Regulatory mechanisms used by ZmMYB39 to enhance drought tolerance in maize (Zea mays) seedlings.

Drought is a significant abiotic stressor that limits maize (Zea mays L.) growth and development. Thus, enhancing drought tolerance is critical for promoting maize production. Our findings demonstrated that ZmMYB39 is an MYB transcription factor with transcriptional activation activity. Drought stress experiments involving ZmMYB39 overexpression and knockout lines indicated that ZmMYB39 positively regulated drought stress tolerance in maize. DAP-Seq, EMSA, dual-LUC, and RT-qPCR provided initial insights into the molecular regulatory mechanisms by which ZmMYB39 enhances drought tolerance in maize. ZmMYB39 directly promoted the expression of ZmP5CS1, ZmPOX1, ZmSOD2, ZmRD22, ZmNAC49, and ZmDREB2A, which are involved in stress resistance. ZmMYB39 enhanced drought tolerance by interacting with and promoting the expression of ZmFNR1, ZmHSP20, and ZmDOF6. Our study offers a theoretical basis for understanding the molecular regulatory networks involved in maize drought stress response. Furthermore, ZmMYB39 serves as a valuable genetic resource for breeding drought-resistant maize.

Association between dietary magnesium intake and gallstones: the mediating role of atherogenic index of plasma.

BACKGROUND: Dyslipidemia and abnormalities in cholesterol metabolism are commonly observed in individuals with gallstone disease. Previous research has demonstrated that dietary magnesium can influence lipid metabolism. The atherogenic index of plasma (AIP) has emerged as a novel lipid marker. This study aimed to examine the possible correlation between dietary magnesium intake and gallstones and the potential mediating role of AIP in US adults. METHODS: A total of 4,841 adults were included in this study from the National Health and Nutrition Examination Survey (NHANES) conducted from 2017 to 2020. A variety of statistical techniques such as logistic regression, subgroup analysis, smoothed curve fitting, and causal mediation analysis were utilized to analyze the information collected from the participants. RESULTS: In the fully adjusted model, a statistically noteworthy inverse relationship was observed between dietary magnesium intake and the presence of gallstones, as indicated by an odds ratio (OR) of 0.58 and a 95% confidence interval (CI) of (0.42, 0.81). Causal intermediary analysis revealed that the association between magnesium intake and gallstones was partially mediated by AIP, with a mediation ratio of 3.2%. CONCLUSION: According to this study, dietary magnesium intake had a significant linear negative association with the prevalence of gallstones, in which AIP played a mediating role. This discovery offers novel perspectives on the prevention and management of gallstones.

Generative Variational-Contrastive Learning for Self-Supervised Point Cloud Representation.

Self-supervised representation learning for 3D point clouds has attracted increasing attention. However, existing methods in the field of 3D computer vision generally use fixed embeddings to represent the latent features, and impose hard constraints on the embeddings to make the latent feature values of the positive samples converge to consistency, which limits the ability of feature extractors to generalize over different data domains. To address this issue, we propose a Generative Variational-Contrastive Learning (GVC) model, where Gaussian distribution is used to construct a continuous, smoothed representation of the latent features. A distribution constraint and cross-supervision are constructed to improve the transfer ability of the feature extractor over synthetic and real-world data. Specifically, we design a variational contrastive module to constrain the feature distribution instead of feature values corresponding to each sample in the latent space. Moreover, a generative cross-supervision module is introduced to preserve the invariance features and promote the consistency of feature distribution among positive samples. Experimental results demonstrate that GVC achieves SOTA on different downstream tasks. In particular, with only pre-training on the synthetic dataset, GVC achieves a lead of 8.4% and 14.2% when transferring to the real-world dataset in the linear classification and few-shot classification.

Epigallocatechin-3-gallate protects sepsis-induced myocardial dysfunction by inhibiting the nuclear factor-κB signaling pathway.

