Shrinkage Operator Research Articles

Dimension Reduction of MUSIC Database for Tailored Clustering Using Lasso

The tailored clustering enabled regionalization (TCER) framework has shown that the prediction error of design parameters at a target site can be reduced by constructing a transformation model from a cluster that only contains sites with geotechnical properties similar to the target site. These sites are retrieved from a global/regional database. However, the similarity between any two sites will reduce when the number of properties increases. TCER will be ineffective if too many properties are included in the database (e.g., structural health monitoring and remote sensing databases). The current study presents a least absolute shrinkage and selection operator for databases with incomplete records (denoted by HMLasso; a dimension reduction method) to improve the performance of TCER in high dimensional databases. HMLasso is utilized to construct a dimensionally reduced database, which includes only the properties relevant to the design parameters, from a high-dimensional database. TCER is then adopted to retrieve a cluster from the dimensionally reduced database for inferring the design parameters at a target site. The capability of HMLasso to enhance the performance of TCER in terms of identifying clusters similar to the target site and reducing the prediction error of design parameters are demonstrated using real-world geotechnical databases.

Nomogram for predicting proliferative lupus nephritis in patients with low-level proteinuria

Abstract Objectives Proliferative lupus nephritis (LN) is not uncommon in individuals with proteinuria < 0.5 g/24h, highlighting the importance of predicting proliferative nephritis for effective clinical management. We aimed to develop a predictive model for proliferative nephritis in this population. Methods The enrolled 671 biopsy-proven LN patients were divided into low-level proteinuria (< 0.5 g/24 h) and high-level proteinuria (≥ 0.5 g/24 h) groups. The clinical features, pathological characteristics and long-term outcomes of the two groups were compared. The least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression analysis were used to construct a predictive nomogram for proliferative nephritis in low-level proteinuria patients and internal validation was performed using bootstrap-resampling. Results 103 of 671 (15.4%) LN patients had low-level proteinuria, while 43 (41.7%) of whom showed proliferative LN, and the activity index and chronicity index were 5 (IQR [4,7]) and 3 (IQR [2,4]), respectively. The long-term adverse renal events-free survival was preferable in the low-level proteinuria group. The LASSO-logistic regression identified that age, sex, mean arterial pressure, haemoglobin, platelet, 24-h proteinuria and anti-double strands DNA antibodies positivity were associated with proliferative nephritis in those with low-level proteinuria. The predictive model showed an area under curve of 0.900 (95% CI: 0.840–0.960) and a bootstrapped result of 0.894 (95% CI: 0.832–0.965), with good calibration. Conclusion 41.7% of the patients with low-level proteinuria exhibited proliferative LN when biopsied. The nomogram including clinical, urinary, and laboratory parameters might help with the prediction of proliferative LN before biopsy among patients with low-level proteinuria.

Determinants of medication adherence in patients with acute coronary syndrome: a secondary analysis of a randomised clinical trial

BackgroundCoronary heart disease (CHD) remains a leading cause of mortality and disability worldwide. Approximately half of the patients who have had a prior hospital admission for CHD will have a...

Development of a nomogram for overall survival in patients with esophageal carcinoma: A prospective cohort study in China

BACKGROUND Esophageal carcinoma (EC) presents a significant public health issue in China, with its prognosis impacted by myriad factors. The creation of a reliable prognostic model for the overall survival (OS) of EC patients promises to greatly advance the customization of treatment approaches. AIM To create a more systematic and practical model that incorporates clinically significant indicators to support decision-making in clinical settings. METHODS This study utilized data from a prospective longitudinal cohort of 3127 EC patients treated at Chongqing University Cancer Hospital between January 1, 2018, and December 12, 2020. Utilizing the least absolute shrinkage and selection operator regression alongside multivariate Cox regression analyses helped pinpoint pertinent variables for constructing the model. Its efficacy was assessed by concordance index (C-index), area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). RESULTS Nine variables were determined to be significant predictors of OS in EC patients: Body mass index (BMI), Karnofsky performance status, TNM stage, surgery, radiotherapy, chemotherapy, immunotherapy, platelet-to-lymphocyte ratio, and albumin-to-globulin ratio (ALB/GLB). The model demonstrated a C-index of 0.715 (95%CI: 0.701-0.729) in the training cohort and 0.711 (95%CI: 0.689-0.732) in the validation cohort. In the training cohort, AUCs for 1-year, 3-year, and 5-year OS predictions were 0.773, 0.787, and 0.750, respectively; in the validation cohort, they were 0.772, 0.768, and 0.723, respectively, illustrating the model's precision. Calibration curves and DCA verified the model's predictive accuracy and net benefit. CONCLUSION A novel prognostic model for determining the OS of EC patients was successfully developed and validated to help clinicians in devising individualized treatment schemes for EC patients.

