Area Under The ROC Curve Research Articles

Unveiling the shared genes between systemic sclerosis and lung cancer

The risk of lung cancer is significantly increased in patients with systemic sclerosis (SSc), yet the specific genes underlying this association remain unexplored. Our study aims to identify genes shared by SSc and lung cancer. We identified differentially expressed genes (DEGs) from SSc and lung adenocarcinoma (LUAD) datasets (SSc: GSE95065, LUAD: GSE136043) in the GEO database. We found shared genes by intersecting top genes in protein–protein interaction networks by the STRING database. The area under the ROC curve (AUC) was calculated for each shared gene in validation datasets (SSc: GSE231692; LUAD: GSE43458), identifying PRKG2 as the core shared gene. We used the UALCAN platform to assess PRKG2 expression in LUAD patients at various stages and lymph node metastasis states, and compared disease-free survival (DFS) between low and high PRKG2 expression LUAD groups. PRKG2 was overexpressed in A549 cells to study its impact on lung cancer cell proliferation and invasion in vitro. We identified seven shared genes (SCN7A, AGTR1, WIF1, PRKG2, LTF, AQP4, COL10A1), with the AUC for PRKG2 exceeding 0.93 in both diseases (SSc AUC = 0.973; LUAD AUC = 0.939). The PRKG2 expression levels of LUAD patients with different clinical stages and lymph node metastasis states were consistently lower than those observed in normal individuals. The DFS of LUAD patients in the high PRKG2 expression group was higher than that in the low expression group (p = 0.028). In vitro experiments confirmed elevated PRKG2 expression inhibits the proliferation and invasion of lung cancer cells. PRKG2 is one of the genes shared by SSc and lung cancer, affecting the proliferation and invasion of lung cancer cells.

Construction and validation of a readmission risk prediction model for elderly patients with coronary heart disease

BackgroundTo investigate the risk factors for readmission of elderly patients with coronary artery disease, and to construct and validate a predictive model for readmission risk of elderly patients with coronary artery disease within 3 years by applying machine learning method.MethodsWe selected 575 elderly patients with CHD admitted to the Affiliated Lu’an Hospital of Anhui Medical University from January 2020 to January 2023. Based on whether patients were readmitted within 3 years, they were divided into two groups: those readmitted within 3 years (215 patients) and those not readmitted within 3 years (360 patients). Lasso regression and multivariate logistic regression were used to compare the predictive value of these models. XGBoost, LR, RF, KNN and DT algorithms were used to build prediction models for readmission risk. ROC curves and calibration plots were used to evaluate the prediction performance of the model. For external validation, 143 patients who were admitted between February and June 2023 from a different associated hospital in Lu'an City were also used.ResultsThe XGBoost model demonstrated the most accurate prediction performance out of the five machine learning techniques. Diabetes, Red blood cell distribution width (RDW), and Triglyceride glucose-body mass index (TyG-BMI), as determined by Lasso regression and multivariate logistic regression. Calibration plot analysis demonstrated that the XGBoost model maintained strong calibration performance across both training and testing datasets, with calibration curves closely aligning with the ideal curve. This alignment signifies a high level of concordance between predicted probabilities and observed event rates. Additionally, decision curve analysis highlighted that both decision trees and XGBoost models achieved higher net benefits within the majority of threshold ranges, emphasizing their significant potential in clinical decision-making processes. The XGBoost model's area under the ROC curve (AUC) reached 0.903, while the external validation dataset yielded an AUC of 0.891, further validating the model's predictive accuracy and its ability to generalize across different datasets.ConclusionTyG-BMI, RDW, and diabetes mellitus at the time of admission are the factors affecting readmission of elderly patients with coronary artery disease, and the model constructed based on the XGBoost algorithm for readmission risk prediction has good predictive efficacy, which can provide guidance for identifying high-risk patients and timely intervention strategies.

Investigating the Use of Novel Blood Processing Methods to Boost the Identification of Biomarkers for Non-Small Cell Lung Cancer: A Proof of Concept.

