Development and external validation of a nomogram for predicting lymph node metastasis in 1-3 cm lung adenocarcinoma.

Aim: This study aimed to investigate the risk factors for lymph node metastasis in 1-3 cm adenocarcinoma and develop a new nomogram to predict the probability of lymph node metastasis.Materials & methods: This study collected clinical data from 1656 patients for risk factor analysis and an additional 500 patients for external validation. The logistic regression analyses were employed for risk factor analysis. The least absolute shrinkage and selection operator regression was used to select variables, and important variables were used to construct the nomogram and an online calculator.Results: The nomogram for predicting lymph node metastasis comprises six variables: tumor size (mediastinal window), consolidation tumor ratio, tumor location, lymphadenopathy, preoperative serum carcinoembryonic antigen level and pathological grade. According to the predicted results, the risk of lymph node metastasis was divided into low-risk group and high-risk group. We confirmed the exceptional clinical efficacy of the model through multiple evaluation methods.Conclusion: The importance of intraoperative frozen section is increasing. We discussed the risk factors for lymph node metastasis and developed a nomogram to predict the probability of lymph node metastasis in 1-3 cm adenocarcinomas, which can guide lymph node resection strategies during surgery.

[Box: see text].

Optimization of drug solubility inside the supercritical CO2 system via numerical simulation based on artificial intelligence approach.

In this research paper, we explored the predictive capabilities of three different models of Polynomial Regression (PR), Extreme Gradient Boosting (XGB), and LASSO to estimate the density of supercritical carbon dioxide (SC-CO2) and the solubility of niflumic acid as functions of the input variables of temperature and pressure. The optimization of hyperparameters for these models is achieved using the innovative Barnacles Mating Optimizer (BMO) algorithm. For SC-CO2 density estimation, PR exhibits remarkable accuracy, showing an R-squared value of 0.99207 for data fitting. XGB performs admirably with an R2 of 0.92673, while LASSO model demonstrates good predictive ability, showing an R2 of 0.81917. Furthermore, we assess the models' performance in predicting the solubility of niflumic acid. PR exhibits excellent predictive capabilities with an R2 of 0.96949. XGB also delivers strong performance, yielding an R-squared score of 0.92961. LASSO performs well, achieving an R-squared score of 0.82094. The results indicated promising performance of machine learning models and optimizer in estimating drug solubility in supercritical CO2 as the solvent applicable for pharmaceutical industry.

Construction and validation of a risk prediction model for extrauterine growth restriction in preterm infants born at gestational age less than 34 weeks.

Objective: This study aimed to develop and validate a model for predicting extrauterine growth restriction (EUGR) in preterm infants born ≤34 weeks gestation. Methods: Preterm infants from Guangxi Maternal and Child Health Hospital (2019-2021) were randomly divided into training (80%) and testing (20%) sets. Collinear clinical variables were excluded using Pearson correlation coefficients. Predictive factors were identified using Lasso regression. Random forest (RF), support vector machine (SVM), and logistic regression (LR) models were then built and evaluated using the confusion matrix, area under the curve (AUC), and the F1 score. Additionally, calibration curves and decision curve analysis (DCA) were plotted to assess the performance and practical utility of the models. Results: The study included 387 infants, with no significant baseline differences between training (n = 310) and testing (n = 77) sets. LR identified gestational age, birth weight, premature rupture of membranes, patent ductus arteriosus, cholestasis, and neonatal sepsis as key EUGR predictors. The RF model (19 variables) demonstrated an accuracy of greater than 90% during training, and superior AUC (0.62), F1 score (0.80), and accuracy (0.72) in testing compared to other models. Conclusions: Gestational age, birth weight, premature rupture of membranes, patent ductus arteriosus, cholestasis, and neonatal sepsis are significant EUGR predictors in preterm infants ≤34 weeks. The model shows promise for early EUGR prediction in clinical practice, potentially enhancing screening efficiency and accuracy, thus saving medical resources.

The lasso peptide produced by Bacillus licheniformis MCC 2514 demonstrates efficacy in treating in-vivo Salmonella Typhimurium infection.