Establishment and internal validation of a model to predict the efficacy of Adalimumab in Crohn’s disease

Background: Clinically, the ability to distinguish which Crohn’s disease patients can benefit from Adalimumab is limited. Aims: This study aimed to develop a model for predicting clinical remission probability for Crohn’s disease patients with Adalimumab at 12 weeks. The model assists clinicians in identifying which Crohn’s disease patients are likely to benefit from Adalimumab treatment before starting therapy, thus optimizing individualized treatment strategies. Methods: Demographic and clinical characteristics of Crohn’s disease patients were utilized to develop a model for clinical remission probability. LASSO regression was used to select predictive factors, and predictions were made using a logistic regression model. The model was internally validated using the bootstrap method (resampling 1000 times). Results: 68 patients with Crohn’s disease were enrolled in this study. Clinical remission was observed in 55.9% at 12 weeks. Three variables were selected through the least absolute shrinkage and selection operator regression method, including Adalimumab-positive cell count, disease duration, and neutrophil count of Crohn’s disease patients. A predictive model was constructed by multivariate logistic regression (Adalimumab-positive cell count (OR, 1.143; 95%CI, 1.056–1.261), disease duration (OR, 0.967; 95%CI, 0.937–0.986), and neutrophil count (×109/L) (OR, 1.274; 95%CI,1.014–1.734)). The predictive model yielded an area under the curve of 0.866 (95%CI, 0.776–0.956), and in the internal validation, the area under the curve was 0.870 (95%CI, 0.770–0.940). Conclusions: This model provides a convenient tool to assess the likelihood of patient remission prior to Adalimumab treatment, thereby supporting the development of personalized treatment plans.

Cardiac Computed Tomography Measurements in Pulmonary Embolism Associated with Clinical Deterioration

Introduction: Most pulmonary embolism response teams (PERT) use a radiologist-determined right ventricle to left ventricle ratio (RV:LV) cut-off of 1.0 to risk-stratify pulmonary embolism (PE) patients. Continuous measurements from computed tomography pulmonary angiograms (CTPAs) may improve risk stratification. We assessed associations of CTPA cardiac measurements with acute clinical deterioration and use of advanced PE interventions. Methods: This was a retrospective study of a PE registry used by eight affiliated emergency departments. We used an artificial intelligence (AI) algorithm to measure RV:LV on anonymized CTPAs from registry patients for whom the PERT was activated (2018–2023) by institutional guidelines. Primary outcome was in-hospital PE-related clinical deterioration defined as cardiac arrest, vasoactive medication use for hypotension, or rescue respiratory interventions. Secondary outcome was advanced intervention use. We used bivariable and multivariable analyses. For the latter, we used least absolute shrinkage and selection operator (LASSO) and random forest (RF) to determine associations of all candidate variables with the primary outcome (clinical deterioration), and the Youden index to determine RV:LV optimal cut-offs for primary outcome. Results: Artificial intelligence analyzed 1,467 CTPAs, with 88% agreement on RV:LV categorization with radiologist reports (kappa 0.36, 95% confidence interval [CI] 0.28–0.43). Of 1,639 patients, 190 (11.6%) had PE-related clinical deterioration, and 314 (19.2%) had advanced interventions. Mean RV:LV were 1.50 (0.39) vs 1.30 (0.32) for those with and without clinical deterioration and 1.62 (0.33) vs 1.35 (0.32) for those with and without advanced intervention use. The RV:LV cut-off of 1.0 by AI and radiologists had 0.02 and 0.53 P-values for clinical deterioration, respectively. With adjusted LASSO, top clinical deterioration predictors were cardiac arrest at presentation, lowest systolic blood pressure, and intensive care unit admission. The RV:LV measurement was a top 10 predictor of clinical deterioration by RF. Optimal cut-off for RV:LV was 1.54 with odds ratio of 2.50 (1.85, 3.45) and area under the curve 0.6 (0.66, 0.70). Conclusion: Artifical intelligence-derived RV:LV measurements ≥1.5 on initial CTPA had strong associations with in-hospital clinical deterioration and advanced interventions in a large PERT database. This study points to the potential of capitalizing on immediately available CTPA RV:LV measurements for gauging PE severity and risk stratification.

