Internal Validity Research Articles

Machine learning and single-cell RNA sequencing reveal relationship between intratumor CD8+ T cells and uveal melanoma metastasis

PurposeUveal melanoma (UM) is adults’ most common primary intraocular malignant tumor. It has been observed that 40% of patients experience distant metastasis during subsequent treatment. While there exist multigene models developed using machine learning methods to assess metastasis and prognosis, the immune microenvironment’s specific mechanisms influencing the tumor microenvironment have not been clarified. Single-cell transcriptome sequencing can accurately identify different types of cells in a tissue for precise analysis. This study aims to develop a model with fewer genes to evaluate metastasis risk in UM patients and provide a theoretical basis for UM immunotherapy.MethodsRNA-seq data and clinical information from 79 μm patients from TCGA were used to construct prognostic models. Mechanisms were probed using two single-cell datasets derived from the GEO database. After screening for metastasis-related genes, enrichment analysis was performed using GO and KEGG. Prognostic genes were screened using log-rank test and one-way Cox regression, and prognostic models were established using LASSO regression analysis and multifactor Cox regression analysis. The TCGA-UVM dataset was used as internal validation and dataset GSE22138 as external validation data. A time-dependent subject work characteristic curve (time-ROC) was established to assess the predictive ability of the model. Subsequently, dimensionality reduction, clustering, pseudo-temporal analysis and cellular communication analysis were performed on GSE138665 and GSE139829 to explore the underlying mechanisms involved. Cellular experiments were also used to validate the relevant findings.ResultsBased on clinical characteristics and RNA-seq transcriptomic data from 79 samples in the TCGA-UVM cohort, 247 metastasis-related genes were identified. Survival models for three genes (SLC25A38, EDNRB, and LURAP1) were then constructed using lasso regression and multifactorial cox regression. Kaplan-Meier survival analysis showed that the high-risk group was associated with poorer overall survival (OS) and metastasis-free survival (MFS) in UM patients. Time-dependent ROC curves demonstrated high predictive performance in 6 m, 18 m, and 30 m prognostic models. Cell scratch assay showed that the 24 h and 48 h migration rates of cells with reduced expression of the three genes were significantly higher than those of the si-NC group. CD8 + T cells may play an important role in tumour metastasis as revealed by immune infiltration analysis. An increase in the percentage of cytotoxic CD8 + T cells in the metastatic high-risk group was found in the exploration of single-cell transcriptome data. The communication intensity of cytotoxic CD8 was significantly enhanced. It was also found that the CD8 + T cells in the two groups were in different states, although the number of CD8 + T cells in the high-risk group increased, they were mostly in the exhausted and undifferentiated state, while in the low-risk group, the CD8 + T cells were mostly in the functional state.ConclusionsWe developed a precise and stable 3-gene model to predict the metastatic risk and prognosis of patients. CD8 + T cells exhaustion in the tumor microenvironment play a crucial role in UM metastasis.

Development and validation of a renal involvement model for patients with hyperuricemia: A cross-sectional study.

To construct a prediction model for renal involvement in patients with hyperuricemia (HUA) based on logistic regression analysis, to achieve early risk stratification. In this cross-sectional study, we collected data from the National Health and Nutrition Examination Survey (NHANES), and constructed a predicted model for renal involvement in HUA patients. The discriminative ability of the model was assessed using the receiver operating characteristic (ROC) curve. Model accuracy was evaluated using the Hosmer-Lemeshow test and calibration curve, while clinical utility was assessed using decision curve analysis (DCA). Furthermore, internal and external validation cohorts were also applied to validate the model. A total of 1669 patients from NHANES between 2007 and 2010 were included in the final analysis for modeling and validation. Six predictive factors including age, Systolic Blood Pressure (SBP), Diastolic Blood Pressure (DBP), Cr, Uric Acid (UA), and sex were identified by binary logistic regression analysis for renal involvement in HUA patients and used to construct a nomogram with good consistency and accuracy. The AUC values for the predictive model, internal validation, and external validation were 0.881 (95% CI: 0.836-0.926), 0.908 (95% CI: 0.871-0.944), and 0.927 (95% CI: 0.897-0.957), respectively. The calibration curves demonstrated consistency between the nomogram and observed values. The DCA curves of the model and validation cohort indicated good clinical utility. This study developed a predictive model for renal involvement in hyperuricemia patients with strong predictive performance and validated by internal and external cohorts, aiding in the early detection of high-risk populations for renal involvement.

