Feature selection (FS) is vitally important for determining the optimum subsets of features with effective information and maximizing the model accuracy. This study proposes a novel FS method based ...on global sensitivity analysis (GSA) for effectively determining the most relevant feature subsets and improving prediction performance of machine learning (ML)based models. Feature ranking is determined based on the results obtained from three global sensitivity analysis (GSA) including Pearson, Sobol’ and PAWN. This novel GSA-based FS method is applied to engineering practice with the combination of ML algorithm random forest (RF) to predict tunnelling-induced settlement prediction model. Meanwhile, the feature extraction method principle component analysis (PCA) is also used to develop RF-based model for comparing the performance of proposed GSA-based FS method. The results indicate the novel GSA-based FS method effectively determines the significance of input variables. The prediction performance of RF-based model with the integration of GSA-based FS methods is enhanced dramatically, and obviously outperforms the model with the integration of PCA-based dimensionality reduction method.
•A novel global sensitivity analysis based feature selection method is proposed.•Proposed feature selection method is integrated with random forest.•The performance of proposed feature selection is compared with principle component analysis.
Obesity is an independent risk factor of development and progression of chronic kidney disease (CKD). Data on the benefits of bariatric surgery in obese patients with impaired kidney function have ...been conflicting.
To explore whether there is improvement in glomerular filtration rate (GFR), proteinuria or albuminuria after bariatric surgery.
We comprehensively searched the databases of MEDLINE, Embase, web of science and Cochrane for randomized, controlled trials and observational studies that examined bariatric surgery in obese subjects with impaired kidney function. Outcomes included the pre- and post-bariatric surgery GFR, proteinuria and albuminuria. In obese patients with hyperfiltration, we draw conclusions from studies using measured GFR (inulin or iothalamate clearance) unadjusted for BSA only. Study quality was evaluated using the Newcastle-Ottawa Scale.
32 observational studies met our inclusion criteria, and 30 studies were included in the meta-analysis. No matter in dichotomous data or in dichotomous data, there were statistically significant reduction in hyperfiltration, albuminuria and proteinuria after bariatric surgery.
The main limitation of this meta-analysis is the lack of randomized controlled trials (RCTs). Another limitation is the lack of long-term follow-up.
Bariatric surgery could prevent further decline in renal function by reducing proteinuria, albuminuria and improving glomerular hyperfiltration in obese patients with impaired renal function. However, whether bariatric surgery reverses CKD or delays ESRD progression is still in question, large, randomized prospective studies with a longer follow-up are needed.
Background
Conditional survival (CS) represents the probability of surviving for additional years after the patient has survived for several years, dynamically describing the survival rate of the ...patient with the varying time of survival. The aim of this study was to evaluate the conditional cause-specific survival (CCSS) after chemotherapy and local treatment for metastatic breast cancer, and to identify the prognostic factors affecting the CCSS.
Methods
Patients diagnosed with primary stage IV breast cancer in the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2015 were included. CS is defined as the probability of additional survival for
y
years after the patient had survived
x
years with the calculation formula CCSS (
x
|
y
) = CSS (
x
+
y
)/CSS (
x
), where CSS(
x
) indicates the patient’s cause-specific survival rate at the time of
x
years. Cox proportional hazard models were used to evaluate predictors of CCSS.
Results
A total of 3,194 patients were included. The 5-year CSS was 39%, whereas the 5-year CCSS increased to 46%, 57%, 71%, and 85% after the diagnosis of 1, 2, 3, and 4 years. For patients with adverse clinical pathological features, CCSS had more pronounced increase with survival time and is more different from the CSS at diagnosis. No matter at the time of diagnosis or 1 year or 3 years after diagnosis, HER2 status, local treatment, and multisite metastasis were independent prognostic factors that affect the long-term survival of patients (all
P
< 0.05).
Conclusion
The 5-year CCSS of patients with stage IV breast cancer was extended as the survival years increased. HER2 status, multisite metastasis, and local treatment were independent prognostic factors even 3 years after diagnosis.
•GlcN yields and ACE inhibitory activities of mushroom hydrolysates were reported.•The parameters of sulfuric acid hydrolysis for GlcN production were optimized.•GlcN and ACE inhibitory peptides were ...fractionated, purified, and characterized.•ASPYAFGL was identified as potent ACE inhibitory peptide.•Mushroom hydrolysates could be considered as potential functional ingredients.