Sepsis-induced myocardial dysfunction (SIMD) has become one of the most lethal complications of sepsis, while the treatment was limited by a shortage of pertinent drugs. Epigallocatechin-3-gallate (EGCG) is the highest content of active substances in green tea, and its application in cardiovascular diseases has broad prospects. This study was conducted to test the hypothesis that EGCG was able to inhibit lipopolysaccharide (LPS) induced myocardial dysfunction and investigate the underlying molecular mechanisms. The cardiac systolic function was assessed by echocardiography. The cardiomyocyte apoptosis was determined by TUNEL staining. The expression of inflammatory factors and apoptosis-related protein, cardiac markers were examined by Western Blot and qRT-PCR. EGCG effectively improve LPS-induced cardiac function damage, enhance left ventricular systolic function, and restore myocardial cell vitality. It can effectively inhibit the upregulation of TLR4 expression induced by LPS and inhibit IκB α/NF- κB/p65 signaling pathway, thereby inhibiting cardiomyocyte apoptosis and improving myocarditis. In conclusion, EGCG protects against SIMD through anti-inflammatory and anti-apoptosis effects; it was mediated by the inhibition of the TLR4/NF-κB signal pathway. Our results demonstrated that EGCG might be a possible medicine for SIMD prevention and treatment.

Associations between per- and polyfluoroalkyl substances exposure and thyroid hormone levels in the elderly.

This study aimed to measure the exposure of the elderly to per- and polyfluoroalkyl substances (PFAS) and explore their effects on thyroid hormone levels. A cross-sectional study of plasma samples from 746 elderly people (aged >60 years) from Taiyuan, China was conducted. Fourteen PFASs were determined using liquid chromatography-tandem mass spectrometry and five thyroid function indicators, thyroid-stimulating hormone (TSH), thyroxine (T4), triiodothyronine (T3), free T4 (FT4), and free T3 (FT3), using an enzyme-linked immunoassay. Descriptive analysis was used to investigate PFC exposure and the toxic equivalent quantity (TEQ) was used to calculate the transthyretin (TTR)-disrupting toxicity of combined exposure to PFAS. Linear additive and multiple linear regression models were used to explore the relationship between PFAS and hormones, using PFC concentration as quartiles and continuous variables. Among the PFAS identified, 12 PFASs had detection rates >80 %, with perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) having the highest concentrations. Perfluorodecanoic acid (PFDA), PFOS, and perfluorononanoic acid (PFNA) were negatively correlated with TSH levels and each interquartile range (IQR) concentration increase caused a reduction in TSH levels by 2.14 %, 1.78 %, and 3.04 %, respectively. Perfluorotridecanoic acid (PFTrA) and perfluoropentanoic acid (PFPA) were positively correlated with T4 and T3 levels, respectively, and levels increased by 4.52 % (T4) and 1.14 % (T3) with IQR concentration increase. Perfluorobutanoic acid (PFBA) was negatively correlated with FT4 levels, which decreased by 1.89 % with IQR concentration increase. A negative correlation was found between the combined exposure indices of TEQ and TSH levels; IQR increase in TEQ decreased the TSH concentration by 1.91 %. In conclusion, exposure to PFAS was common in the elderly population and was associated with decreased TSH and FT4 levels and increased T4 and T3 levels. These results indicated that PFASs may cause thyroid-disrupting effects in the elderly population.

Diabetes Mellitus Mediates the Relationship Between Atherogenic Index of Plasma and Gallstones: A Population-Based Cross-Sectional Study.