A nomogram incorporating linezolid and metabolite concentrations for predicting linezolid induced thrombocytopenia in patients with renal impairment

A nomogram to estimate the risk of linezolid-induced thrombocytopenia in patients with renal impairment is not available. The aim of the study is to develop a nomogram for predicting linezolid-induced thrombocytopenia in patients with renal impairment and to investigate the incremental value of PNU-142300 concentration beyond clinical factors and linezolid trough concentration (Cmin) for risk prediction. Logistic regression was used to identify independent risk factors for linezolid-induced thrombocytopenia in patients with renal impairment and nomograms were established. The performance of the nomograms was assessed in terms of area under the receiver operating characteristic curve (AUROC), net reclassification improvement (NRI), integrated discrimination improvement (IDI) , decision curve analysis (DCA) and calibration. Internal validation and external validation of the nomograms were also performed. Four nomograms were created: nomogram A including total bilirubin, creatinine clearance and concomitant mannitol use; nomogram B containing linezolid Cmin additionally; nomogram C containing total bilirubin, concomitant mannitol use, linezolid Cmin, and PNU142300 concentration; nomogram D including total bilirubin, concomitant mannitol use, and PNU142300 concentration. Nomogram C improved the prediction performance than nomogram A (AUROC 0.881 vs. 0.749; NRI 0.290; IDI 0.226) and nomogram B (AUROC 0.881 vs. 0.812; NRI 0.152; IDI 0.130) in the training cohort. DCA analysis showed that nomogram C yielded a greater net benefit. Compared with nomogram A and nomogram B, nomogram C also showed superior discriminatory efficacy, good calibration and clinical usefulness in the external validation cohort. The nomogram containing PNU-142300 concentration and linezolid Cmin had better predictive capability than that containing linezolid Cmin for predicting linezolid-induced thrombocytopenia in patients with renal impairment.

Aberrant fragmentomic features of circulating cell-free mitochondrial DNA as novel biomarkers for multi-cancer detection.

Fragmentomic features of circulating cell free mitochondrial DNA (ccf-mtDNA) including fragmentation profile, 5' end base preference and motif diversity are poorly understood. Here, we generated ccf-mtDNA sequencing data of 1607 plasma samples using capture-based next generation sequencing. We firstly found that fragmentomic features of ccf-mtDNA were remarkably different from those of circulating cell free nuclear DNA. Furthermore, region-specific fragmentomic features of ccf-mtDNA were observed, which was associated with protein binding, base composition and special structure of mitochondrial DNA. When comparing to non-cancer controls, six types of cancer patients exhibited aberrant fragmentomic features. Then, cancer detection models were built based on the fragmentomic features. Both internal and external validation cohorts demonstrated the excellent capacity of our model in distinguishing cancer patients from non-cancer control, with all area under curve higher than 0.9322. The overall accuracy of tissue-of-origin was 89.24% and 87.92% for six cancer types in two validation cohort, respectively. Altogether, our study comprehensively describes cancer-specific fragmentomic features of ccf-mtDNA and provides a proof-of-principle for the ccf-mtDNA fragmentomics-based multi-cancer detection and tissue-of-origin classification.

Development and Validation of Spiritual Well-Being Scale in India: A Yogic Niyama Perspective.

This study developed and validated the niyama-based spiritual well-being scale (NSWBS) using a mixed-method approach. Niyama is one of the eight limbs of ancient Indian Ashtanga yoga philosophy. There are five niyamas: soucha (internal and external cleanliness and hygiene), santosha (happiness and contentment), tapas (austerity and discipline), swadhyaya (self-study, observation and self-introspection) and ishvara-pranidhana (believe and surrender to the God and cosmic power). Both deductive (using existing spiritual well-being scales) and inductive (focus group discussions) approaches were used to generate items. These statements were first subjected to content validity testing and pre-testing, which resulted in seventeen statements. The psychometric properties of generated items were analyzed with the help of three independent studies that utilized various statistical tests like exploratory factor analysis (EFA), confirmatory factor analysis (CFA), Cronbach's alpha, McDonald Omega, Average variance explained, and Fornell and Larcker criterion. The first study examined the factorial structure of the proposed scale. EFA yielded five factors: Internal and external health and purity, Self-discipline and rigour, Self-introspection and observation, Contentment and gratitude, and Belief in God and higher cosmic power. They explained 62.35% of the total variance. The CFA validated the five factorial structure of the scale. The second study established the proposed scale's internal consistency reliability, convergent, divergent, and predictive validities. Lastly, the third study evaluated and concluded the test-retest reliability of the scale. Overall, the niyama-based spiritual well-being scale depicted appreciable psychometric properties.