Diagnosis of non-small cell lung cancer (NSCLC) currently relies on imaging; however, these methods are not effective for detecting early stage disease. Investigating blood-based protein biomarkers aims to simplify the diagnostic process and identify disease-associated changes before they can be seen by using imaging techniques. In this study, plasma and frozen whole blood cell pellets from NSCLC patients and healthy controls were processed using both classical and novel techniques to produce a unique set of four sample types from a single blood draw. These samples were analyzed using 12 immunoassays and liquid chromatography-mass spectrometry to collectively screen 3974 proteins. Analysis of all fractions produced a set of 522 differentially expressed proteins, with conventional blood analysis (proteomic analysis of plasma) accounting for only 7 of the total. Boosted regression tree analysis of the differentially expressed proteins produced a panel of 13 proteins that were able to discriminate between controls and NSCLC patients, with an area under the ROC curve (AUC) of 0.864 for the set. Our rapid and reproducible (<10% CV for technical replicates) blood preparation and analysis methods enabled the production of high-quality data from only 30 μL of complex samples that typically require significant fractionation prior to proteomic analysis. With our methods, almost 4000 proteins were identified from a single fraction over a 62.5 min gradient by LC-MS/MS.

Effects of PGK1 on immunoinfiltration by integrated single-cell and bulk RNA-sequencing analysis in sepsis

BackgroundSepsis, a life-threatening organ dysfunction caused by a dysregulated immune response to infection, remains a significant global health challenge. Phosphoglycerate kinase 1 (PGK1) has been implicated in regulating inflammation and immune cell infiltration in inflammatory conditions. However, the role of PGK1 in sepsis remains largely unexplored.MethodsFour microarray datasets and a high throughput sequencing dataset were acquired from GEO database to reveal the PGK1 expression in patients of sepsis. Quantitative real-time PCR and western blotting was then used to validate the PGK1 level. Additionally, microarray and single-cell RNA sequencing data integration, including gene set enrichment analysis (GSEA), KEGG and GO functional enrichment analysis, immune infiltration analysis, and single-cell sequencing analysis, were performed to elucidate the role of PGK1 in sepsis.ResultsOur results revealed a significant upregulation of PGK1 in sepsis patients, with the area under the ROC curve (AUC) exceeding 0.9 across multiple datasets, indicating PGK1’s strong potential as a diagnostic biomarker. Notably, PGK1 was enriched in key immune-related pathways, including the TNF signaling pathways, and leukocyte transendothelial migration, suggesting its involvement in immune regulation. Furthermore, PGK1 expression showed a positive correlation with the levels of inflammatory mediators CXCL1, CXCL16, and the chemokine receptor CCR1. In terms of immune cell infiltration, PGK1 was positively correlated with naive B cells, resting memory CD4 T cell, gamma delta T cells, M0 macrophages, eosinophils and negatively correlated with plasma cells, CD8 T cells, activated memory CD4 T cell, Tregs, activated dendritic cells.ConclusionsThis study concluded that PGK1 served as a novel diagnostic biomarker for sepsis, with potential implications for prognosis and immune regulation. The significant upregulation of PGK1 in sepsis patients and its association with immune-related pathways and cell types highlight its potential role in the pathogenesis of sepsis.

A web-based dynamic nomogram for estimating talaromycosis risk in hospitalized HIV-positive patients.

