ago, and several breeders have dedicated improvement programs on this specific germplasm. In this study, we evaluated the feasibility of genomic prediction for yield and malting related traits using 679 breeding lines from two French barley breeders, as well as a set of recently registered varieties. These lines were evaluated in five locations and two harvest years in an unbalanced design. Although the germplasm from the two breeders does show some trend towards differentiation, globally the whole panel did not show a clear-cut genetic structure. Predictive ability of GBLUP was evaluated through random cross-validation within and across breeder sets, and using cross-prediction between breeder sets. Results show moderate to high predictive ability (PA), particularly for malt friability and β-glucan content, for which predictive ability of 0.8 was obtained with training populations as small as 105 registered varieties and across breeding sets. The long range of useful linkage disequilibrium in this particular germplasm allows using as few as 2000 to 5000 markers to obtain high PA. Other prediction methods such as Bayesian LASSO, Bayes Cpi or EGBLUP did not improve predictive ability. These results are very encouraging for implementing genomic prediction of malting quality traits in applied breeding programs.

Predicting NSMR–concrete bond strength using artificial neural networks: A comparative‐analysis study

AbstractNear Surface‐Mounted Fiber Reinforced Polymer (NSM FRP) strips have become a popular technique for the rehabilitation of existing deteriorated concrete structures, the effectiveness of which is vitally related to its bond to concrete. The present Artificial Neural Network (ANN) model, aimed at providing precise predictions for bond strength, utilizes nearly 550 refined dataset points and accounts for more influential parameters, namely elastic modulus, cross‐sectional area, bond length of FRP strips, groove size, concrete edge distance, groove spacing, and concrete's compressive strength. The very few relevant publications for the prediction of bond strength between NSM FRP composites and concrete employing artificial intelligence bond models considered limited database sizes and few key parameters. The Matlab software was utilized to build, train, and test the network model using seven input variables and one targeted output. Detailed statistical analysis was presented to test the model's validity with Taylor charts produced using the Matlab software for the purpose of comparing the performance of the present model against different literature models. The proposed ANN model showed a high prediction ability and a low mean square error. Indeed, the mean square error for the testing and validation data remained below 0.00055, whereas the coefficient of determination exceeded 0.99. The error charts for training and testing database revealed normal distribution of bond strength residuals; supporting further the validity of the proposed model. The trending behavior of the ultimate bond force versus the key parameters was consistent with that captured in the various literature models. The statistical and the sensitively analyses performed stipulated the validity of the present model for producing reliable prediction for bond strength between NSM FRP strips and concrete.

Genomewide selection for fruit quality traits in apple: breeding insights gained from prediction and postdiction.

Many fruit quality traits in apple (Malus domestica Borkh.) are controlled by multiple small-effect quantitative trait loci (QTLs). Genomewide selection (genomic selection) might be an effective breeding approach for highly quantitative traits in woody perennial crops with long generation times like apple. The goal of this study was to determine if genomewide prediction is an effective breeding approach for fruit quality traits in an apple scion breeding program. Representative apple scion breeding germplasm (nindividuals= 955), high-quality single nucleotide polymorphism (SNP) data (nSNPs= 977), and breeding program fruit quality trait data at harvest were analyzed. Breeding parents `Honeycrisp' and `Minneiska' were highly represented. Moderate to high predictive abilities were observed for most fruit quality traits at harvest. For example, when 25% random subsets of the germplasm set were used as training sets, mean predictive abilities ranged from 0.35 to 0.54 across traits. Trait, training and test sets, family size for within family prediction, and number of SNPs per chromosome affected model predictive ability. Inclusion of large-effect QTLs as fixed effects resulted in higher predictive abilities for some traits (e.g. percent red overcolor). Postdiction (i.e. retrospective) analyses demonstrated the impact of culling threshold on selection decisions. The results of this study demonstrate that genomewide selection is a useful breeding approach for certain fruit quality traits in apple.