Necroptosis-Related Gene Signature Predicts Prognosis in Patients with Advanced Ovarian Cancer

Background/Objectives: This study aimed to construct a risk score (RS) based on necroptosis-associated genes to predict the prognosis of patients with advanced epithelial ovarian cancer (EOC). Methods: EOC data from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) series 140082 (GSE140082) were used. Based on known necroptosis-associated genes, clustering was performed to identify molecular subtypes of EOC. A least absolute shrinkage and selection operator (LASSO)–Cox regression analysis identified key genes related to prognosis. The expression of one of them, RIPK3, was analyzed via immunohistochemistry in an EOC cohort. Results: An RS made from ten genes (IDH2, RIPK3, FASLG, BRAF, ITPK1, TNFSF10, ID1, PLK1, MLKL and HSPA4) was developed. Tumor samples were divided into a high-risk group (HRG) and low-risk group (LRG) using the RS. The model is able to predict the overall survival (OS) of EOC and distinguish the prognosis of different clinical subgroups. Immunohistochemical verification of the receptor-interacting serine/threonine-protein kinase (RIPK) 3 confirmed that high nuclear expression is correlated with a longer OS. In addition, the score can predict the response to a programmed death ligand 1 (PD-L1) blockade treatment in selected solid malignancies. Patients from the LRG seem to benefit more from it than patients from the HRG. Conclusions: Our RS based on necroptosis-associated genes might help to predict the prognosis of patients with advanced EOC and gives an idea on how the use of immunotherapy can potentially be guided.

Development of Novel Nomograms to Predict 5- and 7-Year Biochemical-Recurrence-Free Survival in High-Risk Prostate Cancer Patients After Carbon-Ion Radiotherapy and Androgen Deprivation Therapy

Background: The aim of this study was to develop nomograms predicting 5- and 7-year biochemical-recurrence (BCR)-free survival in high-risk prostate cancer (PCa) patients treated with carbon-ion radiotherapy (CIRT) and androgen deprivation therapy (ADT). Methods: We retrospectively evaluated 785 high-risk PCa patients treated with CIRT and ADT. Based on the least absolute shrinkage and selection operator model, two nomograms predicting 5- and 7-year BCR-free survival were developed and internally validated. The ability of each nomogram to predict BCR-free survival was determined by calculating the area under the survival curve (AUC). Results: The 5- and 7-year BCR-free survival rates were 92.1% and 89.3%, respectively. Age, prostate-specific antigen level, clinical T stage, and Gleason score were incorporated into the nomogram predicting 5-year BCR-free survival. In addition to these variables, the percentage of positive biopsy cores was also added to the nomogram predicting 7-year BCR-free survival. The AUC value of each nomogram showed suboptimal-to-good discrimination. Conclusions: We developed the first nomograms accurately predicting BCR-free survival in high-risk PCa patients treated with CIRT and ADT. These nomograms will enable adequate understanding and explanation of BCR-free survival to patients when clinicians use them.

Combination of inflammatory proteins in serum can be used to diagnose papillary thyroid carcinoma with lymph node metastasis.