Our study aimed to develop and validate a nomogram to assess talaromycosis risk in hospitalized HIV-positive patients. Prediction models were built using data from a multicentre retrospective cohort study in China. On the basis of the inclusion and exclusion criteria, we collected data from 1564 hospitalized HIV-positive patients in four hospitals from 2010 to 2019. Inpatients were randomly assigned to the training or validation group at a 7:3 ratio. To identify the potential risk factors for talaromycosis in HIV-infected patients, univariate and multivariate logistic regression analyses were conducted. Through multivariate logistic regression, we determined ten variables that were independent risk factors for talaromycosis in HIV-infected individuals. A nomogram was developed following the findings of the multivariate logistic regression analysis. For user convenience, a web-based nomogram calculator was also created. The nomogram demonstrated excellent discrimination in both the training and validation groups [area under the ROC curve (AUC)=0.883 vs. 0.889] and good calibration. The results of the clinical impact curve (CIC) analysis and decision curve analysis (DCA) confirmed the clinical utility of the model. Clinicians will benefit from this simple, practical, and quantitative strategy to predict talaromycosis risk in HIV-infected patients and can implement appropriate interventions accordingly.

Comparison of 2D and 3D lung lobe quantification with Ventilation/Perfusion Ratio.

In this study, standard 2D lung lobe quantification is compared with two 3D lung lobe quantification software tools to investigate the clinical benefit of a 3D approach. The accuracy of 2D versus 3D lung lobe quantification is evaluated based on the calculated numerical ventilation-perfusion ratio (VQR) using a receiver operating curve (ROC) analysis.A study group of 50 consecutive patients underwent a planar lung scintigraphy (anterior/posterior) as well as ventilation/perfusion single photon emission computed tomography (SPECT/CT) to exclude acute pulmonary embolism. All data were acquired with SPECT OPTIMA NM/CT 640 (GE Healthcare). 2D analysis was performed for all ventilation/perfusion scans using a lung analysis tool (Syngo Workstation, Siemens Healthineers). 3D quantification analysis was performed using QLUNG (Q. Lung, Xeleris 4.0, GE Healthcare) and LLQ (Hermes Hybrid 3D Lung Lobar Quantification, Hermes Medical Solutions). The area under the ROC curve (AUC) served as a decision criterion to find the best agreement between clinical PE findings and calculated PE candidates of the 2D and 3D methods. The significance of the ROC curves was evaluated using the DeLong comparison.A significant difference between 2D/3D could be determined. Both 3D approaches showed robust and comparable results. The AUC range of [0.10, 0.67] was given for 2D lobar analysis, QLUNG AUC range revealed in [0.39,0.74] and LLQ AUC range was [0.42,0.72]. Averaged over all lung lobes an AUC=0.39 was given for 2D analysis and AUC=0.58 was given for LLQ/QLUNG.We could demonstrate the better performance of 3D analysis compared to 2D analysis. Consequently, is recommended to use a 3D approach in clinical practice.

Integrative multi-omics approach using random forest and artificial neural network models for early diagnosis and immune infiltration characterization in ischemic stroke.

Ischemic stroke (IS) is a significant global health issue, causing high rates of morbidity, mortality, and disability. Since conventional Diagnosis methods for IS have several shortcomings. It is critical to create new Diagnosis models in order to enhance existing Diagnosis approaches. We utilized gene expression data from the Gene Expression Omnibus (GEO) databases GSE16561 and GSE22255 to identify differentially expressed genes (DEGs) associated with IS. DEGs analysis using the Limma package, as well as GO and KEGG enrichment analyses, were performed. Furthermore, PPI networks were constructed using DEGs from the String database, and Random Forest models were utilized to screen key DEGs. Additionally, an artificial neural network model was developed for IS classification. Use the GSE58294 dataset to evaluate the effectiveness of the scoring model on healthy controls and ischemic stroke samples. The effectiveness of the scoring model was evaluated through AUC analysis, and CIBERSORT analysis was conducted to estimate the immune landscape and explore the correlation between gene expression and immune cell infiltration. A total of 26 significant DEGs associated with IS were identified. Metascape analysis revealed enriched biological processes and pathways related to IS. 10 key DEGs (ARG1, DUSP1, F13A1, NFIL3, CCR7, ADM, PTGS2, ID3, FAIM3, HLA-DQB1) were selected using Random Forest and artificial neural network models. The area under the ROC curve (AUC) for the IS classification model was found to be near 1, indicating its high accuracy. Additionally, the analysis of the immune landscape demonstrated elevated immune-related networks in IS patients compared to healthy controls. The study uncovers the involvement of specific genes and immune cells in the pathogenesis of IS, suggesting their importance in understanding and potentially targeting the disease.