Glucosamine and ACE inhibitory peptides were produced from five mushrooms by sulfuric acid hydrolysis. Straw mushroom, hydrolysed at 100°C for 4.03h by 5.67molL−1 sulfuric acid, had the highest glucosamine yield (56.81mg/g), while oyster mushroom hydrolysate presented the maximum ACE inhibitory activity (IC50, 64.111mg/mL). Glucosamine was enriched 2-fold in the ethanol-soluble fraction after ultrafiltration and ethanol precipitation. The potent ACE inhibitory peptides were purified 69–175 folds by a four-step purification procedure. All the purified peptides showed satisfactory residual activity against heat, pH, and in vitro gastrointestinal digestion. Eighteen novel peptides were identified, of which ASPYAFGL was found to exhibit high inhibition potency with IC50 values of 1.080×10−7molL−1. These results suggested that mushroom hydrolysates containing glucosamine and ACE inhibitory peptides could be exploited as multifunctional ingredients for drugs or functional foods against osteoarthritis and hypertension.
In this paper, utilizing the same recognition group dinitrophenyl and hydroxyl functional NIR fluorophore hemicyanine, directly-linked probe CyNO2 and self-immolative probe CyBNO2 were developed for ...evaluation of sensing PhSH. Though CyNO2 was easily synthesized and sensitive to mercapto, the probe CyBNO2 showed higher selectivity, broader linear range from 1.0 × 10−7 to 7.0 × 10−6 M with lower detection limit of 22 nM for PhSH. Moreover, CyBNO2 was successfully applied for monitoring PhSH in living cells and in vivo, indicating the great potential of self-immolative probes.
Display omitted
•Comparison between directly-linked probe CyNO2 and self-immolative probe CyBNO2 for sensing thiophenols.•CyBNO2 showed a 160-fold response with a detection limit of 22 nM for thiophenols.•CyBNO2 was successfully applied for visualizing thiophenols in living cells and mice.
The primary treatment for patients with myocardial infarction (MI) is percutaneous coronary intervention (PCI). Despite this, the incidence of major adverse cardiovascular events (MACEs) remains a ...significant concern. Our study seeks to optimize PCI predictive modeling by employing an ensemble learning approach to identify the most effective combination of predictive variables.
We conducted a retrospective, non-interventional analysis of MI patient data from 2018 to 2021, focusing on those who underwent PCI. Our principal metric was the occurrence of 1-year postoperative MACEs. Variable selection was performed using lasso regression, and predictive models were developed using the Super Learner (SL) algorithm. Model performance was appraised by the area under the receiver operating characteristic curve (AUC) and the average precision (AP) score. Our cohort included 3,880 PCI patients, with 475 (12.2%) experiencing MACEs within one year. The SL model exhibited superior discriminative performance, achieving a validated AUC of 0.982 and an AP of 0.971, which markedly surpassed the traditional logistic regression models (AUC: 0.826, AP: 0.626) in the test cohort. Thirteen variables were significantly associated with the occurrence of 1-year MACEs.
Implementing the Super Learner algorithm has substantially enhanced the predictive accuracy for the risk of MACEs in MI patients. This advancement presents a promising tool for clinicians to craft individualized, data-driven interventions to better patient outcomes.
There is considerable potential for data‐driven modelling to describe path‐dependent soil response. However, the complexity of soil behaviour imposes significant challenges on the training efficiency ...and the ability to generalise. This study proposes a novel physics‐constrained hierarchical (PCH) training strategy to deal with existing challenges in using data‐driven models to capture soil behaviour. Different from previous strategies, the proposed hierarchical training involves ‘low‐level’ and ‘high‐level’ neural networks, and linear regression, in which the loss function of the neural network is constructed using physical laws to constrain the solution domain. Feedforward and long short‐term memory (LSTM) neural networks are adopted as baseline algorithms to further enhance the present method. The data‐driven model is then trained on random strain loading paths and subsequently integrated within a custom finite element (FE) analysis to solve boundary value problems by way of validation. The results indicate that the proposed PCH‐LSTM approach improves prediction accuracy, requires much less training data and has a lower performance sensitivity to the exact network architecture compared to traditional LSTM. When coupled with FE analysis, the PCH‐LSTM model is also shown to be a reliable means of modelling soil behaviour under loading‐unloading‐reloading and proportional strain loading paths. In addition, strain localisation and instability failure mechanisms can be accurately identified. PCH‐LSTM modelling overcomes the need for ad‐hoc network architectures thereby facilitating fast and robust model development to capture complex soil behaviours in engineering practice with less experimental and computational cost.