Purpose: Previous studies have shown a correlation between diabetes mellitus and gallstone formation. The atherogenic index of plasma (AIP) is associated with many metabolic diseases. However, insufficient evidence still exists to elucidate the association between AIP and gallstones. The primary objective of this study was to investigate the correlation between AIP and gallstones in US adults, and the secondary objective was to analyze whether diabetes plays a mediating role in the association. Patients and Methods: Using data from the National Health and Nutrition Survey (NHANES) conducted between 2017 and March 2020, this study investigated the association between AIP and gallstone incidence in US adults. A variety of statistical methods were used to analyze the data in this study, including multivariate logistic regression, subgroup analyses, restricted cubic spline curves (RCS), and mediation effects analysis. In addition, two-stage linear regression was used to detect possible threshold and saturation effects. Results: A total of 6952 subjects were enrolled in the trial, of which 748 patients were diagnosed with gallstones. A significant positive association between AIP and gallstones was observed by fully adjusted multivariate logistic regression analysis, with an odds ratio (OR) of 1.45 and a 95% confidence interval (CI) of (1.09, 1.93). In addition, a non-linear positive association and saturation effect between AIP and gallstones were found, with an inflection point of 0.2246. Mediation analysis showed that diabetes had a mediating effect of 16.9% in the association between AIP and gallstones. Conclusion: This study suggests that elevated levels of AIP are linked to an augmented vulnerability to gallstone development, with diabetes serving as a mediating factor. These findings present a novel perspective on clinical approaches to prevent and manage gallstones.

Causal relationship between aspirin consumption and heart failure: a Mendelian randomization study.

AIMS: This study aimed to investigate the causal association of aspirin consumption with the risk of heart failure. METHODS: Our study included a total of 218 208 individuals, with 23 397 cases of heart failure. Genetic summary data on the association between single-nucleotide polymorphisms (SNPs) and aspirin consumption were obtained from a large-scale genome-wide association study involving 462 933 individuals, of which 61 702 people were taking aspirin. After the exclusion of critical confounding factors, we assessed the final and independent association between the aspirin consumption and the risk of heart failure using 3 two-sample Mendelian randomization (MR) methods-inverse variance weighted (IVW), weighted-median, and MR-Egger regression. Sensitivity analyses and directionality test were employed to further validate the stability of the results. RESULTS: After excluding the SNPs that exhibited associations with potential confounders and harmonizing the data, a total of 32 SNPs were finally selected for MR analysis from the initially identified 60 SNPs that displayed strong associations with the exposure. The results of the main method (IVW) showed a significant positive association between aspirin use and the occurrence of heart failure (OR [odds ratio]: 1.085; 95% CI [confidence interval]: 1.015-1.161; P = 0.017), although other methods did not showed statistically significant results (MR-Egger, OR: 1.211, 95% CI: 0.842-1.21, P = 0.896; weighted-median, OR: 1.087, 95% CI: 0.983-1.202, P = 0.105). Heterogeneity test, the MR-Egger intercept, and the funnel plot did not reveal any evidence of heterogeneity (Cochran's Q statistic = 29.263; P = 0.556) or horizontal pleiotropy (intercept = 0.007; P = 0.319). The 'leave-one-out' analysis indicated that no individual SNP exerted a dominant influence on the main estimate. Directionality test confirmed the accuracy of the causal relationship between exposure and outcome direction in our data. CONCLUSIONS: Our results support a potential positive causal relationship between aspirin consumption and the occurrence of heart failure.

Extracellular vesicle-derived silk fibroin nanoparticles loaded with MFGE8 accelerate skin ulcer healing by targeting the vascular endothelial cells.