Use of HBV RNA and to predict change in serological status and disease activity in CHB.

Predicting changes in disease activity and serological endpoints is necessary for the management of patients with chronic hepatitis B (CHB). We examined whether HBV RNA and hepatitis B core-related antigen (HBcrAg), two specialized virological markers proposed to reflect the activity of covalently closed circular DNA, may improve the ability to predict not sustained inactive carrier phase, spontaneous alanine aminotransferase (ALT) flare, HBeAg loss, and HBsAg loss. Among eligible participants enrolled in the North American Hepatitis B Research Network Adult Cohort Study, we evaluated demographic, clinical, and virologic characteristics, including HBV RNA and HBcrAg, to predict not sustained inactive carrier phase, ALT flare, HBeAg loss, and HBsAg loss through a series of Cox proportional hazard or logistic regression models, controlling for antiviral therapy use. Among the study population, 54/103 participants experienced not sustained inactive carrier phase, 41/1006 had a spontaneous ALT flare, 83/250 lost HBeAg, and 54/1127 lost HBsAg. HBV RNA or HBcrAg were predictive of all 4 events. However, their addition to models of the readily available host (age, sex, race/ethnicity), clinical (ALT, use of antiviral therapy), and viral factors (HBV DNA), which had acceptable-excellent accuracy (e.g., AUC = 0.72 for ALT flare, 0.92 for HBeAg loss, and 0.91 for HBsAg loss), provided only small improvements in predictive ability. Given the high predictive ability of readily available markers, HBcrAg and HBV RNA have a limited role in improving the prediction of key serologic and clinical events in patients with CHB.

Potential role of plasma branched-chain amino acids in the differential diagnosis of acute cerebral venous thrombosis

Cerebral venous thrombosis (CVT) is a special and easily misdiagnosed or undiagnosed subtype of stroke. To identify specific biomarkers with a high predictive ability for the diagnosis of acute CVT, we performed metabolomic analysis in plasma samples from acute CVT patients and healthy controls and confirmed the results in validation cohorts. In the discovery stage, there were 343 differential metabolites, and the caffeine metabolism pathway and the biosynthesis pathway for the branched chain amino acids (BCAAs) valine, leucine, and isoleucine were two significant pathways between the CVT and healthy cohorts. The area under the curve (AUC) for metabolites associated with valine, leucine, and isoleucine biosynthesis was 0.934. In the validation stage, the BCAA concentrations demonstrated an AUC of 0.935 to differentiate patients with acute CVT from the control cohort. In addition, BCAAs combined with D-dimer levels were used to establish a diagnostic model for CVT, and the AUC was 0.951, showing good diagnostic efficacy of separating CVT patients from the control cohort. BCAAs as plasma biomarkers deserve to be further studied and even developed in clinical CVT management.

An effective prognostic model for assessing prognosis of non-small cell lung cancer with brain metastases.

Background: Brain metastasis, with an incidence of more than 30%, is a common complication of non-small cell lung cancer (NSCLC). Therefore, there is an urgent need for an assessment method that can effectively predict brain metastases in NSCLC and help understand its mechanism. Materials and methods: GSE30219, GSE31210, GSE37745, and GSE50081 datasets were downloaded from the GEO database and integrated into a dataset (GSE). The integrated dataset was divided into the training and test datasets. TCGA-NSCLC dataset was regarded as an independent verification dataset. Here, the limma R package was used to identify the differentially expression genes (DEGs). Importantly, the RiskScore model was constructed using univariate Cox regression analysis and least absolute shrinkage and selection operator (LASSO) analysis. Moreover, we explored in detail the tumor mutational signature, immune signature, and sensitivity to treatment of brain metastases in NSCLC. Finally, a nomogram was built using the rms package. Results: First, 472 DEGs associated with brain metastases in NSCLC were obtained, which were closely associated with cancer-associated pathways. Interestingly, a RiskScore model was constructed using 11 genes from 472 DEGs, and the robustness was confirmed in GSE test, entire GSE, and TCGA datasets. Samples in the low RiskScore group had a higher gene mutation score and lower immunoinfiltration status. Moreover, we found that the patients in the low RiskScore group were more sensitive to the four chemotherapy drugs. In addition, the predictive nomogram model was able to effectively predict the outcome of patients through appropriate RiskScore stratification. Conclusion: The prognostic RiskScore model we established has high prediction accuracy and survival prediction ability for brain metastases in NSCLC.