Salmonella, a significant pathogen transmitted through food, presents a substantial threat to public health. The issue of antibiotic misuse is causing a quest for alternative replacements. An antimicrobial peptide, predicted as lasso peptide 2514 (LP-2514), derived from the probiotic bacteria Bacillus licheniformis MCC 2514, has demonstrated effectiveness against various foodborne pathogens including Salmonella. This study aims to assess the efficacy of this peptide in vivo. Mice infected with Salmonella Typhimurium received daily oral administration of LP-2514 (30 mg/kg/day) for 2 weeks until the symptoms subsided. After the treatment, biochemical and histopathological parameters were examined. LP-2514 treated mice demonstrated reduced infection, as evidenced by a 5-fold decrease in aspartate aminotransferase concentration and a 10-fold decrease in alanine aminotransferase concentration in plasma. Nitric oxide generation was decreased by 61.23 %, C-reactive protein by 75.9 %, and numerous antioxidant enzymes were elevated to suppress the infection. Increased expression of the anti-inflammatory marker Interleukin-10 (IL-10) by 43-fold was observed in treated mice, while untreated mice displayed elevated expression of pro-inflammatory cytokines indicating the severity of infection. Hence, LP-2514 successfully alleviated the disease symptoms caused by S. Typhimurium, thus exhibiting as a potential replacement for antibiotics or food-grade preservatives.

A Flexible Adaptive Lasso Cox Frailty Model Based on the Full Likelihood.

In this work, a method to regularize Cox frailty models is proposed that accommodates time-varying covariates and time-varying coefficients and is based on the full likelihood instead of the partial likelihood. A particular advantage of this framework is that the baseline hazard can be explicitly modeled in a smooth, semiparametric way, for example, via P-splines. Regularization for variable selection is performed via a lasso penalty and via group lasso for categorical variables while a second penalty regularizes wiggliness of smooth estimates of time-varying coefficients and the baseline hazard. Additionally, adaptive weights are included to stabilize the estimation. The method is implemented in the R function coxlasso, which is now integrated into the package PenCoxFrail, and will be compared to other packages for regularized Cox regression.

Development of a predictive model for patients with bone metastases referred to palliative radiotherapy: Secondary analysis of a multicenter study (the PRAIS trial).

BACKGROUND: The decision to administer palliative radiotherapy (RT) to patients with bone metastases (BMs), as well as the selection of treatment protocols (dose, fractionation), requires an accurate assessment of survival expectancy. In this study, we aimed to develop three predictive models (PMs) to estimate short-, intermediate-, and long-term overall survival (OS) for patients in this clinical setting. MATERIALS AND METHODS: This study constitutes a sub-analysis of the PRAIS trial, a longitudinal observational study collecting data from patients referred to participating centers to receive palliative RT for cancer-induced bone pain. Our analysis encompassed 567 patients from the PRAIS trial database. The primary objectives were to ascertain the correlation between clinical and laboratory parameters with the OS rates at three distinct time points (short: 3 weeks; intermediate: 24 weeks; prolonged: 52 weeks) and to construct PMs for prognosis. We employed machine learning techniques, comprising the following steps: (i) identification of reliable prognostic variables and training; (ii) validation and testing of the model using the selected variables. The selection of variables was accomplished using the LASSO method (Least Absolute Shrinkage and Selection Operator). The model performance was assessed using receiver operator characteristic curves (ROC) and the area under the curve (AUC). RESULTS: Our analysis demonstrated a significant impact of clinical parameters (primary tumor site, presence of non-bone metastases, steroids and opioid intake, food intake, and body mass index) and laboratory parameters (interleukin 8 [IL-8], chloride levels, C-reactive protein, white blood cell count, and lymphocyte count) on OS. Notably, different factors were associated with the different times for OS with only IL-8 included both in the PMs for short- and long-term OS. The AUC values for ROC curves for 3-week, 24-week, and 52-week OS were 0.901, 0.767, and 0.806, respectively. CONCLUSIONS: We successfully developed three PMs for OS based on easily accessible clinical and laboratory parameters for patients referred to palliative RT for painful BMs. While our findings are promising, it is important to recognize that this was an exploratory trial. The implementation of these tools into clinical practice warrants further investigation and confirmation through subsequent studies with separate databases.

Development and validation of a predictive model for failure of ureteral access sheath placement in patients with ureteral calculi.