BACKGROUND: Reduced supplies of oxygen and nutrients caused by vascular injury lead to difficult-to-heal pressure ulcers (PU) in clinical practice. Rapid vascular repair in the skin wound is the key to the resolution of this challenge, but clinical measures are still limited. We described the beneficial effects of extracellular vesicle-derived silk fibroin nanoparticles (NPs) loaded with milk fat globule EGF factor 8 (MFGE8) on accelerating skin blood vessel and PU healing by targeting CD13 in the vascular endothelial cells (VECs). METHODS: CD13, the specific targeting protein of NGR, and MFGE8, an inhibitor of ferroptosis, were detected in VECs and PU tissues. Then, NPs were synthesized via silk fibroin, and MFGE8-coated NPs (NPs@MFGE8) were assembled via loading purified protein MFGE8 produced by Chinese hamster ovary cells. Lentivirus was used to over-express MFGE8 in VECs and obtained MFGE8-engineered extracellular vesicles (EVs-MFGE8) secreted by these VECs. The inhibitory effect of EVs-MFGE8 or NPs@MFGE8 on ferroptosis was detected in vitro. The NGR peptide cross-linked with NPs@MFGE8 was assembled into NGR-NPs@MFGE8. Collagen and silk fibroin were used to synthesize the silk fibroin/collagen hydrogel. After being loaded with NGR-NPs@MFGE8, silk fibroin/collagen hydrogel sustained-release carrier was synthesized to investigate the repair effect on PU in vivo. RESULTS: MFGE8 was decreased, and CD13 was increased in PU tissues. Similar to the effect of EVs-MFGE8 on inhibiting ferroptosis, NPs@MFGE8 could inhibit the mitochondrial autophagy-induced ferroptosis of VECs. Compared with the hydrogels loaded with NPs or NPs@MFGE8, the hydrogels loaded with NGR-NPs@MFGE8 consistently released NGR-NPs@MFGE8 targeting CD13 in VECs, thereby inhibiting mitochondrial autophagy and ferroptosis caused by hypoxia and accelerating wound healing effectively in rats. CONCLUSIONS: The silk fibroin/collagen hydrogel sustained-release carrier loaded with NGR-NPs@MFGE8 was of great significance to use as a wound dressing to inhibit the ferroptosis of VECs by targeting CD13 in PU tissues, preventing PU formation and promoting wound healing.

Hunter: Exploring High-Order Consistency for Point Cloud Registration With Severe Outliers.

After decades of investigation, point cloud registration is still a challenging task in practice, especially when the correspondences are contaminated by a large number of outliers. It may result in a rapidly decreasing probability of generating a hypothesis close to the true transformation, leading to the failure of point cloud registration. To tackle this problem, we propose a transformation estimation method, named Hunter, for robust point cloud registration with severe outliers. The core of Hunter is to design a global-to-local exploration scheme to robustly find the correct correspondences. The global exploration aims to exploit guided sampling to generate promising initial alignments. To this end, a hypergraph-based consistency reasoning module is introduced to learn the high-order consistency among correct correspondences, which is able to yield a more distinct inlier cluster that facilitates the generation of all-inlier hypotheses. Moreover, we propose a preference-based local exploration module that exploits the preference information of top- k promising hypotheses to find a better transformation. This module can efficiently obtain multiple reliable transformation hypotheses by using a multi-initialization searching strategy. Finally, we present a distance-angle based hypothesis selection criterion to choose the most reliable transformation, which can avoid selecting symmetrically aligned false transformations. Experimental results on simulated, indoor, and outdoor datasets, demonstrate that Hunter can achieve significant superiority over the state-of-the-art methods, including both learning-based and traditional methods (as shown in Fig. 1). Moreover, experimental results also indicate that Hunter can achieve more stable performance compared with all other methods with severe outliers.

Identification of high-risk factors for recurrence of colon cancer following complete mesocolic excision: An 8-year retrospective study.