A comparison of the diagnostic capabilities of the ratio of acceleration time to total left ventricular ejection time (AT/ET) in determining the severity of aortic stenosis in patients with bicuspid and tricuspid aortic valve: retrospective comparative study

Objective. We aimed to compare the diagnostic capabilities of the ratio of acceleration time to total left ventricular ejection time (AT/ET) in determining the severity of aortic stenosis (AS) in patients with bicuspid and tricuspid aortic valves (AV).
 Material and methods. We retrospectively analyzed the data of 187 patients with moderate and severe AS who underwent diagnostic examination at the Penza Federal Center for Cardiovascular Surgery. The patients were divided into 2 groups based on whether their AV was tricuspid or bicuspid. Visual assessment of the AV structure was performed using transthoracic echocardiography (TTE). In indeterminate cases, computed tomography was used for the assessment.
 Results. A comparative analysis of the echocardiographic characteristics of patients with tricuspid and bicuspid AV did not reveal a statistically significant difference between the patient groups (p 0.05). Linear regression analysis in patients with a tricuspid AV demonstrated a statistically significant correlation between AT/ET scores and peak gradient (Gmax) (r=0.68, р=0.03), mean gradient (Gmean) (r=0.78, р=0.01), effective orifice area (EOA) (r=0.7, р=0.03), and doppler velocity index (DVI) scores (r=0.72, р=0.02). In patients with a bicuspid AV, a similarly significant correlation was found between the AT/ET index and Gmax (r=0.67, р=0.02), Gmean (r=0.8, р 0.001), EOA (r=0.72, р=0.04), and DVI (r=0.75, р=0.01). The receiver operating characteristic analysis demonstrated a high predictive ability of AT/ET for severe aortic valve stenosis (with a value 0.35). The area under the curve in patients with tricuspid and bicuspid AV was 84 (p 0.001) and 86 (p 0.001), respectively. For determining severe AV stenosis in patients with a tricuspid AV, the sensitivity and specificity of AT/ET 0.35 was 84% and 75%, respectively; and in patients with a bicuspid AV, it was 87% and 78%, respectively.
 Conclusion. The AT/ET ratio has comparable diagnostic capabilities in determining severe AS in patients with tricuspid and bicuspid AV structures. The AT/ET 0.35 is a highly sensitive parameter for defining severe AS for both morphologies of AV.

Abstract 3047: Association of 16S rRNA fecal microbiome data with colorectal cancer risk: A pooled analysis