Accurately distinguishing lymph node metastases (LNM) from papillary thyroid carcinomas (PTC) is crucial in clinical practice. The role of the immune system in PTC-LNM has attracted increasing attention. The aim of the present study was to evaluate the differential expression of 92 immune-related proteins in the serum and identify their potential diagnostic effects in patients with PTC-LNM. The 92 immune-related proteins were analyzed using a proximity extension assay. In addition, logistic regression and least absolute shrinkage and selection operator regression methods were used to develop combined diagnostic markers for thyroid cancer. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic validity. Gene Ontology, Kyoto Encyclopedia of Genes and Genomes, and Gene Set Enrichment Analysis were used to analyze the potential regulatory pathways. Five proteins, including IL-22RA1, IL-12B, CCL4, CCL3, and IL-1α, were significantly elevated in the serum of patients with LNM. The combined diagnosis of these five proteins demonstrated excellent diagnostic performance in distinguishing patients with PTC-LNM (area under the curve = 0.967, sensitivity = 0.941, and specificity = 0.889). Further analysis revealed that IL12B and IL1A mRNAs were significantly overexpressed in patients with PTC-LNM. This study also showed that the IL12B and IL1A was closely related to the PI3K-AKT, NF-κB, and MAPK signaling pathways. The combination of IL-22RA1, IL-12B, CCL4, CCL3, and IL-1α represents a promising diagnostic panel for PTC-LNM. These findings provide a novel set of diagnostic markers for PTC-LNM based on serum inflammatory protein levels.

Therapeutic implications and comprehensive insights into cellular senescence and aging in the tumor microenvironment of sarcoma

Sarcoma (SARC), a diverse group of stromal tumors arising from mesenchymal tissues, is often associated with a poor prognosis. Emerging evidence indicates that senescent cells within the tumor microenvironment (TME) significantly contribute to cancer progression and metastasis. Although the influence of senescence on SARC has been partially acknowledged, it has yet to be fully elucidated. In this study, we revealed that senescence level and age were associated with TME, immune treatment indicators, and prognosis in SARC. Utilizing the weighted gene co-expression network analysis and least absolute shrinkage and selection operator algorithm, we identified senescence-related genes and developed a senescence predictor. Three genes, RAD54, PIK3IP1, and TRIP13, were selected to construct a multiple linear regression model. Validation cohorts, immunohistochemistry, and quantitative reverse transcription polymerase chain reaction confirmed that the predictor derived from these three genes possessed prognostic and pathological relevance. Our senescence predictor provides comprehensive insights into the molecular mechanisms of SARC and identifies potential biomarkers for prognosis, paving the way for effective treatments. The results of this study hold promise for developing therapeutic strategies tailored to the unique characteristics of SARC.

Predicting colorectal adenomatous polyps in patients with chronic liver disease: A novel nomogram.

Colorectal polyps are commonly observed in patients with chronic liver disease (CLD) and pose a significant clinical concern because of their potential for malignancy. To explore the clinical characteristics of colorectal polyps in patients with CLD, a nomogram was established to predict the presence of adenomatous polyps (AP). Patients with CLD who underwent colonoscopy at Tianjin Second People's Hospital from January 2020 to May 2023 were evaluated. Clinical data including laboratory results, colonoscopy findings, and pathology reports were collected. Key variables for the nomogram were identified through least absolute shrinkage and selection operator regression, followed by multivariate logistic regression. The performance of the model was evaluated using the area under the receiver area under curve, as well as calibration curves and decision curve analysis. The study enrolled 870 participants who underwent colonoscopy, and the detection rate of AP in patients with CLD was 28.6%. Compared to individuals without polyps, six risk factors were identified as predictors for AP occurrence: Age, male sex, body mass index, alcohol consumption, overlapping metabolic dysfunction-associated steatotic liver disease, and serum ferritin levels. The novel nomogram (AP model) demonstrated an area under curve of 0.801 (95% confidence interval: 0.756-0.845) and 0.785 (95% confidence interval: 0.712-0.858) in the training and validation groups. Calibration curves indicated good agreement among predicted and actual probabilities (training: χ 2 = 11.860, P = 0.157; validation: χ 2 = 7.055, P = 0.530). The decision curve analysis underscored the clinical utility of the nomogram for predicting the risk of AP. The AP model showed reasonable accuracy and provided a clinical foundation for predicting the occurrence of AP in patients with CLD, which has a certain predictive value.