AI-enabled thermal monitoring of commercial (PHEV) Li-ion pouch cells with Feature-Adapted Unsupervised Anomaly Detection

Distributed temperature profiling of lithium-ion batteries provides valuable insights, aiding thermal management and minimising risk of battery failures. Highlighted by Batteries Europe as crucial for battery safety, advances in thermal monitoring are imperative to continuous safe adoption of battery technology. Deep Learning techniques have recently emerged as powerful tools for anomaly detection (AD) in many thermal mapping applications. These data-driven methods can handle common challenges like data unavailability or environment variations. Our study devises a methodology to leverage Deep Learning with thermal data from commercially available pouch cells and an infrared camera. We explain the building blocks of FAUAD (Feature-Adapted Unsupervised Anomaly Detection), which models the normality of the input data and synthesizes anomalies in its feature space. The resulting model is benchmarked against some of the latest state-of-the-art methods and achieves high anomaly detection capability; Area Under the ROC Curve (AUROC) score of 0.971 on simulated data, 0.990 on contaminated real data, and a perfect score of 1.0 on real clean data. While maintaining a compact size of 15 MB. FAUAD offers a notable advancement in unsupervised anomaly detection for battery thermal monitoring. The proposed method is cell chemistry agnostic and open to usage scenarios beyond this works’ scope.

Utilization and validation of dried blood spot-based metabolomics in plasma-derived diagnostic models

Metabolomics has rapidly advanced in life sciences, enhancing our understanding of physiological conditions. Plasma stands out as a predominant sample type in metabolomics studies. Dried blood spot (DBS) has been recognized as a valuable strategy due to its unique characteristics, such as minimal invasiveness, small sample volume, and easy transport. Most large cohort studies construct the diagnostic model using plasma-based metabolomics rather than DBS. However, the comparability of metabolite measurements between DBS and plasma samples, as well as the integration of DBS-based metabolomics into established plasma-derived diagnostic models, remains unclear. We collected plasma and DBS samples from the same 12 healthy controls under fasting and non-fasting conditions and performed ultra-high performance liquid chromatography-mass spectrometry (UPLC-MS)-based metabolomics. We identified 277 metabolites present in both plasma and DBS samples, with 229 (82.7 %) showing a strong correlation between the two sample types. Furthermore, pre-processing datasets from plasma and DBS samples before combining them effectively diminished inconsistencies between plasma- and DBS-based metabolomics without compromising the inherent classification within the data. For the clinical application, the plasma samples were obtained from the gastric cancer patients (n = 204) and the healthy controls (n = 128) to serve as the training cohorts. Additionally, DBS samples were collected from different participants of gastric cancer (n = 19) and healthy controls (n = 12) to validate the diagnostic model. Finally, DBS-based metabolomics were integrated into established plasma-derived gastric cancer diagnostic models, yielding an area under the ROC curve (AUC) of 0.934. Overall, incorporating DBS-based metabolomics into an established plasma-derived diagnostic model holds promise, offering valuable opportunities and insights for diagnostic research and clinical practice.

A Novel Prognostic Risk Score Model Based on RNA Editing Level in Lower-Grade Glioma