Abstract Changes in the fecal microbial composition may have a potential role in colorectal carcinogenesis and could provide insights into an individual’s risk of developing colorectal cancer (CRC). Although the previous studies observed the association between fecal microbiome and CRC risk, the evidence from large-scale population-based studies are lacking and the causality is still yet to be established. To address this gap, we conducted a microbiome-wide association study (MWAS) to identify CRC associated unique microbial markers among Koreans. Three publicly available datasets were downloaded from NCBI Sequence Read Archive under accession numbers PRJNA747730 (70 CRC cases), PRJNA797640 (200 CRC cases), and PRJNA644479 (179 healthy controls). Additionally, 295 CRC patients were recruited from the Center for Colorectal Cancer, National Cancer Center (NCC) hospital, Korea. Total study population comprised with 565 CRC cases and 179 healthy controls. Fecal samples of NCC cases were collected and 16S rRNA gene sequencing was performed. Forward sequence reads from all four datasets were used for quality control and taxonomic annotation using QIIME2 pipeline. Gaussian graphical models were applied to derive a microbial network and microbial communities were detected within the network using R package “linkcomm”. Microbial community-specific scores were calculated for each cluster and the association between community-specific score and CRC risk was observed using unconditional logistic regression models. Microbial communities were evaluated using machine learning models to identify CRC specific bacterial markers. There was a significant divergence of the microbial composition between CRC patients and controls (PERMANOVA p=0.001). Faecalibacterium and Megamonas were enriched in healthy controls whereas Escherichia-Shigella was enriched in CRC cases. Out of 24 communities, community 6: odds ratio (OR=2.68), 95% confidence intervals (CI=1.18-6.09, p-trend=0.006), 7 (OR=4.33, 95% CI=1.86-10.05, p-trend=0.001), and 11 (OR=7.28, 95% CI=3.35-15.81) were positively associated while community 19 (OR=0.34, 95% CI=0.15-0.78, p-trend=0.015) was negatively associated with CRC risk. Random forest showed the best performance to predict CRC risk based on 10 most important microbial communities (AUC=0.97). Community 20 with five genera had highest predictive ability (AUC=0.89) whereas Christensenellaceae_R_7 genus was identified as a unique microbial marker to distinguish CRC patients from healthy controls (AUC=0.85). This MWAS could help to identify Christensenellaceae_R_7 as the unique marker to distinguish CRC patients from healthy controls. Thus, it could work as a promising indicator for CRC diagnosis among Koreans specifically implementing in CRC screening programs. Citation Format: Madhawa Gunathilake, Jeongseon Kim. Association of 16S rRNA fecal microbiome data with colorectal cancer risk: A pooled analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3047.

Elevated Serum NOX2 Levels Contribute to Delayed Cerebral Ischemia and a Poor Prognosis After Aneurysmal Subarachnoid Hemorrhage: A Prospective Cohort Study.

NADPH oxidase 2 (NOX2) is highly expressed in injured brain tissues. We determined serum NOX2 levels of aneurysmal subarachnoid hemorrhage (aSAH) patients and further investigated correlation of serum NOX2 levels with disease severity, delayed cerebral ischemia (DCI) plus prognosis after aSAH. Serum NOX2 levels were measured in 123 aSAH patients and 123 healthy controls. World Federation of Neurological Surgeons scale (WFNS) score and modified Fisher (mFisher) score were utilized to assess disease severity. Modified Rankin scale (mRS) score was used to evaluate the clinical prognosis at 90 days after aSAH. Relations of serum NOX2 levels to DCI and 90-day poor prognosis (mRS score of 3-6) were analyzed using multivariate analysis. Receiver operating characteristic curve (ROC) was built to evaluate the prognostic predictive capability. Serum NOX2 levels in aSAH patients, compared with healthy controls, were significantly increased, and were independently correlated with WFNS score, mFisher score and post-stroke 90-day mRS score. Patients with poor prognosis or DCI had significantly higher serum NOX2 levels than other remainders, and serum NOX2 levels independently predicted 90-day poor prognosis and DCI. Serum NOX2 had high prognosis and DCI predictive abilities, and their areas under ROC curve were similar to those of WFNS score and mFisher score. Serum NOX2 levels are significantly associated with hemorrhage severity, poor 90-day prognosis and DCI in aSAH patients. Hence, complement NOX2 may serve as a potential prognostic biomarker after aSAH.