Prognostic model for esophagogastric variceal rebleeding after endoscopic treatment in liver cirrhosis: A Chinese multicenter study.

Rebleeding after recovery from esophagogastric variceal bleeding (EGVB) is a severe complication that is associated with high rates of both incidence and mortality. Despite its clinical importance, recognized prognostic models that can effectively predict esophagogastric variceal rebleeding in patients with liver cirrhosis are lacking. To construct and externally validate a reliable prognostic model for predicting the occurrence of esophagogastric variceal rebleeding. This study included 477 EGVB patients across 2 cohorts: The derivation cohort (n = 322) and the validation cohort (n = 155). The primary outcome was rebleeding events within 1 year. The least absolute shrinkage and selection operator was applied for predictor selection, and multivariate Cox regression analysis was used to construct the prognostic model. Internal validation was performed with bootstrap resampling. We assessed the discrimination, calibration and accuracy of the model, and performed patient risk stratification. Six predictors, including albumin and aspartate aminotransferase concentrations, white blood cell count, and the presence of ascites, portal vein thrombosis, and bleeding signs, were selected for the rebleeding event prediction following endoscopic treatment (REPET) model. In predicting rebleeding within 1 year, the REPET model exhibited a concordance index of 0.775 and a Brier score of 0.143 in the derivation cohort, alongside 0.862 and 0.127 in the validation cohort. Furthermore, the REPET model revealed a significant difference in rebleeding rates (P < 0.01) between low-risk patients and intermediate- to high-risk patients in both cohorts. We constructed and validated a new prognostic model for variceal rebleeding with excellent predictive performance, which will improve the clinical management of rebleeding in EGVB patients.

Identification and mechanistic insights of cell senescence-related genes in psoriasis

Background Psoriasis is a chronic inflammatory skin disease affecting 2–3% of the global population, characterised by red scaly patches that significantly affect patients’ quality of life. Recent studies have suggested that cell senescence, a state in which cells cease to divide and secrete inflammatory mediators, plays a critical role in various chronic diseases, including psoriasis. However, the involvement and mechanisms of action of senescence-related genes in psoriasis remain unclear. Methods This study aimed to identify senescence-related genes associated with psoriasis and explore their molecular mechanisms. RNA sequencing data from psoriasis and control samples were obtained from the GEO database. Differential expression analysis was performed using DESeq2 to identify differentially expressed genes (DEGs). The intersection of DEGs with cell senescence-related genes from the CellAge database was used to identify the candidate genes. Protein-protein interaction networks, Gene Ontology, and Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted to explore the functions and pathways of these genes. Machine learning algorithms, including Least Absolute Shrinkage and Selection Operator (LASSO) regression and Support vector machine-recursive feature elimination (SVE-RFE), were used to select feature genes that were validated by qRT-PCR. Additionally, an immune cell infiltration analysis was performed to understand the roles of these genes in the immune response to psoriasis. Results This study identified 4,913 DEGs in psoriasis, of which 46 were related to cell senescence. Machine learning highlighted four key genes, CXCL1, ID4, CCND1, and IRF7, as significant. These genes were associated with immune cell infiltration and validated by qRT-PCR, suggesting their potential as therapeutic targets for psoriasis. Conclusions This study identified and validated key senescence-related genes involved in psoriasis, providing insights into their molecular mechanisms and potential therapeutic targets and offering a foundation for developing targeted therapies for psoriasis.

Profiling Exosomal Metabolomics as a Means for Diagnosis and Researching Early-Stage Hypertensive Nephropathy