BackgroundLower-grade glioma (LGG) refers to WHO grade 2 and 3 gliomas. Surgery combined with radiotherapy and chemotherapy can significantly improve the prognosis of LGG patients, but tumor progression is still unavoidable. As a form of posttranscriptional regulation, RNA editing (RE) has been reported to be involved in tumorigenesis and progression and has been intensively studied recently. MethodsSurvival data and RE data were subjected to univariate and multivariate Cox regression analysis and lasso regression analysis to establish an RE risk score model. A nomogram combining the risk score and clinicopathological features was built to predict the 1-, 3-, and 5-year survival probability of patients. The relationship among ADAR1, SOD2 and SOAT1 was verified by reverse transcription-quantitative polymerase chain reaction (RT-qPCR) ResultsA risk model associated with RE was constructed and patients were divided into different risk groups based on risk scores. The model demonstrated strong prognostic capability, with the area under the ROC curve (AUC) values of 0.882, 0.938, and 0.947 for 1-, 3-, and 5-year survival predictions, respectively. Through receiver operating characteristic curve (ROC) curves and calibration curves, it was verified that the constructed nomogram had better performance than age, grade, and risk score in predicting patient survival probability. Apart from this functional analysis, the results of correlation analyses between risk differentially expressed genes (RDEGs) and RE help us to understand the underlying mechanism of RE in LGG. ADAR may regulate the expression of SOD2 and SOAT1 through gene editing. ConclusionIn conclusion, this study establishes a novel and accurate 17-RE model and a nomogram for predicting the survival probability of LGG patients. ADAR may affect the prognosis of glioma patients by influencing gene expression.

Validation of the Node Reporting and Data System (Node-RADS) for standardized CT evaluation of regional lymph nodes in esophageal squamous cell carcinoma patients.

The accurate identification of positive lymph nodes in esophageal squamous cell carcinoma (ESCC) influences patient risk assessment and treatment decisions, but there is no standardized approach for radiological evaluation. The aim of this study was to verify the diagnostic performance of the new Node Reporting and Data System 1.0 (Node-RADS) in the assessment of lymph node metastasis in patients with ESCC, as verified by final histopathology. Node-RADS is a scoring system composed of different criteria for evaluating lymph node metastasis, with scores ranging from 1 to 5, corresponding to the degree of suspicion of lymph node involvement. In this single-center study, Node-RADS was used to retrospectively evaluate regional lymph nodes in 173 ESCC patients who underwent computed tomography (CT) before radical resection. In addition, the area under the ROC curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for the Node-RADS score and individual criteria. Node-RADS outperformed the individual assessment criteria (AUC: 94.3%, sensitivity: 96.5%, specificity: 92.0%), with scores ≥ 3 indicating the maximum diagnostic effectiveness. The diagnostic efficiency of the highest Node-RADS score surpassed that of the short axis score (AUC: 94.3% vs. 81.9%, p < 0.001). Our results indicated that the best diagnostic cut-off points for the short axis, long axis and short axis/long axis ratio were 9 mm, 11 mm, and 0.74, respectively. Node-RADS has emerged as a practical, repetitive method for the early identification of high-risk metastatic lymph nodes, providing therapeutic guidance and predicting disease prognosis in ESCC patients. Question How does the Node Reporting and Data System 1.0 (Node-RADS) perform in the assessment of lymph node metastasis in patients with esophageal squamous cell carcinoma (ESCC)? Findings The maximum diagnostic efficiency was achieved with a Node-RADS score of ≥ 3. Clinical relevance The Node-RADS has improved diagnostic efficiency for distinguishing lymph node metastasis in patients with ESCC.

Identification of a machine learning-based diagnostic model for axial spondyloarthritis in rheumatological routine care using a random forest approach

ObjectivesIn axial spondyloarthritis (axSpA), early diagnosis is crucial, but diagnostic delay remains long and diagnostic criteria do not exist. We aimed to identify a diagnostic model that distinguishes patients with...

A data-driven hybrid approach towards developing a circular economy diffusion model for the building construction industry