Prediction of ciprofloxacin resistance in hospitalized patients using machine learning

BackgroundCiprofloxacin is a widely used antibiotic that has lost efficiency due to extensive resistance. We developed machine learning (ML) models that predict the probability of ciprofloxacin resistance in hospitalized patients.MethodsData were collected from electronic records of hospitalized patients with positive bacterial cultures, during 2016-2019. Susceptibility results to ciprofloxacin (n = 10,053 cultures) were obtained for Escherichia coli, Klebsiella pneumoniae, Morganella morganii, Pseudomonas aeruginosa, Proteus mirabilis and Staphylococcus aureus. An ensemble model, combining several base models, was developed to predict ciprofloxacin resistant cultures, either with (gnostic) or without (agnostic) information on the infecting bacterial species.ResultsThe ensemble models’ predictions are well-calibrated, and yield ROC-AUCs (area under the receiver operating characteristic curve) of 0.737 (95%CI 0.715–0.758) and 0.837 (95%CI 0.821–0.854) on independent test-sets for the agnostic and gnostic datasets, respectively. Shapley additive explanations analysis identifies that influential variables are related to resistance of previous infections, where patients arrived from (hospital, nursing home, etc.), and recent resistance frequencies in the hospital. A decision curve analysis reveals that implementing our models can be beneficial in a wide range of cost-benefits considerations of ciprofloxacin administration.ConclusionsThis study develops ML models to predict ciprofloxacin resistance in hospitalized patients. The models achieve high predictive ability, are well calibrated, have substantial net-benefit across a wide range of conditions, and rely on predictors consistent with the literature. This is a further step on the way to inclusion of ML decision support systems into clinical practice.

Development of advanced machine learning models for optimization of methyl ester biofuel production from papaya oil: Gaussian process regression (GPR), multilayer perceptron (MLP), and K-nearest neighbor (KNN) regression models

Data-driven machine learning (ML) methods are extensively employed for modeling and simulation of highly complicated processes. ML techniques confirmed their great predictive capability compared to conventional techniques for modeling and management of non-linear relationships between input and output parameters. Biofuels as renewable sources of energy are a significant potential alternative to fossil fuels. Due to the non-linearity and complexity of biofuels production processes and increasing energy conversion, accurate and fast modeling tools are necessary for design and optimization of these processes. Hence, in this research, ML modeling techniques were developed for simulation of biofuel production from energy conversion of Papaya oil through transesterification process. In order to simulate and optimize the content Papaya oil methyl ester (POME) production, Gaussian Process Regression (GPR), Multilayer perceptron (MLP), and K-nearest neighbor (KNN) regression models, as well as adaptive boosting for amplification, were employed. The temperature of reaction, catalyst quantity, time of process, and methanol to oil molar ratio were considered as the inputs of models while the POME yield was the model output. The obtained results showed that the R2-score of 0.988, 0.993, and 0.994 were obtained for Boosted MLP, Boosted GPR, and Boosted KNN, respectively, which demonstrate the high predictive ability of these models. Also, the RMSE metric error rates of 9.8071, 4.8150, and 6.5180 corresponded to Boosted MLP, Boosted GPR, and Boosted KNN, respectively. We examined performance using another metric, MAE: 8.38008, 2.3184, and 5.21954 errors were observed for Boosted MLP, Boosted GPR, and Boosted KNN, respectively. The optimized POME production yield of 99.89% was observed at temperature of 62.5 °C, 6.47 min of reaction, catalyst quantity of 0.8125 wt% and methanol to oil molar ratio of 10.33. The obtained results of this study show that the ML techniques are highly recommended for prediction of biofuels production as cost and time saving methods.

Data analysis and machine learning in the context of customer churn prediction

Due to the fierce competition in the market, customers are often faced with multiple choices when choosing products and services. So many industries, including banking, are now facing the problem of how to address customer churn. At the same time, in order to improve the quality of service for users, banks and other institutions need to conduct in-depth research on the characteristics of customers. This paper provides solutions to the above two problems by using data analysis and mining technology and machine learning technology in artificial intelligence. The study provides an in-depth exploration of cus-tomer churn likelihood by analyzing customer behavior and characteristics. This study used data analysis methods such as Chi-square test, Mann Whitney U test, linear regres-sion, and machine learning methods such as logistic regression, random forest, and XGBoost. This research uses public datasets on Kaggle. This study uses data analysis techniques to provide recommendations on how the banking industry should improve service quality, and establishes 6 models with better performance to predict customer churn. In addition, this paper also uses a variety of evaluation indicators to compare the model performance, and selects the random forest model with high predictive ability as the most suitable model. In addition, the order of importance of the factors responsible for customer churn was successfully derived.