Whole blood transcriptional profiles and the pathogenesis of tuberculous meningitis.

Mortality and morbidity from tuberculous meningitis (TBM) are common, primarily due to inflammatory response to Mycobacterium tuberculosis infection, yet the underlying mechanisms remain poorly understood. We aimed to uncover genes and pathways associated with TBM pathogenesis and mortality, and determine the best predictors of death, utilizing whole-blood RNA sequencing from 281 Vietnamese adults with TBM, 295 pulmonary tuberculosis (PTB), and 30 healthy controls. Through weighted gene co-expression network analysis, we identified hub genes and pathways linked to TBM severity and mortality, with a consensus analysis revealing distinct patterns between HIV-positive and HIV-negative individuals. We employed multivariate elastic-net Cox regression to select candidate predictors of death, then logistic regression and internal bootstrap validation to choose best predictors. Increased neutrophil activation and decreased T and B cell activation pathways were associated with TBM mortality. Among HIV-positive individuals, mortality associated with increased angiogenesis, while HIV-negative individuals exhibited elevated TNF signaling and impaired extracellular matrix organization. Four hub genes-MCEMP1, NELL2, ZNF354C, and CD4-were strong TBM mortality predictors. These findings indicate that TBM induces a systemic inflammatory response similar to PTB, highlighting critical genes and pathways related to death, offering insights for potential therapeutic targets alongside a novel four-gene biomarker for predicting outcomes.

Psychometric Evaluation of the Chinese Version of the Vulvovaginal Symptom Questionnaire in Women With Breast Cancer.

To evaluate the psychometric properties of the Chinese version of the Vulvovaginal Symptoms Questionnaire for assessing vulvovaginal symptoms and symptom-related influences in women with breast cancer. A methodological study. Women with breast cancer (n = 202) were recruited from the outpatient department of a hospital. Data were collected between July 2020 and October 2021. Psychometric properties, including internal consistency, test-retest reliability and construct validity, were tested after the translation of the original English-language instrument. The construct validity was examined by testing the hypothesised relationships between the Chinese version of the Vulvovaginal Symptoms Questionnaire with validated instruments associated with quality of life and sexual function and by Confirmatory Factor Analysis. The internal consistency and test-retest reliability for the Chinese version of the Vulvovaginal Symptoms Questionnaire's total scale and four subscales were satisfactory. The construct validity was confirmed by significant correlations between scores on the Chinese version of the Vulvovaginal Symptoms Questionnaire with the Chinese version of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Cancer 30 and Quality of Life Questionnaire-Breast 23 and the Chinese version of the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire 12. The Confirmatory Factor Analysis verification results showed that the traditional Chinese-language questionnaire's three- and four-factor models had acceptable model fit indices. We obtained the Chinese version of the Vulvovaginal Symptoms Questionnaire's preliminary and satisfactory psychometric properties. It can help worldwide healthcare professionals adequately assess vulvovaginal symptoms and their influences experienced by Chinese-speaking women with breast cancer. The Chinese version of the Vulvovaginal Symptoms Questionnaire can help healthcare professionals and researchers concurrently identify vulvovaginal symptoms and related influences, leading to timely and appropriate management. Well-designed and accessible healthcare services on vulvovaginal and sexual health after breast cancer diagnosis are essential for both healthcare professionals and this population. We adhered to the STROBE checklist of cross-sectional studies. No patient or public engagement..

Whole blood transcriptional profiles and the pathogenesis of tuberculous meningitis