Regression loss function in object detection model plays an important factor during training procedure. The IoU based loss functions, such as CIOU loss, achieve remarkable performance, but still have ...some inherent shortages that may cause slow convergence speed. The paper proposes a Scale-Sensitive IOU(SIOU) loss for the object detection in multi-scale targets, especially the remote sensing images to solve the problem where the gradients of current loss functions tend to be smooth and cannot distinguish some special bounding boxes during training procedure in multi-scale object detection, which may cause unreasonable loss value calculation and impact the convergence speed. A new geometric factor affecting the loss value calculation, namely area difference, is introduced to extend the existing three factors in CIOU loss; By introducing an area regulatory factor <inline-formula> <tex-math notation="LaTeX">\gamma </tex-math></inline-formula> to the loss function, it could adjust the loss values of the bounding boxes and distinguish different boxes quantitatively. Furthermore, we also apply our SIOU loss to the oriented bounding box detection and get better optimization. Through extensive experiments, the detection accuracies of YOLOv4, Faster R-CNN and SSD with SIOU loss improve much more than the previous loss functions on two horizontal bounding box datasets, i.e, NWPU VHR-10 and DIOR, and on the oriented bounding box dataset, DOTA, which are all remote sensing datasets. Therefore, the proposed loss function has the state-of-the-art performance on multi-scale object detection.
Deep learning technology has been widely used in the military field, which have achieved great success. The traditional method for painting camouflage either using the background information or the ...artificial pattern. None of the traditional methods can both consider the background information and camouflage rules. In this paper, a new automatic camouflage generation framework is proposed. A method for generating camouflage pattern is designed. The imitation camouflage pattern is synthesized from the features of both background and artificial pattern. In our method, the texture feature of both background and traditional pattern patches are extracted from the feature maps of shallow neural network (SNN). Based on the feature maps, statistic information of second order differential and mean subtracted contrast normalized coefficients for texture and color is extracted. By iterating to optimize the imitation camouflage to be generated, the statistical information of the imitation camouflage can approximate the characteristic statistical information of the background and pattern. The new generated camouflage pattern can contain the color and texture information of background; besides, it can maintain the traditional patch camouflage criteria. Our approach makes camouflage painting more flexible and allows the target to better infuse into the background. And our method is designed for the preparation of painting camouflage.
Display omitted
•The deformation and stress characteristics of existing twin tunnels induced by EPBS under-crossing were obtained.•MJS columns reinforcement method was adopted to protect the existing ...tunnels.•The settlement development of the existing tunnels during EPBS tunneling was found to experience four stages.•Induced hoop stress of the existing tunnel during EPBS under-crossing was found to change sharply.•A skewed ovalisation of the final cross-section of the existing tunnel was observed.
This paper presented a case study on the deformation and stress characteristics of twin tunnels induced by close distance earth pressure balance shield (EPBS) under-crossing in sandy soil stratum, located in Changsha, China. The horizontal columns constructed by Metro Jet System (MJS) method were used to stabilize the sandy soil below the existing twin tunnels. The deformation and the stress of existing tunnels which affected by the construction of new tunnels were systematically monitored. The settlement development of the existing tunnels was found to experience four stages: (i) subsidence, (ii) heave, (iii) second subsidence, and (iv) steady state, respectively. The settlement profiles of the existing tunnels induced by the second tunnel under-crossing were found to be asymmetric with respect to the second tunnel centerline, and the location of the maximum settlement point deviated toward the twin tunnels center. The settlement caused by the second shield under-crossing was found significantly larger than that caused by the first tunnel under-crossing. After the two shields passing, the final settlement profile of the existing tunnels displayed a “U” shape. The induced hoop stress of the existing tunnels changed sharply during the shield under-crossing. The cross-section of the existing tunnels changed into a skewed oval shape. The final rotation direction of the existing left and right tunnels above the first and second tunnels centerline was opposite. Both the existing tunnels rotated toward the location of the large volume loss. The shifted Gaussian distribution curve was adopted to model the settlement profiles of the existing tunnels. The differences of the settlement profiles caused by the first and second tunnels excavation were also explained.