Systemic inflammatory prognostic scores in advanced pancreatic adenocarcinoma.

Systemic inflammatory scores may aid prognostication and patient selection for trials. We compared five scores in advanced pancreatic adenocarcinoma (PDAC). Unresectable/metastatic PDAC patients enrolled in the Comprehensive Molecular Characterisation of Advanced Pancreatic Ductal Adenocarcinoma for Better Treatment Selection trial (NCT02750657) were included. Patients had pre-treatment biopsies for whole genome and RNA sequencing. CD8 immunohistochemistry was available in a subset. The neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio, Prognostic Nutritional Index, Gustave Roussy Immune Score (GRIm-S), and Memorial Sloan Kettering Prognostic Score (MPS) were calculated. Overall survival (OS) was estimated using Kaplan-Meier methods. Associations between inflammatory scores, clinical/genomic characteristics, and OS were analysed. We analysed 263 patients. High-risk NLR, GRIm-S and MPS were poorly prognostic. The GRIm-S had the highest predictive ability: median OS 6.4 vs. 10 months for high risk vs. low-risk (P < 0.001); HR 2.26 (P < 0.001). ECOG ≥ 1, the basal-like subtype, and low-HRDetect were additional poor prognostic factors (P < 0.01). Inflammatory scores did not associate with RNA-based classifiers or homologous recombination repair deficiency genotypes. High-risk MPS (P = 0.04) and GRIm-S (P = 0.02) patients had lower median CD8 + tumour-infiltrating lymphocytes. Inflammatory scores incorporating NLR have prognostic value in advanced PDAC. Understanding immunophenotypes of poor-risk patients and using these scores in trials will advance the field.

Experimental study and fretting fatigue life prediction in dovetail assembly considering stress–temperature nonlinear coupling effect

AbstractAs a paramount failure form of dovetail assembly, high‐temperature fretting fatigue seriously restricts the actual service life. Aiming at the problem of fretting fatigue life prediction of dovetail assembly of the aero‐engine compressor under the coexistence of stress and high temperature, based on D‐M contact theory and continuum damage mechanics theory, a fretting fatigue life prediction model considering stress–temperature nonlinear coupling effect is proposed in this paper by, respectively, introducing the fretting fatigue stress σfretting and the high‐temperature influence factor λ. Meanwhile, the judgment criterion of fretting fatigue crack initiation and failure is raised based on the strain–displacement method, and the fretting fatigue test is designed and carried out for experimental verification. The results show that the proposed fretting fatigue life prediction model has high life prediction ability, whose life prediction results are within ±1.7 times the error band, which can provide a theoretical reference for the structural design optimization and damage tolerance design of dovetail assembly.

Exploring nonlinear strengthening in polycrystalline metallic materials by machine learning methods and heterostructure design