Mortality and morbidity from tuberculous meningitis (TBM) are common, primarily due to inflammatory response to Mycobacterium tuberculosis infection, yet the underlying mechanisms remain poorly understood. We aimed to uncover genes and pathways associated with TBM pathogenesis and mortality, and determine the best predictors of death, utilizing whole-blood RNA sequencing from 281 Vietnamese adults with TBM, 295 pulmonary tuberculosis (PTB), and 30 healthy controls. Through weighted gene co-expression network analysis, we identified hub genes and pathways linked to TBM severity and mortality, with a consensus analysis revealing distinct patterns between HIV-positive and HIV-negative individuals. We employed multivariate elastic-net Cox regression to select candidate predictors of death, then logistic regression and internal bootstrap validation to choose best predictors. Increased neutrophil activation and decreased T and B cell activation pathways were associated with TBM mortality. Among HIV-positive individuals, mortality associated with increased angiogenesis, while HIV-negative individuals exhibited elevated TNF signaling and impaired extracellular matrix organization. Four hub genes—MCEMP1, NELL2, ZNF354C, and CD4—were strong TBM mortality predictors. These findings indicate that TBM induces a systemic inflammatory response similar to PTB, highlighting critical genes and pathways related to death, offering insights for potential therapeutic targets alongside a novel four-gene biomarker for predicting outcomes.

Characterising impaired awareness of hypoglycaemia and associated risks through HypoA-Q: findings from a T1D Exchange cohort.

We aimed to: (1) externally validate the five-item Hypoglycaemia Awareness Questionnaire (HypoA-Q) impaired awareness subscale (HypoA-Q IA); (2) examine how impaired awareness of hypoglycaemia (IAH) relates to the risk of severe hypoglycaemia and level 2 hypoglycaemia; and (3) identify factors associated with IAH. Nationwide survey of T1D Exchange registrants was conducted to collect data on demographics, 6 month severe-hypoglycaemia history, hypoglycaemia awareness status (via HypoA-Q IA, the Gold instrument and the Clarke instrument) and continuous glucose monitor (CGM) measures. The Clarke hypoglycaemia awareness factor (Clarke-HAF) was calculated to exclude severe-hypoglycaemia history items. Analyses included Cronbach's α, Spearman correlations and logistic regression. Valid survey responses were collected from N=1580 adults with type 1 diabetes (median age, 44 years; 52% female participants; median HbA1c, 48 mmol/mol [6.5%]). Of these, 94% of participants were using CGMs and 69% were using hybrid closed-loop (HCL) systems; 30% had at least one severe-hypoglycaemia episode in the past 6 months. The HypoA-Q IA had satisfactory internal reliability (α=0.79) and construct validity. Higher HypoA-Q IA scores were independently associated with greater risk of severe hypoglycaemia (p<0.001), performing comparably to the Gold instrument and the Clarke-HAF instrument. HypoA-Q IA-determined IAH was independently associated with 88% higher odds of developing severe hypoglycaemia (p<0.001) and twofold higher odds for spending ≥1% of time in level 2 hypoglycaemia (p=0.011). Higher age and longer diabetes duration were associated with higher IAH risk (p<0.001). CGM and HCL use was associated with lower IAH risk (p<0.001). The HypoA-Q IA is a brief, valid and reliable tool for assessing IAH in today's technology-oriented era. IAH was independently associated with severe hypoglycaemia and level 2 hypoglycaemia in a cohort with high prevalence of advanced diabetes technology use and HbA1c within the recommended range. CGM and HCL use was related to lower IAH risk.

Barriers and facilitators to conducting randomised controlled trials within routine care of neurorehabilitation centres: a qualitative study

BackgroundRandomised controlled trials (RCTs) are considered the gold standard for generating clinical evidence. The focus on high internal validity in RCTs challenges the external validity and generalisability of findings, potentially hindering their application in routine care. In neurorehabilitation, limited literature addresses conducting RCTs feasibly and efficiently. We investigated barriers and facilitators to conducting RCTs within routine care of neurorehabilitation centres from the perspective of stakeholders in neurorehabilitation in Germany and Austria.MethodsWe conducted semi-structured interviews with stakeholders in neurorehabilitation from four centres in Germany and Austria, informed by the Theoretical Domains Framework (TDF) and the Capability, Opportunity, Motivation and Behaviour model (COM-B). Employing a hybrid approach, the interview analysis integrated both deductive, theory-driven analysis based on the TDF domains and COM-B model and inductive, reflexive thematic analysis.ResultsTwelve stakeholders (4 physicians, 4 therapy managers, 4 therapists; 5 females, 7 males; with research experience spanning 0–40 years) were interviewed. Key barriers to conducting RCTs in neurological rehabilitation centres include limited financial, human, and time resources, high clinical workloads, and a lack of interest of some therapists. Ineffective leadership, perceived lack of research expertise, and communication issues were also significant barriers. Social influence factors such as lack of employer support and inadequate training access further contributed to the challenges. Additionally, barriers included insufficient research infrastructure, limited space, internal power struggles, and rigid cost bearer specifications. Key facilitators included physicians’ and therapists’ motivation to advance the field, contribute to knowledge, and to prioritise patient health. Support from supervisors, joint decision-making, and efficient organisation were crucial facilitators. Flexible therapy planning, mutual support, and interdisciplinary collaboration also played important roles.ConclusionOur results suggest that increasing professional development and understanding, along with providing adequate financial, human, time, and spatial resources to support research endeavours, implementing effective communication strategies to enhance interdisciplinary collaboration and coordination among team members may contribute to increased motivation and facilitate RCTs within the setting of neurorehabilitation centres.Trial registrationThis study was prospectively registered with the German Clinical Trials Register (08.04.2021 DRKSID DRKS00024982).