Aims/Background Hypertension (HT) is a prevalent medical condition showing an increasing incidence rate in various populations over recent years. Long-term hypertension increases the risk of the occurrence of hypertensive nephropathy (HTN), which is also a health-threatening disorder. Given that very little is known about the pathogenesis of HTN, this study was designed to identify disease biomarkers, which enable early diagnosis of the disease, through the utilization of high-throughput untargeted metabolomics strategies. Methods The participants of this study were patients admitted to The Second Affiliated Hospital of Chengdu Medical College, Nuclear Industry 416 Hospital, who were randomly divided into three groups: Normal group (n = 11), HT group (n = 10), and HTN group (n = 12). Urine exosomes were extracted, purified, and subjected to untargeted metabolomics analysis. Differential metabolites and their significantly enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were identified. The least absolute shrinkage and selection operator (LASSO) regression analysis was then employed to establish a diagnostic model for early-stage HTN. Finally, logistic regression and receiver operating characteristic (ROC) curve analysis were performed to identify biomarkers related to early HTN. Results Orthogonal partial least squares-discriminant analysis (OPLS-DA) revealed significant differences in the metabolic profiles of the three patient groups. Compared to subjects of the Normal group, the HT and HTN groups exhibited significantly upregulated and downregulated profiles of differential metabolites, respectively. LASSO regression analysis results indicated that 4-hydroxyphenylacetic acid, bilirubin, uracil, and iminodiacetic acid are potential biomarkers for HTN or HT. Conclusion With untargeted metabolomics analysis, we successfully identified differential metabolites in HTN. A further LASSO regression analysis revealed that four key metabolites, namely 4-hydroxyphenylacetic acid, bilirubin, uracil, and iminodiacetic acid, hold promise for the diagnosis of early-stage HTN.

Network Analysis of Internalizing and Externalizing Symptoms in Arab Gifted Children: A Cross-Sectional Study.

The current paper aimed to estimate the network structure of general psychopathology (internalizing and externalizing symptoms/disorders) among 239 gifted children in Jordan. This cross-sectional study with a convenience sampling method was conducted between September 2023 and October 2024 among gifted children aged 7-12. The Child Behavior Checklist (CBCL) was employed to assess six symptom clusters: conduct problems, attention-deficit/hyperactivity disorder (ADHD), and oppositional defiant problems as externalizing symptoms, and affective problems, anxiety issues, and somatic complaints as internalizing symptoms. We used the network analysis perspective by graphical least absolute shrinkage and selection operator (gLASSO) and the Extended Bayesian Information Criterion (EBIC). These methods were used to determine network structure and important nodes in the estimated network. "Sleeps less" (centrality strength = 2.04, edge weight = 0.33) was the central symptom in the affective cluster. In contrast, "worries" (centrality strength = 1.89, edge weight = 0.28) and "headaches" (centrality strength = 2.35, edge weight = 0.41) were pivotal in the anxiety and somatic clusters, respectively. The findings suggested that these symptoms had critical roles in the context of the general psychopathology among gifted children. Accordingly, the mentioned symptoms should be assessed and targeted among gifted children. Future studies could evaluate the results of targeting these symptoms on gifted children's well-being and daily functions.

Identification and validation of mitochondria-related LncRNA signatures as a novel prognostic model for glioma.

A predictive model for long-term survival is needed, and mitochondrial dysfunction is a key feature of cancer metabolism, though its link to glioma is not well understood. The aim of this study was to identify the molecular characteristics associated with glioma prognosis and explore its potential function. We analyzed RNA-seq data from The Cancer Genome Atlas and identified differentially expressed mitochondrial long noncoding RNAs (lncRNAs) using R's 'limma' package. A prognostic model was developed using 10 selected lncRNAs and validated with Cox regression and least absolute shrinkage and selection operator algorithm. The model's efficacy was assessed using Kaplan-Meier and receiver operating characteristic curve analyses, and its correlation with immune cell profiles and drug sensitivity was explored. A 10-mitochondria-related LncRNA signature was generated. The median risk score values are used to classify glioma samples into low-risk and high-risk groups. In breast patients, the signature-based risk score demonstrated a more potent ability to predict survival than conventional clinicopathological features. Furthermore, we noted a substantial disparity in the number of immune cells, including B cells, CD8+T cells, and macrophages, between the two groups. In addition, the high-risk group exhibited lower half-maximal inhibitory concentration values for specific chemotherapy medications, including bortezomib, luminespib, rapamycin, and 5-fluorouracil. Our study elucidates the diagnostic and prognostic value of mitochondria-related-lncRNAs in the promotion, suppression, and treatment of glioma.