To improve the strength and plasticity of structural materials, researchers often introduce various strengthening mechanisms such as second-phase strengthening, dislocation strengthening, and back stress strengthening (HDI). Due to the interaction of multiple mechanisms, the linear superposition relationship has a poor fitting effect and is only used for rough calculations of the strengthening mechanisms. In this study, the transfer learning data was used to optimize the deep learning network structure (Re-CNN) based on the residual algorithm, and the yield strength prediction physical neural informed model (PNIM) of polycrystalline metallic materials was established. To promote the industrial application of the heterostructure design method, a medium carbon steel heterostructure design strategy based on the existing equipment of the factory was proposed. Medium-carbon heterostructure materials (MHSM) with mixed strengthening mechanisms were successfully prepared. MHSM exhibits excellent comprehensive mechanical properties. When a linear relationship is used to describe the MHSM yield strength, there is a large error, while Re-CNN shows satisfactory prediction accuracy. The linear relationship is incompatible with homogeneous structure materials and heterogeneous structure materials, and its universality is lower than that of nonlinear Re-CNN. Re-CNN shows high cross-scale prediction ability and can be compatible with homogeneous microstructures and heterogeneous microstructures. Using the heterogeneity evolution characteristics of MHSM, the key factors deviating from the linear relationship were revealed. The overestimation and underestimation of the linear relation are demonstrated by Taylor factor and TEM analysis to be caused by the multiscale properties of ferrite, the behavior of the second phase particles, and the interaction of various mechanisms. This study provides a new idea for the cross-scale calculation of the mechanical properties of polycrystalline metallic materials.

Multicentric development and evaluation of 18F-FDG PET/CT and MRI radiomics models to predict para-aortic lymph node involvement in locally advanced cervical cancer.

To develop machine learning models to predict para-aortic lymph node (PALN) involvement in patients with locally advanced cervical cancer (LACC) before chemoradiotherapy (CRT) using 18F-FDG PET/CT and MRI radiomics combined with clinical parameters. We retrospectively collected 178 patients (60% for training and 40% for testing) in 2 centers and 61 patients corresponding to 2 further external testing cohorts with LACC between 2010 to 2022 and who had undergone pretreatment analog or digital 18F-FDG PET/CT, pelvic MRI and surgical PALN staging. Only primary tumor volumes were delineated. Radiomics features were extracted using the Radiomics toolbox®. The ComBat harmonization method was applied to reduce the batch effect between centers. Different prediction models were trained using a neural network approach with either clinical, radiomics or combined models. They were then evaluated on the testing and external validation sets and compared. In the training set (n = 102), the clinical model achieved a good prediction of the risk of PALN involvement with a C-statistic of 0.80 (95% CI 0.71, 0.87). However, it performed in the testing (n = 76) and external testing sets (n = 30 and n = 31) with C-statistics of only 0.57 to 0.67 (95% CI 0.36, 0.83). The ComBat-radiomic (GLDZM_HISDE_PET_FBN64 and Shape_maxDiameter2D3_PET_FBW0.25) and ComBat-combined (FIGO 2018 and same radiomics features) models achieved very high predictive ability in the training set and both models kept the same performance in the testing sets, with C-statistics from 0.88 to 0.96 (95% CI 0.76, 1.00) and 0.85 to 0.92 (95% CI 0.75, 0.99), respectively. Radiomic features extracted from pre-CRT analog and digital 18F-FDG PET/CT outperform clinical parameters in the decision to perform a para-aortic node staging or an extended field irradiation to PALN. Prospective validation of our models should now be carried out.

Banking on a new understanding: translational opportunities from veterinary biobanks.

Current advances in geroscience are due in part to the discovery of biomarkers with high predictive ability in short-lived laboratory animals such as flies and mice. These model species, however, do not always adequately reflect human physiology and disease, highlighting the need for a more comprehensive and relevant model of human aging. Domestic dogs offer a solution to this obstacle, as they share many aspects not only of the physiological and pathological trajectories of their human counterpart, but also of their environment. Furthermore, they age at a considerably faster rate. Studying aging in the companion dog provides an opportunity to better understand the biological and environmental determinants of healthy lifespan in our pets, and to translate those findings to human aging. Biobanking, the systematic collection, processing, storage, and distribution of biological material and associated data has contributed to basic, clinical, and translational research by streamlining the management of high-quality biospecimens for biomarker discovery and validation. In this review, we discuss how veterinary biobanks can support research on aging, particularly when integrated into large-scale longitudinal studies. As an example of this concept, we introduce the Dog Aging Project Biobank.