Although there is a growing body of knowledge about circular economy (CE) practices in the building construction industry (BCI), the prediction of CE diffusion within the BCI remains insufficiently explored. This paper aims to explore a hybrid approach towards predicting the diffusion of CE practices in the BCI of a developing economy. This study utilized the technology-organization-environment (TOE) framework to identify the essential factors influencing CE diffusion in the BCI. Survey data collected from 303 experts were analyzed using partial least squares structural equation modelling (PLS-SEM) to test the hypothesis regarding these influencing factors. Subsequently, machine learning (ML) algorithms were employed to develop a predictive model for CE diffusion in the BCI. SHapley Additive exPlanation (SHAP) was then applied to interpret the contributions of each essential factor to the predictive model. The PLS-SEM results advocated that four major factors, namely technological compatibility, relative technological advantages, top management support, and organizational readiness, significantly and positively influence CE diffusion in the BCI. Furthermore, random forest algorithm was identified as the optimal ML model for predicting CE diffusion, achieving an accuracy of 80.33% and ROC AUC (area under the Curve receiver operating characteristics) of 80.27%. According to the SHAP results, the three most essential features contributing to the random forest model prediction are organizational readiness, top management support, and the relative technological advantages of CE adoption. This study advances the existing literature on CE diffusion by offering a comprehensive, data-driven approach that stakeholders can leverage to forecast trends and patterns in CE practices. It also equips decision makers with strategic insights and pragmatic plans to foster CE diffusion in the BCI, particularly within the context of developing countries.

Predicting ICU Admission for Hospitalized COVID-19 Patients by Artificial Neural Network Combined with Elastic Net

Background: Machine learning models could assist physicians in identifying high-risk COVID-19 patients. This study aimed to predict the Intensive Care Unit (ICU) admission in COVID-19 hospitalized patients by the Artificial Neural Network (ANN) model combined with Elastic Net algorithm. Methods: In this prospective study, the data of 139 COVID-19 patients admitted to Imam Reza Hospital in Tabriz between 20 March and 5 April 2020, were analyzed. The Elastic Net method was used to choose features with high importance. The chosen variables were standardized and ANN was fitted to the data with one hidden layer based on the descending gradient algorithm. To validate the model, the training and test group method with a ratio of 70 to 30 was used. The model’s predictive power was reported by calculating the overall accuracy, sensitivity, specificity, and Area Under the ROC curve (AUC). Results: According to the results of the Elastic Net, the ANN model was constructed based on age, sex, body mass index, diabetes mellitus status, history of heart disease, and vital signs including systolic and diastolic blood pressure, saturated oxygen level, pulse rate, respiration rate, and body temperature. The overall accuracy of this model was 93.15%, sensitivity 80%, specificity 95.8%, and AUC 0.90. Saturated oxygen level, pulse rate, and age were the most important and predictive variables. Conclusion: In the investigated sample of patients admitted with COVID-19, the fitted ANN model had acceptable performance to predict the ICU admission. This finding could be useful for physicians and policy makers.

Serum homocysteine and left ventricular hypertrophy in adults with chronic kidney disease: A case–control study

Hyperhomocysteinemia (serum homocysteine concentration &gt; 15 μmol/L) is of high prevalence in chronic kidney disease (CKD). And myocardial hypertrophy is a common complication of CKD. Given that both hyperhomocysteinemia and cardiac hypertrophy have an association with CKD, we hypothesized that high level of plasma homocysteine (Hcy) is associated with a higher prevalence of ventricular hypertrophy(LVH) in adults with CKD. The registration number of the case-control study is ChiCTR2200064834. The information of inpatients with CKD including Echocardiograms and analysis of plasma Hcy concentrations were collected. We performed linear and logistic regression to investigate the association of plasma Hcy with left ventricular hypertrophy (LVH) (LVMI ≥ 95th percentile), adjusted for levels of hemoglobin, ferritin, cystatin C and β-adrenergic blocker therapy. Further, a stratified analysis of the relationship between plasma Hcy and LVH was carried out according to eGFR. The case records for 1068 inpatients with CKD were collected. After data soring and case-control matching, there were 374 samples screened for statistical analysis. Univariate logistic regression indicated a high level of serum Hcy had an association with LVH (OR, 1.16; 95% CI, 1.11–1.20). Finally, multivariable logistic regression suggested that hyperhomocysteinemia was independently associated with LVH (OR, 1.14; 95% CI, 1.10–1.19) after adjustment for hemoglobin, ferritin, cystatin C, and β-adrenergic receptor blocker therapy. We constructed a predicting model including the variable of Hcy for cardiac hypertrophy in CKD. The model had an area under the ROC curve (AUC) of 0.86 (95% CI: 0.82–0.89, P &lt; .001). The decision curve analysis (DCA) showed a superior net clinical benefit of model with Hcy over model without Hcy. Elevated level of serum Hcy is closely associated with LVH in adults with CKD.