OBJECTIVE: The Ureteral Access Sheath (UAS) has notable benefits but may fail to traverse the ureter in some cases. Our objective was to develop and validate a dynamic online nomogram for patients with ureteral stones who experienced UAS placement failure during retrograde intrarenal surgery (RIRS). METHODS: This study is a retrospective cohort analysis using medical records from the Second Hospital of Tianjin Medical University. We reviewed the records of patients with ureteral stones who underwent RIRS in 2022 to identify risk factors associated with UAS placement failure. Lasso combined logistic regression was utilized to identify independent risk factors associated with unsuccessful UAS placement in individuals with ureteral stones. Subsequently, a nomogram model was developed to predict the likelihood of failed UAS placement in this patient cohort. The model's performance was assessed through Receiver Operating Characteristic Curve (ROC) analysis, calibration curve assessment, and Decision Curve Analysis (DCA). RESULTS: Significant independent risk factors for unsuccessful UAS placement in patients with ureteral stones included age (OR = 0.95, P < 0.001), male gender (OR = 2.15, P = 0.017), body mass index (BMI) (OR = 1.12, P < 0.001), history of stone evacuation (OR = 0.35, P = 0.014), and ureteral stone diameter (OR = 0.23, P < 0.001). A nomogram was constructed based on these variables. Model validation demonstrated an area under the ROC curve of 0.789, indicating good discrimination. The calibration curve exhibited strong agreement, and the decision curve analysis revealed a favorable net clinical benefit for the model. CONCLUSIONS: Young age, male sex, high BMI, no history of stone evacuation, and small diameter of ureteral stones were independent risk factors for failure of UAS placement in patients with ureteral stones, and the dynamic nomogram established with these 5 factors was clinically effective in predicting the outcome of UAS placement.

Methods for multi-omic data integration in cancer research.

Multi-omics data integration is a term that refers to the process of combining and analyzing data from different omic experimental sources, such as genomics, transcriptomics, methylation assays, and microRNA sequencing, among others. Such data integration approaches have the potential to provide a more comprehensive functional understanding of biological systems and has numerous applications in areas such as disease diagnosis, prognosis and therapy. However, quantitative integration of multi-omic data is a complex task that requires the use of highly specialized methods and approaches. Here, we discuss a number of data integration methods that have been developed with multi-omics data in view, including statistical methods, machine learning approaches, and network-based approaches. We also discuss the challenges and limitations of such methods and provide examples of their applications in the literature. Overall, this review aims to provide an overview of the current state of the field and highlight potential directions for future research.

Computed tomography-based radiomics nomogram for prediction of lympho-vascular and perineural invasion in esophageal squamous cell cancer patients: a retrospective cohort study.

PURPOSE: Lympho-vascular invasion (LVI) and perineural invasion (PNI) have been established as prognostic factors in various types of cancers. The preoperative prediction of LVI and PNI has the potential to guide personalized medicine strategies for patients with esophageal squamous cell cancer (ESCC). This study investigates whether radiomics features derived from preoperative contrast-enhanced CT could predict LVI and PNI in ESCC patients. METHODS AND MATERIALS: A retrospective cohort of 544 ESCC patients who underwent esophagectomy were included in this study. Preoperative contrast-enhanced CT images, pathological results of PNI and LVI, and clinical characteristics were collected. For each patient, the gross tumor volume (GTV-T) and lymph nodes volume (GTV-N) were delineated and four categories of radiomics features (first-order, shape, textural and wavelet) were extracted from GTV-T and GTV-N. The Mann-Whitney U test was used to select significant features associated with LVI and PNI in turn. Subsequently, radiomics signatures for LVI and PNI were constructed using LASSO regression with ten-fold cross-validation. Significant clinical characteristics were combined with radiomics signature to develop two nomogram models for predicting LVI and PNI, respectively. The area under the curve (AUC) and calibration curve were used to evaluate the predictive performance of the models. RESULTS: The radiomics signature for LVI prediction consisted of 28 features, while the PNI radiomics signature comprised 14 features. The AUCs of the LVI radiomics signature were 0.77 and 0.74 in the training and validation groups, respectively, while the AUCs of the PNI radiomics signature were 0.69 and 0.68 in the training and validation groups. The nomograms incorporating radiomics signatures and significant clinical characteristics such as age, gender, thrombin time and D-Dimer showed improved predictive performance for both LVI (AUC: 0.82 and 0.80 in the training and validation group) and PNI (AUC: 0.75 and 0.72 in the training and validation groups) compared to the radiomics signature alone. CONCLUSION: The radiomics features extracted from preoperative contrast-enhanced CT of gross tumor and lymph nodes have demonstrated their potential in predicting LVI and PNI in ESCC patients. Furthermore, the incorporation of clinical characteristics has shown additional value, resulting in improved predictive performance.