Establishment and validation of a nomogram for coronary artery lesions in children with Kawasaki disease

BackgroundThe nomogram is a powerful and robust tool in disease risk prediction that summarizes complex variables into a visual model that is interpretable with a quantified risk probability. In the current study, a nomogram was developed to predict the occurrence of coronary artery lesions (CALs) among patients with Kawasaki disease (KD). This is especially valuable in the early identification of the risk of CALs, which will lead to proper diagnosis and treatment to reduce their associated complications.MethodsRetrospective clinical data of 677 children diagnosed with KD who were treated in the Children's Hospital Affiliated with Shandong University were analyzed. All the participants were divided into the CAL group and no CAL group according to their coronary echocardiography results. Least absolute shrinkage and selection operator (LASSO) regression was applied for the identification of the most informative predictors of CAL. Based on this, a nomogram was developed for accurate risk estimation.ResultsThe data were divided into a training set and a validation set. Receiver operating characteristic analysis, calibration curves, and decision curve analysis all supported the high accuracy and clinical utility of this model. LASSO regression highlighted five key predictors: sodium, hemoglobin, platelet count, D-dimer, and cystatin C. A nomogram based on these predictors was established and successfully validated in both datasets. In the training set, the AUC was 0.819 and in the validation set it was 0.844. The C-index of the calibration curve in the training set was 0.820, while in the validation set it was 0.844. In the decision curve analysis, the predictive benefit of the model was greater than zero when the threshold probability was below 95% in the training set and below 92% in the validation set.ConclusionThe predictive factors identified through the LASSO regression approach and the development of the nomogram are important contributions in this respect. This model had a high predictive accuracy and reliability for identifying high-risk children in the very early stage of disease with remarkable precision, laying the foundation for personalized treatment strategies and targeted treatment and providing a strong scientific basis for precise therapeutic intervention.

Association Between Urinary Metal Levels and Chronic Kidney Dysfunction in Rural China: A Study on Sex-Specific Differences

Background: While current epidemiological studies have documented associations between environmental metals and renal dysfunction, the majority have concentrated on plasma metal levels. The relationship between urinary metal exposure and chronic kidney disease (CKD) remains contentious, particularly within specific demographic groups. Methods: This cross-sectional study included 2919 rural Chinese adults recruited between 2018 and 2019. Urine metals were measured by ICP-MS. Least absolute shrinkage and selection operator (LASSO) regression was employed to identify metals significantly associated with CKD. Then, we used binary logistic regression, along with restricted cubic spline (RCS) models, to assess the individual exposure effects of specific metals on CKD. Quantile g-computation, weighted quantile sum regression, and Bayesian kernel machine regression (BKMR) models were applied to evaluate combined effects of metal exposures on CKD. Gender-stratified analyses were also conducted to explore these associations. Results: LASSO identified seven metals (V, Cu, Rb, Sr, Ba, W, Pb) with significant impacts on CKD. In single-metal models, Cu and W exhibited a positive correlation with CKD, whereas V, Rb, Sr, Ba, and Pb showed significant negative correlations (all p &lt; 0.05). RCS analysis revealed nonlinear associations between V, Cu, Ba, Pb, and CKD (all p-nonlinear &lt; 0.05). In the multi-metal model, quantile-based g-computation demonstrated a collective negative association with CKD risk for the seven mixed urinary metal exposures (OR (95% CI) = −0.430 (−0.656, −0.204); p &lt; 0.001), with V, Rb, Sr, Ba, and Pb contributing to this effect. The WQS model analysis further confirmed this joint negative association (OR (95% CI): −0.885 (−1.083, −0.899); p &lt; 0.001), with V as the main contributor. BKMR model analysis indicated an overall negative impact of the metal mixture on CKD risk. Interactions may exist between V and Cu, as well as Cu and Sr and Pb. The female subgroup in the BKMR model demonstrated consistency with the overall association. Conclusions: Our study findings demonstrate a negative association between the urinary metal mixture and CKD risk, particularly notable in females. Joint exposure to multiple urinary metals may involve synergistic or antagonistic interactions influencing renal function. Further research is needed to validate these observations and elucidate underlying mechanisms.