A machine learning model revealed that exosome small RNAs may participate in the development of breast cancer through the chemokine signaling pathway

BackgroundExosome small RNAs are believed to be involved in the pathogenesis of cancer, but their role in breast cancer is still unclear. This study utilized machine learning models to screen for key exosome small RNAs and analyzed and validated them.MethodPeripheral blood samples from breast cancer screening positive and negative people were used for small RNA sequencing of plasma exosomes. The differences in the expression of small RNAs between the two groups were compared. We used machine learning algorithms to analyze small RNAs with significant differences between the two groups, fit the model through training sets, and optimize the model through testing sets. We recruited new research subjects as validation samples and used PCR-based quantitative detection to validate the key small RNAs screened by the machine learning model. Finally, target gene prediction and functional enrichment analysis were performed on these key RNAs.ResultsThe machine learning model incorporates six small RNAs: piR-36,340, piR-33,161, miR-484, miR-548ah-5p, miR-4282, and miR-6853-3p. The area under the ROC curve (AUC) of the machine learning model in the training set was 0.985 (95% CI = 0.948-1), while the AUC in the test set was 0.972 (95% CI = 0.882–0.995). RT-qPCR was used to detect the expression levels of these key small RNAs in the validation samples, and the results revealed that their expression levels were significantly different between the two groups (P < 0.05). Through target gene prediction and functional enrichment analysis, it was found that the functions of the target genes were enriched mainly in the chemokine signaling pathway.ConclusionThe combination of six plasma exosome small RNAs has good prognostic value for women with positive breast cancer by imaging screening. The chemokine signaling pathway may be involved in the early stage of breast cancer. It is worth further exploring whether small RNAs mediate chemokine signaling pathways in the pathogenesis of breast cancer through the delivery of exosomes.

Risk factors and nomogram predictive models for postsurgical progression/hyperprogression recurrence in hepatocellular carcinoma with macroscopic vascular invasion

PurposeThis study aimed to develop postsurgical progression/hyperprogression recurrence (type III-IV recurrence) prediction models for hepatocellular carcinoma (HCC) patients with macroscopic vascular invasion (MaVI) and to guide treatment strategies in the accurate healthcare era.Patients and methods393 HCC patients with MaVI from two central hospitals made up the entire study population. In developmental (290 patients) and validation (103 patients) cohorts, all patients were randomized into one or the other. Two prediction models for type III-IV recurrence were developed, based on the findings of univariate and multivariate analysis in the development cohort, and multidimensional verification was carried out in both cohorts.ResultsThe postoperative recurrence rate of type III-IV in 393 HCC patients with MaVI was 70.9%. Young age, large tumor size (≥ 10 cm), node number, incomplete tumor capsule, postoperative complications, and high Ki67 index were the independent risk factors for relapse of type III-IV. In the development cohort, two nomograms (pre- and postoperative) had the Area Under the ROC curve (AUC) of 0.827 and 0.891, respectively. The two nomograms performed well, according to multidimensional verification methods such as clinical impact curves, decision curve analysis (DCA), and calibration curves. The validation cohort saw similar encouraging results. Both nomograms could separate patients into two distinct prognosis subgroups with ideal cutoff values of 170.3 presurgery and 175.0 postsurgery (both P < 0.05).ConclusionWe constructed two novel and potentially clinically valuable models for predicting type III-IV recurrence. These two models can develop strategies for treating those suffering from HCC with MaVI owing to their strong prediction performance and availability.