A data science approach to optimize ADHD assessment with the BRIEF-2 questionnaire.

Attention deficit hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder. A key challenge associated with this condition is achieving an early diagnosis. The current study seeks to anticipate and delineate the assessments offered by both parents and teachers concerning a child's behavior and overall functioning with the Behavior Rating Inventory of Executive Function-2 (BRIEF-2). Mothers, fathers, and teachers of 59 children diagnosed or in the process of being assessed for ADHD participated in this study. The responses provided by 59 mothers, 59 fathers, and 57 teachers to the BRIEF-2 questionnaire were collected. The performance of various feature selection techniques, including Lasso, decision trees, random forest, extreme gradient boosting, and forward stepwise regression, was evaluated. The results indicate that Lasso stands out as the optimal method for our dataset, striking an ideal balance between accuracy and interpretability. A repeated validation analysis reveals an average positive correlation exceeding 0.5 between the inattention/hyperactivity scores reported by informants (mother, father, or teacher) and the predictions derived from Lasso. This performance is achieved using only approximately 18% of the BRIEF-2 items. These findings underscore the usefulness of variable selection techniques in accurately characterizing a patient's condition while employing a small subset of assessment items. This efficiency is particularly valuable in time-constrained settings and contributes to improving the comprehension of ADHD.

Development of a predictive model for the U-shaped relationship between the triglyceride glycemic index and depression using machine learning (NHANES 2009-2018).

Background: At present, the relationship between depression and the triglyceride glycemic (TyG) index remains a topic of debate. This study sought to elucidate the relationship between depression and the TyG index to create a predictive model that would help doctors diagnose patients. Methods: We conducted a cross-sectional study utilizing the National Health and Nutrition Examination Survey (NHANES) dataset, which comprises data from 2009 to 2018. The analysis involved 11,222 adults with a Patient Health Questionnaire-9 (PHQ-9) score of 5 or higher, indicating the presence of depression. As part of the analysis, multiple regression models were used to test whether a linear relationship existed between the TyG index and depression. A threshold effects analysis was used to generate smoothed curves and detect nonlinear correlations. Additionally, the Least Absolute Shrinkage and Selection Operator (LASSO) regression were employed to identify the key risk factors associated with depression. The factors identified were then used to construct the risk prediction nomogram. Finally, Receiver Operating Characteristic (ROC) curves were used to evaluate the discriminative performance of the model. Results: Multivariable linear regression analysis indicated a strong positive correlation between depression and the TyG index (ß: 0.38, 95 % CI: 0.16-0.60, p = 0.0008). A U-shaped relationship with an inflection point was observed at a TyG index of 8.16. The nomogram model, constructed using risk factors identified by LASSO, exhibited a significant predictive value (AUC = 0.888). Conclusions: The results of this investigation point to a U-shaped association between depression risk and the TyG index among Americans. Those with a TyG index of over 8.16 are significantly more likely to develop depression. These results suggest a possible causal relationship and emphasize the importance of monitoring the TyG index in depression risk assessment.

Integrated machine learning and Mendelian randomization reveal PALMD as a prognostic biomarker for nonspecific orbital inflammation.