BACKGROUND: Colon cancer recurrence is a common adverse outcome for patients after complete mesocolic excision (CME) and greatly affects the near-term and long-term prognosis of patients. This study aimed to develop a machine learning model that can identify high-risk factors before, during, and after surgery, and predict the occurrence of postoperative colon cancer recurrence. METHODS: The study included 1187 patients with colon cancer, including 110 patients who had recurrent colon cancer. The researchers collected 44 characteristic variables, including patient demographic characteristics, basic medical history, preoperative examination information, type of surgery, and intraoperative information. Four machine learning algorithms, namely extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM), and k-nearest neighbor algorithm (KNN), were used to construct the model. The researchers evaluated the model using the k-fold cross-validation method, ROC curve, calibration curve, decision curve analysis (DCA), and external validation. RESULTS: Among the four prediction models, the XGBoost algorithm performed the best. The ROC curve results showed that the AUC value of XGBoost was 0.962 in the training set and 0.952 in the validation set, indicating high prediction accuracy. The XGBoost model was stable during internal validation using the k-fold cross-validation method. The calibration curve demonstrated high predictive ability of the XGBoost model. The DCA curve showed that patients who received interventional treatment had a higher benefit rate under the XGBoost model. The external validation set's AUC value was 0.91, indicating good extrapolation of the XGBoost prediction model. CONCLUSION: The XGBoost machine learning algorithm-based prediction model for colon cancer recurrence has high prediction accuracy and clinical utility.

Distributions of total mercury and methylmercury and regulating factors in lake water and surface sediment in the cold-arid Wuliangsuhai Lake region.

This study aimed to understand the occurrence of mercury in the water environment of typical cold and arid lakes and the regulating environmental factors. Water and surface sediment samples were collected from July to August, 2022 in the Wuliangsuhai Lake region for the analysis of total mercury (THg) and total methylmercury (TMeHg). Lake water THg and TMeHg ranged between 19.20 ~ 668.10 and 0.10 ~ 11.40 ng/L, respectively, exceeding China's environmental quality standards and contents of other lakes and reservoirs in China and other areas. Surface sediments showed lower mean THg and TMeHg of 261.85 and 0.18 µg/kg, respectively, with the former significantly exceeding the background value of Inner Mongolia and unpolluted natural lakes but lower than those of lakes affected by human factors, such as aquaculture. Sediments showed relatively low methylation and TMeHg (0.01-0.21%) concentrations. Correlation analysis identified salinity, total dissolved solids, conductivity, and redox potential as important factors affecting mercury speciation in water, whereas those in surface sediments were organic matter, pH, and total iron content. This study conducted preliminary research on the different species of Hg in Wuliangsuhai Lake water environment, which can provide scientific evidence for the specific treatment of Hg pollution in agriculture, or industry and other related fields. Our results suggest that upstream and downstream regulatory agencies should strengthen the regulation of agricultural and industrial production, moderately reduce human activities, and reduce the use of mercury-rich substances such as pesticides.

Effects of CpG sites methylation modification of HPV16 integration essential gene on the proliferation of cervical cancer cells

Purpose The mechanism of methylation of HPV CpG sites in the occurrence and prognosis of cervical carcinogenesis remains unclear. We investigated the effects of demethylation of the CpG sites of E2 and E6, essential genes of HPV16 integration, on cervical cancer cell expression, integration, and proliferation. Materials and Methods HPV16-positive (Caski) cells were treated with different concentrations of the demethylation compound 5-aza-dc (0, 5, 10, 20 μmol/l) in vitro. After the intervention, the methylation statuses of HPV16 E2 and E6 were detected by TBS, the expression levels of E2 and E6 mRNA and protein were detected by real-time PCR and western blot, cell proliferation activity was detected by CCK8, and cell cycle and apoptosis were determined by FCM. GraphPad Prism version 8.4.2 and R version 4.2.3 were used for relevant data analyses. Results The methylation levels of HPV16 E2 and E6 CpG sites decreased gradually with increasing 5-aza-dc intervention concentrations. With decreasing E2 and E6 methylation rates, E2 expression increased, the E2/E6 ratio increased, E6 expression decreased, and the growth inhibition rate of Caski cells increased. E2 and E6 expression were negatively and positively correlated with their degrees of methylation respectively, while the E2/E6 mRNA to protein ratio was negatively correlated with the methylation degrees of E2 and E6. Conclusion Demethylation can be used as a prospective treatment to affect HPV expression and persistent infection, providing a new theoretical basis for the clinical treatment of viral infections (AU)

Association between CZT­SPECT myocardial blood flow and coronary stenosis: A cross­sectional study.