Combining interferon-γ release assays and metagenomic next-generation sequencing for diagnosis of pulmonary tuberculosis: a retrospective study

BackgroundRapid diagnosis of pulmonary tuberculosis (PTB) is urgently needed. We aimed to improve diagnosis rates by combining tuberculosis-interferon (IFN)-γ release assays (TB-IGRA) with metagenomic next-generation sequencing (mNGS) for PTB diagnosis.MethodsA retrospective study of 29 PTB and 32 non-TB patients from our hospital was conducted between October 2022 and June 2023. Samples were processed for TB-IGRA and mNGS tests according to the manufacturer’s protocol.ResultsThe levels of IFN-γ release in PTB patients were significantly higher than those in non-TB patients (604.15 ± 112.18 pg/mL, and 1.04 ± 0.38 pg/mL, respectively; p < 0.0001). Regarding presenting symptoms or signs, cough and thoracalgia were less common in PTB patients than in non-TB patients (p = 0.001 and p = 0.024, respectively). Total protein and albumin levels in the sera of PTB patients were significantly elevated compared to non-TB patients (p = 0.039 and p = 0.004, respectively). The area under the ROC curve (AUC) for TB-IGRA in PTB diagnosis was 0.939. With an optimal IFN-γ cut-off value of 14.3 pg/mL (Youden’s index 0.831), sensitivity was 86.2% and specificity was 96.9%. ROC curve analysis for mNGS and TB-IGRA combined with mNGS showed AUCs of 0.879 and 1, respectively. The AUC of TB-IGRA combined with mNGS was higher than that of TB-IGRA and mNGS alone.ConclusionsTB-IGRA combined with mNGS may be an effective method for diagnosing tuberculosis, and can be used in the clinical diagnosis of PTB.

Using logistic regression model to predict the future coronary heart disease

Abstract. Coronary Heart Disease (CHD) has become a significant concern due to the global rise in cardiovascular diseases, prompting a need for effective predictive tools. This study employs logistic regression modeling, a machine learning technique, to predict the risk of CHD and identify key influencing factors. Initially, complex data was preprocessed to handle missing values and other issues. The accuracy of the logistic regression model was then evaluated using confusion matrices and classification reports. Additionally, the model's performance was validated through the calculation of odds ratios and the analysis of ROC curves. The results showed that logistic regression performed well in the prediction of CHD, with a model prediction accuracy of 0.86 and an area under the ROC curve (AUC) of 0.73. The model identified the main factors that affect the development of CHD and pointed out potential room for improvement in predictive accuracy. However, although the model has achieved a high accuracy, there is still room for further improvement. In addition, focusing on improving data preprocessing techniques, especially in dealing with missing values, may improve the discriminatory ability of the model.

Predicting Mortality in Severe Burns: A Comparison of Four Mortality Prediction Scores and the Role of Organizational Changes in the Croatian Burn Center

Background: The primary aim of this study was to evaluate the performance of four burn prognostic scores—Abbreviated Burn Severity Index (ABSI), Ryan, Belgium Outcome Burn Injury (BOBI), and revised Baux score (rBaux) in a Croatian burn center. A secondary aim was to compare patient outcomes before and after the organizational and protocol changes. Methods: A retrospective study and comparison of four prediction scores was conducted over a nine-year period in burn patients with ≥20% total body surface area (TBSA) burned. Additionally, outcomes before and after organizational changes were compared. Results: A total of 149 patients were included, with the mean patient age of 54.62 ± 19.38 years, the mean of TBSA of 42.98 ± 19.90, and an overall mortality rate of 48.99%. The area under the ROC curve (AUROC) was 0.79 for the rBaux and ABSI score, 0.77 for the BOBI score, and 0.76 for the Ryan score. The duration of mechanical ventilation and length of stay (LOS) in burn intensive care units (BICU) decreased after the organizational changes, though survival rates remained similar. Conclusions: Prognostic scores are good predictors of mortality but with moderate predictive accuracy. Continuity of care in intensive care could be important for better outcomes.