Development and internal and external validation of a nomogram model for predicting the risk of chronic kidney disease progression in IgA nephropathy patients.

IgA nephropathy (IgAN) is the most common primary glomerular disease in chronic kidney disease (CKD), exhibiting significant heterogeneity in both clinical and pathological presentations. We aimed to explore the risk factors influencing short-term prognosis (≥90 days) and to construct a nomogram model for evaluating the risk of CKD progression in IgAN patients. Clinical and pathological data of patients diagnosed with IgAN through biopsy at two centers were retrospectively collected. Logistic regression was employed to analyze the training cohort dataset and identify the independent predictors to construct a nomogram model based on the final variables. The predictive model was validated both internally and externally, with its performance assessed using the area under the curve (AUC), calibration curves, and decision curve analysis. Out of the patients in the modeling group, 129 individuals (41.6%) did not achieve remission following 3 months of treatment, indicating a high risk of CKD progression. A multivariate logistic regression analysis demonstrated that body mass index, urinary protein excretion, and tubular atrophy/interstitial fibrosis were identified as independent predictors for risk stratification. A nomogram model was formulated utilizing the final variables. The AUCs for the training set, internal validation set, and external validation set were 0.746 (95% confidence intervals (CI) [0.691-0.8]), 0.764 (95% CI [0.68-0.85]), and 0.749 (95% CI [0.65-0.85]), respectively. The validation of the subgroup analysis also demonstrated a satisfactory AUC. This study developed and validated a practical nomogram that can individually predict short-term treatment outcomes (≥90 days) and the risk of CKD progression in IgAN patients. It provides reliable guidance for timely and personalized intervention and treatment strategies.

Ambulation recovery prediction after hip fracture surgery using the Hip Fracture Short-Term Ambulation Prediction tool.

To develop models for predicting postoperative ambulation recovery at 3 months following fragility hip fracture surgery. Cross-sectional study. Fragility hip fracture patients aged ≥ 50 years who underwent operative treatment and completed a 3-month follow-up. Potential predictors were collected from eligible patients, while ambulation at 3 months after injury was assessed using the modified functional ambulation classification. These factors were used to develop the Hip Fracture Short-Term Ambulation Prediction, consisting of 2 models: Model 1 for postoperative ambulation and Model 2 for preinjury status recovery. Among the 275 patients, 55 (20.0%) achieved good ambulation, and 59 (21.5%) returned to their preinjury status at 3 months. Age, preinjury ambulatory status, and discharge ambulatory status were identified as significant predictors of 3-month postoperative ambulation. The tool presented (Models 1 and 2) showed strong performance (area under the curve of 0.86 and 0.85, respectively) and good internal validity. Age, preinjury ambulatory status, and discharge ambulatory status significantly predict postoperative ambulation and preinjury status recovery at 3 months after fragility hip fracture surgery. The tool presented may aid clinicians in identifying patients who could benefit from targeted rehabilitation interventions during this crucial period.

In silico aquatic toxicity prediction of chemicals toward Daphnia magna and fathead minnow using Monte Carlo approaches