BACKGROUND: Nonspecific Orbital Inflammation (NSOI) remains a perplexing enigma among proliferative inflammatory disorders. Its etiology is idiopathic, characterized by distinctive and polymorphous lymphoid infiltration within the orbital region. Preliminary investigations suggest that PALMD localizes within the cytosol, potentially playing a crucial role in cellular processes, including plasma membrane dynamics and myogenic differentiation. The potential of PALMD as a biomarker for NSOI warrants meticulous exploration. METHODS: PALMD was identified through the intersection analysis of common DEGs from datasets GSE58331 and GSE105149 from the GEO database, alongside immune-related gene lists from the ImmPort database, using Lasso regression and SVM-RFE analysis. GSEA and GSVA were conducted with gene sets co-expressed with PALMD. To further investigate the correlation between PALMD and immune-related biological processes, the CIBERSORT algorithm and ESTIMATE method were employed to evaluate immune microenvironment characteristics of each sample. The expression levels of PALMD were subsequently validated using GSE105149. RESULTS: Among the 314 DEGs identified, several showed significant differences. Lasso and SVM-RFE algorithms pinpointed 15 hub genes. Functional analysis of PALMD emphasized its involvement in cell-cell adhesion, leukocyte migration, and leukocyte-mediated immunity. Enrichment analysis revealed that gene sets positively correlated with PALMD were enriched in immune-related pathways. Immune infiltration analysis indicated that resting dendritic cells, resting mast cells, activated NK cells, and plasma cells positively associate with PALMD expression. Conversely, naive B cells, activated dendritic cells, M0 and M1 macrophages, activated mast cells, activated CD4 memory T cells, and naive CD4 T cells showed a negative correlation with PALMD expression. PALMD demonstrated significant diagnostic potential in differentiating NSOI. CONCLUSIONS: This study identifies PALMD as a potential biomarker linked to NSOI, providing insights into its pathogenesis and offering new avenues for tracking disease progression.

Bioinformatics-Based Identification of Key Prognostic Genes in Neuroblastoma with a Focus on Immune Cell Infiltration and Diagnostic Potential of VGF.

Objective: This study aims to identify differentially expressed genes (DEGs) in neuroblastoma (NB) through comprehensive bioinformatics analysis and machine learning techniques. We seek to elucidate these DEGs' biological functions and associated signaling pathways. Furthermore, our objective extends to predicting upstream microRNAs (miRNAs) and relevant transcription factors of pivotal genes, with the ultimate goal of guiding clinical diagnostics and informing future treatment strategies for Neuroblastoma. Methods: In this study, we sourced datasets GSE49710 and TARGET from the GEO and UCSC-XENA databases, respectively. Differentially expressed genes (DEGs) were identified using the R language "limma" package. Subsequent Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of these DEGs were conducted using the "clusterProfiler" package. We employed Weighted Gene Co-expression Network Analysis (WGCNA) to isolate the most significant modules associated with death and MYCN amplification, specifically MEpink and MEbrown modules. These modules were then cross-referenced with the DEGs for further GO and KEGG pathway analyses. LASSO regression analysis, facilitated by the "glmnet" package, was utilized to pinpoint three hub genes. We performed differential analysis on these genes and constructed Receiver Operating Characteristic (ROC) curves for disease diagnosis purposes. Immune infiltration analysis was conducted using the "GSVA" package's ssGSEA function. Additionally, single-gene Gene Set Enrichment Analysis (GSEA) on the hub gene was carried out based on Reactome and KEGG databases. Upstream miRNA and transcription factors associated with the hub gene were predicted using RegNetwork, with visual representations created in Cytoscape. Furthermore, to validate the three identified markers in neuroblastoma tissues, quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) analysis was conducted. Results: We identified 483 differentially expressed genes (DEGs) in neuroblastoma. These genes predominantly function in protein translation, membrane composition, and RNA transcription regulation, and are implicated in multiple signaling pathways relevant to neurodegenerative diseases. Utilizing LASSO regression analysis, we pinpointed three hub genes: VGF, DGKD, and C19orf52. The Receiver Operating Characteristic (ROC) curve analysis yielded Area Under Curve (AUC) values of 0.751 and 0.722 for VGF, 0.79 and 0.656 for DGKD, and 0.8 and 0.753 for C19orf52, respectively. Our immune infiltration analysis revealed significant correlations among monocytes, follicular helper T cells, and CD4+ T cells. Notably, in the death group, we observed heightened infiltration levels of activated CD4+ T cells, macrophages, and Th2 cells. C19orf52 exhibited a close association with the infiltration of monocytes, CD4+ T cells, and Th2 cells, with P-values less than 0.05. Furthermore, qRT-PCR analysis corroborated the upregulation of VGF in neuroblastoma tissues, further validating our findings. Conclusion: The hub genes (VGF, DGKD, and C19orf52) of neuroblastoma are screened. VGF, one of the hub genes, may have a high diagnostic value and is involved in the immune cell infiltration in neuroblastoma tissue, which may be used as a biomarker for the diagnosis of neuroblastoma and provides a new direction for clinical prognosis prediction and management improvement.

Risk factors and predictive model for nosocomial infections by extensively drug-resistant Acinetobacter baumannii.