The association between the quantitative and semi-quantitative parameters of myocardial blood flow obtained using cadmium-zinc-telluride single photon emission computed tomography (CZT-SPECT) and coronary stenosis remains unclear. Therefore, the objective of the present study was to evaluate the diagnostic value of two parameters obtained using CZT-SPECT in patients with suspected or known coronary artery disease. A total of 24 consecutive patients who underwent CZT-SPECT and coronary angiography within 3 months of each other were included in the study. To evaluate the predictive ability of the regional difference score (DS), coronary flow reserve (CFR), and the combination thereof for positive coronary stenosis at the vascular level, receiver operating characteristic (ROC) curves were plotted and the area under the curves (AUCs) were calculated. Comparisons of the reclassification ability for coronary stenosis between different parameters were assessed by calculating the net reclassification index (NRI) and the integrated discrimination improvement (IDI). The 24 participants (median age: 65 years; range: 46-79 years; 79.2% male) included in this study had a total of 72 major coronary arteries. When stenosis ≥50% was defined as the criteria for positive coronary stenosis, the AUCs and the 95% confidence interval (CI) for regional DS, CFR, and the combination of the two indices were 0.653 (CI, 0.541-0.766), 0.731 (CI, 0.610-0.852) and 0.757 (CI, 0.645-0.869), respectively. Compared with single DS, the combination of DS and CFR increased the predictive ability for positive stenosis, with an NRI of 0.197-1.060 (P<0.01) and an IDI of 0.0150-0.1391 (P<0.05). When stenosis ≥75% was considered as the criteria, the AUCs were 0.760 (CI, 0.614-0.906), 0.703 (CI, 0.550-0.855), and 0.811 (CI, 0.676-0.947), respectively. Compared with DS, CFR had an IDI of -0.3392 to -02860 (P<0.05) and the combination of DS and CFR also enhanced the predictive ability, with an NRI of 0.0313-1.0758 (P<0.01). In conclusion, both regional DS and CFR had diagnostic values for coronary stenosis, but the diagnostic abilities differed in distinguishing between different degrees of stenosis, and the efficiency was improved with a combination of DS and CFR.

Identification of high-risk factors associated with mortality at 1-, 3-, and 5-year intervals in gastric cancer patients undergoing radical surgery and immunotherapy: an 8-year multicenter retrospective analysis.

Background: Combining immunotherapy with surgical intervention is a prevailing and radical therapeutic strategy for individuals afflicted with gastric carcinoma; nonetheless, certain patients exhibit unfavorable prognoses even subsequent to this treatment regimen. This research endeavors to devise a machine learning algorithm to recognize risk factors with a high probability of inducing mortality among patients diagnosed with gastric cancer, both prior to and during their course of treatment. Methods: Within the purview of this investigation, a cohort of 1015 individuals with gastric cancer were incorporated, and 39 variables encompassing diverse features were recorded. To construct the models, we employed three distinct machine learning algorithms, specifically extreme gradient boosting (XGBoost), random forest (RF), and k-nearest neighbor algorithm (KNN). The models were subjected to internal validation through employment of the k-fold cross-validation technique, and subsequently, an external dataset was utilized to externally validate the models. Results: In comparison to other machine learning algorithms employed, the XGBoost algorithm demonstrated superior predictive capacity regarding the risk factors that affect mortality after combination therapy in gastric cancer patients for a duration of one year, three years, and five years posttreatment. The common risk factors that significantly impacted patient survival during the aforementioned time intervals were identified as advanced age, tumor invasion, tumor lymph node metastasis, tumor peripheral nerve invasion (PNI), multiple tumors, tumor size, carcinoembryonic antigen (CEA) level, carbohydrate antigen 125 (CA125) level, carbohydrate antigen 72-4 (CA72-4) level, and H. pylori infection. Conclusion: The XGBoost algorithm can assist clinicians in identifying pivotal prognostic factors that are of clinical significance and can contribute toward individualized patient monitoring and management.