Prognostic value of multi-PLD ASL radiomics in acute ischemic stroke

IntroductionEarly prognosis prediction of acute ischemic stroke (AIS) can support clinicians in choosing personalized treatment plans. The aim of this study is to develop a machine learning (ML) model that uses multiple post-labeling delay times (multi-PLD) arterial spin labeling (ASL) radiomics features to achieve early and precise prediction of AIS prognosis.MethodsThis study enrolled 102 AIS patients admitted between December 2020 and September 2024. Clinical data, such as age and baseline National Institutes of Health Stroke Scale (NIHSS) score, were collected. Radiomics features were extracted from cerebral blood flow (CBF) images acquired through multi-PLD ASL. Features were selected using least absolute shrinkage and selection operator regression, and three models were developed: a clinical model, a CBF radiomics model, and a combined model, employing eight ML algorithms. Model performance was assessed using receiver operating characteristic curves and decision curve analysis (DCA). Shapley Additive exPlanations was applied to interpret feature contributions.ResultsThe combined model of extreme gradient boosting demonstrated superior predictive performance, achieving an area under the curve (AUC) of 0.876. Statistical analysis using the DeLong test revealed its significant outperformance compared to both the clinical model (AUC = 0.658, p &amp;lt; 0.001) and the CBF radiomics model (AUC = 0.755, p = 0.002). The robustness of all models was confirmed through permutation testing. Furthermore, DCA underscored the clinical utility of the combined model. The prognostic prediction of AIS was notably influenced by the baseline NIHSS score, age, as well as texture and shape features of CBF.ConclusionThe integration of clinical data and multi-PLD ASL radiomics features in a model offers a secure and dependable approach for predicting the prognosis of AIS, particularly beneficial for patients with contraindications to contrast agents. This model aids clinicians in devising individualized treatment plans, ultimately enhancing patient prognosis.

Automatic segmentation model and machine learning model grounded in ultrasound radiomics for distinguishing between low malignant risk and intermediate-high malignant risk of adnexal masses

ObjectiveTo develop an automatic segmentation model to delineate the adnexal masses and construct a machine learning model to differentiate between low malignant risk and intermediate-high malignant risk of adnexal masses based on ovarian-adnexal reporting and data system (O-RADS).MethodsA total of 663 ultrasound images of adnexal mass were collected and divided into two sets according to experienced radiologists: a low malignant risk set (n = 446) and an intermediate-high malignant risk set (n = 217). Deep learning segmentation models were trained and selected to automatically segment adnexal masses. Radiomics features were extracted utilizing a feature analysis system in Pyradiomics. Feature selection was conducted using the Spearman correlation analysis, Mann–Whitney U-test, and least absolute shrinkage and selection operator (LASSO) regression. A nomogram integrating radiomic and clinical features using a machine learning model was established and evaluated. The SHapley Additive exPlanations were used for model interpretability and visualization.ResultsThe FCN ResNet101 demonstrated the highest segmentation performance for adnexal masses (Dice similarity coefficient: 89.1%). Support vector machine achieved the best AUC (0.961, 95% CI: 0.925–0.996). The nomogram using the LightGBM algorithm reached the best AUC (0.966, 95% CI: 0.927–1.000). The diagnostic performance of the nomogram was comparable to that of experienced radiologists (p > 0.05) and outperformed that of less-experienced radiologists (p < 0.05). The model significantly improved the diagnostic accuracy of less-experienced radiologists.ConclusionsThe segmentation model serves as a valuable tool for the automated delineation of adnexal lesions. The machine learning model exhibited commendable classification capability and outperformed the diagnostic performance of less-experienced radiologists.Critical relevance statementThe ultrasound radiomics-based machine learning model holds the potential to elevate the professional ability of less-experienced radiologists and can be used to assist in the clinical screening of ovarian cancer.Key PointsWe developed an image segmentation model to automatically delineate adnexal masses.We developed a model to classify adnexal masses based on O-RADS.The machine learning model has achieved commendable classification performance.The machine learning model possesses the capability to enhance the proficiency of less-experienced radiologists.We used SHapley Additive exPlanations to interpret and visualize the model.Graphical