Background: Extensively drug-resistant Acinetobacter baumannii (XDRAB) has become a significant pathogen in hospital environments, particularly in intensive care units (ICUs). XDRAB's resistance to conventional antimicrobial treatments and ability to survive on various surfaces pose a substantial threat to patient health, often resulting in severe infections such as ventilator-associated pneumonia (VAP) and bloodstream infections (BSI). Methods: We retrospectively analyzed clinical data from 559 patients with XDRAB infections admitted to Jinhua Central Hospital between January 2021 and December 2023. Patients were randomly divided into a training set (391 cases) and a testing set (168 cases). Variables were selected using Lasso regression and logistic regression analysis, and a predictive model was constructed and validated internally and externally. Model performance and clinical utility were evaluated using the Hosmer-Lemeshow test, C-index, ROC curve, decision curve analysis (DCA), and clinical impact curve (CIC). Results: Lasso regression analysis was used to screen 35 variables, selecting features through 10-fold cross-validation. We chose lambda.1se=0.03450 (log(lambda.1se)=-3.367), including 10 non-zero coefficient features. These features were then included in a multivariate logistic regression analysis, identifying 8 independent risk factors for XDRAB infection: ICU stay of 1-7 days (OR=3.970, 95%CI=1.586-9.937), ICU stay >7 days (OR=12.316, 95%CI=5.661-26.793), hypoproteinemia (OR=3.249, 95%CI=1.679-6.291), glucocorticoid use (OR=2.371, 95%CI=1.231-4.564), urinary catheterization (OR=2.148, 95%CI=1.120-4.120), mechanical ventilation (OR=2.737, 95%CI=1.367-5.482), diabetes mellitus (OR=2.435, 95%CI=1.050-5.646), carbapenem use (OR=6.649, 95%CI=2.321-19.048), and ß-lactamase inhibitor use (OR=4.146, 95%CI=2.145-8.014). These 8 factors were used to construct a predictive model visualized through a nomogram. The model validation showed a C-index of 0.932 for the training set and 0.929 for the testing set, with a Hosmer-Lemeshow test p-value of 0.47, indicating good calibration. Furthermore, the DCA curve demonstrated good clinical decision-making performance, and the CIC curve confirmed the model's reliable clinical impact. Conclusion: Regression analysis identified ICU stay duration, hypoproteinemia, glucocorticoid use, urinary catheterization, mechanical ventilation, diabetes mellitus, carbapenem use, and ß-lactamase inhibitor use as independent risk factors for XDRAB infection. The corresponding predictive model demonstrated high accuracy and stability.

A Broad Spectrum Lasso Peptide Antibiotic Targeting the Bacterial Ribosome.

Lasso peptides, biologically active molecules with a distinct structurally constrained knotted fold, are natural products belonging to the class of ribosomally-synthesized and posttranslationally modified peptides (RiPPs). Lasso peptides act upon several bacterial targets, but none have been reported to inhibit the ribosome, one of the main antibiotic targets in the bacterial cell. Here, we report the identification and characterization of the lasso peptide antibiotic, lariocidin (LAR), and its internally cyclized derivative, lariocidin B (LAR-B), produced by Paenabacillussp. M2, with broad-spectrum activity against many bacterial pathogens. We show that lariocidins inhibit bacterial growth by binding to the ribosome and interfering with protein synthesis. Structural, genetic, and biochemical data show that lariocidins bind at a unique site in the small ribosomal subunit, where they interact with the 16S rRNA and aminoacyl-tRNA, inhibiting translocation and inducing miscoding. LAR is unaffected by common resistance mechanisms, has a low propensity for generating spontaneous resistance, shows no human cell toxicity, and has potent in vivo activity in a mouse model of Acinetobacter baumannii infection. Our finding of the first ribosome-targeting lasso peptides uncovers new routes toward discovering alternative protein synthesis inhibitors and offers a new chemical scaffold for developing much-needed antibacterial drugs.

Establishment and validation of apnea risk prediction models in preterm infants: a retrospective case control study.