The gut microbiome of Helicoverpa armigera enhances immune response to baculovirus infection via suppression of Duox-mediated reactive oxygen species.

BACKGROUND: Baculoviruses such as Helicoverpa armigera nucleopolyhedrovirus (HearNPV) infect their lepidopteran hosts via the larval midgut where they interact with host immune responses and gut microbiota. Here we demonstrate that gut microbiota proliferating in response to HearNPV infection promote larval immune responses which impede the infection process. RESULTS: The microbial load of the larval midgut increased following HearNPV infection, due primarily to increases in Enterococcus spp., whereas most other bacterial genera declined, with Firmicutes replacing Proteobacteria as the dominant phylum. Injection of abdominal prolegs of infected larvae with H2 O2 promoted viral infection, diminished microbial abundance, and accelerated larval death, mimicking the effects of HearNPV infection, which up-regulated dual oxidase (Duox) expression, increasing H2 O2 levels in the midgut. Knockdown of Duox with RNAi reduced H2 O2 production in the guts of infected larvae, increased bacterial loads, decreased viral replication, and improved larval survival. Germ-free larvae were more susceptible to HearNPV than control larvae, exhibiting greater expression of Duox, higher levels of H2 O2 , and lower survival. Replenishment of gut bacteria in germ-free larvae restored the base-line immunity to HearNPV observed in normal larvae. Enterococcus spp., Levilactobacillus brevis, and Lactobacillus sp. bacteria were isolated and implicated in immunity restoration via replenishment in germ-free larvae. CONCLUSION: These findings illuminate how gut microbiota play important roles in larval defense against oral baculovirus infection, and suggest novel avenues of investigation to enhance the efficacy of baculoviruses and improve control of lepidopteran pests. © 2023 Society of Chemical Industry.

Ten-Year Multicenter Retrospective Study Utilizing Machine Learning Algorithms to Identify Patients at High Risk of Venous Thromboembolism After Radical Gastrectomy.

Purpose: This study aims to construct a machine learning model that can recognize preoperative, intraoperative, and postoperative high-risk indicators and predict the onset of venous thromboembolism (VTE) in patients. Patients and Methods: A total of 1239 patients diagnosed with gastric cancer were enrolled in this retrospective study, among whom 107 patients developed VTE after surgery. We collected 42 characteristic variables of gastric cancer patients from the database of Wuxi People's Hospital and Wuxi Second People's Hospital between 2010 and 2020, including patients' demographic characteristics, chronic medical history, laboratory test characteristics, surgical information, and patients' postoperative conditions. Four machine learning algorithms, namely, extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN), were employed to develop predictive models. We also utilized Shapley additive explanation (SHAP) for model interpretation and evaluated the models using k-fold cross-validation, receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), and external validation metrics. Results: The XGBoost algorithm demonstrated superior performance compared to the other three prediction models. The area under the curve (AUC) value for XGBoost was 0.989 in the training set and 0.912 in the validation set, indicating high prediction accuracy. Furthermore, the AUC value of the external validation set was 0.85, signifying good extrapolation of the XGBoost prediction model. The results of SHAP analysis revealed that several factors, including higher body mass index (BMI), history of adjuvant radiotherapy and chemotherapy, T-stage of the tumor, lymph node metastasis, central venous catheter use, high intraoperative bleeding, and long operative time, were significantly associated with postoperative VTE. Conclusion: The machine learning algorithm XGBoost derived from this study enables the development of a predictive model for postoperative VTE in patients after radical gastrectomy, thereby assisting clinicians in making informed clinical decisions.

Combined serum CXCL8, CXCL9 and CXCL13 tests for the prediction of microvascular invasion in hepatocellular carcinoma.