BACKGROUND: Apnea is common in preterm infants and can be accompanied with severe hypoxic damage. Early assessment of apnea risk can impact the prognosis of preterm infants. We constructed a prediction model to assess apnea risk in premature infants for identifying high-risk groups. METHODS: A total of 162 and 324 preterm infants with and without apnea who were admitted to the neonatal intensive care unit of Xiamen University between January 2018 and December 2021 were selected as the case and control groups, respectively. Demographic characteristics, laboratory indicators, complications of the patients, pregnancy-related factors, and perinatal risk factors of the mother were collected retrospectively. The participants were randomly divided into modeling (n = 388) and validation (n = 98) sets in an 8:2 ratio. Least Absolute Shrinkage and Selection Operator (LASSO) and multivariate logistic regression analyses were used to independently filter variables from the modeling set and build a model. A nomogram was used to visualize models. The calibration and clinical utility of the model was evaluated using consistency index, receiver operating characteristic (ROC) curve, calibration curve, and decision curve, and the model was verified using the validation set. RESULTS: Results of LASSO combined with multivariate logistic regression analysis showed that gestational age at birth, birth length, Apgar score, and neonatal respiratory distress syndrome were predictors of apnea development in preterm infants. The model was presented as a nomogram and the Hosmer-Lemeshow goodness of fit test showed a good model fit (χ2=5.192, df=8, P=0.737), with Nagelkerke R2 of 0.410 and C-index of 0.831. The area under the ROC curve and 95% CI were 0.831 (0.787-0.874) and 0.829 (0.722-0.935), respectively. Delong's test comparing the AUC of the two data sets showed no significant difference (P=0.976). The calibration curve showed good agreement between the predicted and actual observations. The decision curve results showed that the threshold probability range of the model was 0.07-1.00, the net benefit was high, and the constructed clinical prediction model had clinical utility. CONCLUSIONS: Our risk prediction model based on gestational age, birth length, Apgar score 10 min post-birth, and neonatal respiratory distress syndrome was validated in many aspects and had good predictive efficacy and clinical utility.

Development and validation of a dynamic nomogram for high care dependency during the hospital-family transition periods in older stroke patients.

BACKGROUND: This research aimed to develop and validate a dynamic nomogram for predicting the risk of high care dependency during the hospital-family transition periods in older stroke patients. METHODS: 309 older stroke patients in the hospital-family transition periods who were treated in the Department of Neurology outpatient clinics of three general hospitals in Jinzhou, Liaoning Province from June to December 2023 were selected as the training set. The patients were investigated with the General Patient Information Questionnaire, the Care Dependency Scale (CDS), the Tilburg Frailty Inventory (TFI), the Hamilton Anxiety Rating Scale (HAMA), the Hamilton Depression Rating Scale-17 (HAMD-17), and the Mini Nutrition Assessment Short Form (MNA-SF). Lasso-logistic regression analysis was used to screen the risk factors for high care dependency in older stroke patients during the hospital-family transition period, and a dynamic nomogram model was constructed. The model was uploaded in the form of a web page based on Shiny apps. The Bootstrap method was employed to repeat the process 1000 times for internal validation. The model's predictive efficacy was assessed using the calibration plot, decision curve analysis curve (DCA), and area under the curve (AUC) of the receiver operator characteristic (ROC) curve. A total of 133 older stroke patients during the hospital-family transition periods who visited the outpatient department of Neurology of three general hospitals in Jinzhou from January to March 2024 were selected as the validation set for external validation of the model. RESULTS: Based on the history of stroke, chronic disease, falls in the past 6 months, depression, malnutrition, and frailty, build a dynamic nomogram. The AUC under the ROC curves of the training set was 0.830 (95% CI: 0.784-0.875), and that of the validation set was 0.833 (95% CI: 0.766-0.900). The calibration curve was close to the ideal curve, and DCA results confirmed that the nomogram performed well in terms of clinical applicability. CONCLUSION: The online dynamic nomogram constructed in this study has good specificity, sensitivity, and clinical practicability, which can be applied to senior stroke patients as a prediction and assessment tool for high care dependency. It is of great significance to guide the development of early intervention strategies, optimize resource allocation, and reduce the care burden on families and society.

KLF15 Transcription Activated Tachykinin Precursor 1 to Induce Perineural Invasion in Head and Neck Squamous Cell Carcinoma.

OBJECTIVE: In this study, we aimed to identify novel biomarkers related to Peripheral Neural Invasion (PNI) in head and neck squamous cell carcinoma (HNSCC). METHODS: The PNI-related differentially expressed mRNAs (DE-mRNAs) in HNSCC were identified to construct a PNI-related risk score model. The expression level and ROC curve for Tachykinin Precursor 1 (TAC1) were calculated. Additionally, two kinds of in vitro models of PNI were established for investigation, including the Matrigel-PNI model and the Transwell-PNI model. Furthermore, the transcription factor of the TAC1 was predicted and verified by qRTPCR. RESULTS: A total of 139 DE-mRNAs were identified in PNI positive and negative groups of HNSCC patients. The risk-score marker model incorporating 20 PNI-related DE-mRNAs was established. The TAC1 was identified as a potential highly expressed PNI marker, which exhibited good performance in predicting PNI events. Patients with higher TAC1 expressions demonstrated significantly shorter survival rates compared to those with lower TAC1 expressions in HNSCC. Besides, the knockdown of TAC1 significantly repressed neural invasion in HNSCC cells in vitro, according to the Matrigel-PNI model and Transwell-PNI model. Furthermore, KLF15 was predicted and verified as a transcription activator of TAC1 in HNSCC. CONCLUSION: This study highlights that the activation of KLF15 transcription of TAC1 promotes PNI in HNSCC cells, which provides guidance regarding the molecular diagnosis of PNI in HNSCC cells.

A Machine Learning Model for Predicting the Propagation Rate Coefficient in Free-Radical Polymerization.

The propagation rate coefficient (kp) is one of the most crucial kinetic parameters in free-radical polymerization (FRP) as it directly governs the rate of polymerization and the resulting molecular weight distribution. The kp in FRP can typically be obtained through experimental measurements or quantum chemical calculations, both of which can be time consuming and resource intensive. Herein, we developed a machine learning model based solely on the structural features of monomers involved in FRP, utilizing molecular embedding and a Lasso regression algorithm to predict kp more efficiently and accurately. The result shows that the model achieves a mean absolute percentage error (MAPE) of only 5.49% in the predictions for four new monomers, which indicates that the model exhibits strong generalization capabilities and provides reliable and robust predictions. In addition, this model can accurately predict the influence of the ester side chain length of (meth)acrylates on kp, aligning well with established scientific knowledge. This approach offers a straightforward and practical model for other researchers to rapidly obtain accurate kp values by employing monomer structural information. The model is sufficiently general to apply to a wide range of (meth)acrylate and butadiene FRP monomers, thereby supporting kinetic modeling of polymerization reactions.

A comprehensive prediction model of drug-refractory epilepsy based on combined clinical-EEG microstate features.

Background: Epilepsy is a chronic neurological disorder characterized by recurrent seizures that significantly impact patients' quality of life. Identifying predictors is crucial for early intervention. Objective: Electroencephalography (EEG) microstates effectively describe the resting state activity of the human brain using multichannel EEG. This study aims to develop a comprehensive prediction model that integrates clinical features with EEG microstates to predict drug-refractory epilepsy (DRE). Design: Retrospective study. Methods: This study encompassed 226 patients with epilepsy treated at the epilepsy center of a tertiary hospital between October 2020 and May 2023. Patients were categorized into DRE and non-DRE groups. All patients were randomly divided into training and testing sets. Lasso regression combined with Stepglm [both] algorithms was used to screen independent risk factors for DRE. These risk factors were used to construct models to predict the DRE. Three models were constructed: a clinical feature model, an EEG microstate model, and a comprehensive prediction model (combining clinical-EEG microstates). A series of evaluation methods was used to validate the accuracy and reliability of the prediction models. Finally, these models were visualized for display. Results: In the training and testing sets, the comprehensive prediction model achieved the highest area under the curve values, registering 0.99 and 0.969, respectively. It was significantly superior to other models in terms of the C-index, with scores of 0.990 and 0.969, respectively. Additionally, the model recorded the lowest Brier scores of 0.034 and 0.071, respectively, and the calibration curve demonstrated good consistency between the predicted probabilities and observed outcomes. Decision curve analysis revealed that the model provided significant clinical net benefit across the threshold range, underscoring its strong clinical applicability. We visualized the comprehensive prediction model by developing a nomogram and established a user-friendly website to enable easy application of this model (https://fydxh.shinyapps.io/CE_model_of_DRE/). Conclusion: A comprehensive prediction model for DRE was developed, showing excellent discrimination and calibration in both the training and testing sets. This model provided an intuitive approach for assessing the risk of developing DRE in patients with epilepsy.