Aim: This work is to explore the predictive and diagnostic value of chemokine C-X-C motif ligand 8 (CXCL8), CXCL9 and CXCL13 combined detections for microvascular invasion (MVI) in hepatocellular carcinoma (HCC) patients. Materials & methods: A total of 82 HCC patients with MVI were recruited as the MVI group and 154 patients with non MVI were recruited as the non MVI group. Results: In HCC patients with MVI, CXCL8, CXCL9, CXCL13 levels were significantly elevated. Child-Pugh scores and serum α-fetoprotein level had positive correlation with CXCL8, CXCL9 and CXCL13 levels. The serum levels of CXCL8, 9 and 13 were effective in predicting MVI in HCC patients. Conclusion: CXCL8, CXCL9 and CXCL13 levels in HCC patients are valuable parameters in the prediction of MVI.

Identification of High-Risk Patients for Postoperative Myocardial Injury After CME Using Machine Learning: A 10-Year Multicenter Retrospective Study.

Purpose: The occurrence of myocardial injury, a grave complication post complete mesocolic excision (CME), profoundly impacts the immediate and long-term prognosis of patients. The aim of this inquiry was to conceive a machine learning model that can recognize preoperative, intraoperative and postoperative high-risk factors and predict the onset of myocardial injury following CME. Patients and Methods: This study included 1198 colon cancer patients, 133 of whom experienced myocardial injury after surgery. Thirty-six distinct variables were gathered, encompassing patient demographics, medical history, preoperative examination characteristics, surgery type, and intraoperative details. Four machine learning algorithms, namely, extreme gradient boosting (XGBoost), random forest (RF), multilayer perceptron (MLP), and k-nearest neighbor algorithm (KNN), were employed to fabricate the model, and k-fold cross-validation, ROC curve, calibration curve, decision curve analysis (DCA), and external validation were employed to evaluate it. Results: Out of the four predictive models employed, the XGBoost algorithm demonstrated the best performance. The ROC curve findings indicated that the XGBoost model exhibited remarkable predictive accuracy, with an area under the curve (AUC) value of 0.997 in the training set and 0.956 in the validation set. For internal validation, the k-fold cross-validation method was utilized, and the XGBoost model was shown to be steady. Furthermore, the calibration curves demonstrated the XGBoost model's high predictive capability. The DCA curve revealed higher benefit rates for patients who underwent interventional treatment under the XGBoost model. The AUC value for the external validation set was 0.74, which indicated that the XGBoost prediction model possessed good extrapolative capacity. Conclusion: The myocardial injury prediction model for patients undergoing CME that was developed using the XGBoost machine learning algorithm in this study demonstrates both high predictive accuracy and clinical utility.

An Explainable Student Fatigue Monitoring Module with Joint Facial Representation.

Online fatigue estimation is, inevitably, in demand as fatigue can impair the health of college students and lower the quality of higher education. Therefore, it is essential to monitor college students' fatigue to diminish its adverse effects on the health and academic performance of college students. However, former studies on student fatigue monitoring are mainly survey-based with offline analysis, instead of using constant fatigue monitoring. Hence, we proposed an explainable student fatigue estimation model based on joint facial representation. This model includes two modules: a spacial-temporal symptom classification module and a data-experience joint status inferring module. The first module tracks a student's face and generates spatial-temporal features using a deep convolutional neural network (CNN) for the relevant drivers of abnormal symptom classification; the second module infers a student's status with symptom classification results with maximum a posteriori (MAP) under the data-experience joint constraints. The model was trained on the benchmark NTHU Driver Drowsiness Detection (NTHU-DDD) dataset and tested on an Online Student Fatigue Monitoring (OSFM) dataset. Our method outperformed the other methods with an accuracy rate of 94.47% under the same training-testing setting. The results were significant for real-time monitoring of students' fatigue states during online classes and could also provide practical strategies for in-person education.