The fast-increasing use of chemicals led to large numbers of chemical compounds entering the aquatic environment, raising concerns about their potential effects on ecosystems. Therefore, assessment of the ecotoxicological features of organic compounds on aquatic organisms is very important. Daphnia magna and Fathead minnow are two aquatic species that are commonly tested as standard test organisms for aquatic risk assessment and are typically chosen as the biological model for the ecotoxicology investigations of chemical pollutants. Herein, global quantitative structure–toxicity relationship (QSTR) models have been developed to predict the toxicity (pEC(LC)50) of a large dataset comprising 2106 chemicals toward Daphnia magna and Fathead minnow. The optimal descriptor of correlation weights (DCWs) is calculated using the notation of simplified molecular input line entry system (SMILES) and is used to construct QSTR models. Three target functions, TF1, TF2, and TF3 are utilized to generate 12 QSTR models from four splits, and their statistical characteristics are also compared. The designed QSTR models are validated using both internal and external validation criteria and are found to be reliable, robust, and excellently predictive. Among the models, those generated using the TF3 demonstrate the best statistical quality with R 2 values ranging from 0.9467 to 0.9607, Q 2 values ranging from 0.9462 to 0.9603 and RMSE values ranging from 0.3764 to 0.4413 for the validation set. The applicability domain and the mechanistic interpretations of generated models were also discussed.

Development and validation of a French job-exposure matrix for healthcare workers: JEM Soignances.

This study aimed to develop and evaluate a job-exposure matrix (JEM) specific to healthcare workers, JEM Soignances, based on self-reported data. The JEM was constructed using data from healthcare workers within the CONSTANCES cohort (N=12 489). Job titles and sectors of activity (eg, hospital activities) defined occupational groups. We assessed 24 exposures covering organizational, psychosocial, physical, chemical and biological factors. Several methods (group-based frequency, CART, random forest, extreme gradient boosting machine) were applied using a 70% training sample. Performance was evaluated on the remaining 30% using area under the ROC curve (AUC) and Cohen's Kappa (κ). Two alternative JEM were proposed using only job titles or adding healthcare establishment size and type (public/private) to define occupational groups. All methods offered similar discriminatory power (AUC). We selected the group-based frequency method as it was the most understandable and easiest to implement. Of the 24 included exposures, 15 demonstrated satisfactory performance, with nine showing good discriminatory power and fair-to-moderate agreement, such as physical effort at work (AUC=0.861, κ=0.556), ionizing radiation exposure (AUC=0.865, κ=0.457), carrying heavy loads (AUC=0.840, κ=0.402), shift work (AUC=0.807, κ=0.383), and formaldehyde exposure (AUC=0.847, κ=0.289). The remaining nine exposures mainly showed poor-to-moderate discriminatory power and poor agreement. Compared to JEM Soignances, the job title-only JEM performed poorly, while the one incorporating healthcare establishment size and type showed similar results. JEM Soignances provides good internal performance and validity. Future research will assess its external validity by comparing it with existing JEM and examining its predictive validity regarding known associations between exposures and health outcomes (eg, long working hours and strokes).

Development and internal validation of prediction model for rebleeding within one year after endoscopic treatment of cirrhotic varices: consideration from organ-based CT radiomics signature

BackgroundRebleeding after endoscopic treatment for esophagogastric varices (EGVs) in cirrhotic patients remains a significant clinical challenge, with high mortality rates and limited predictive tools. Current methods, relying on clinical indicators, often lack precision and fail to provide personalized risk assessments. This study aims to develop and validate a novel, non-invasive prediction model based on CT radiomics to predict rebleeding risk within one year of treatment, integrating radiomic features from key organs and clinical data.Methods123 patients were enrolled and divided into rebleeding (n = 44) and non-bleeding group (n = 79) within 1 year after endoscopic treatment of EGVs. The liver, spleen, and the lower part of the esophagus were segmented and the extracted radiomics features were selected to construct liver/spleen/esophagus radiomics signatures based on logistic regression. Clinic-radiomics combined models and multi-organ combined radiomics models were constructed based on independent model scores using logistic regression. The model performance was evaluated by ROC analysis, calibration and decision curves. The continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indices were analyzed.ResultsThe clinical-liver combined model had the highest AUC of 0.931 (95% CI: 0.887–0.974), which was followed by the liver-based model with AUC of 0.891 (95% CI: 0.835–0.74). The decision curves also showed that the clinical-liver combined model afforded a greater net benefit compared to other models within the threshold probability of 0.45 to 0.80. Significant improvements in discrimination (IDI, P < 0.05) and reclassification (NRI, P < 0.05) were obtained for clinical-liver combined model compared with the independent ones.ConclusionThe independent and combined liver-based CT radiomics models performed well in predicting rebleeding within 1 year after endoscopic treatment of EGVs.

Psychometric properties of the Persian version of the workplace fun scale among nurses: a cross-sectional study

Previous research has highlighted the importance of workplace fun in enhancing employee satisfaction and performance, particularly in high-stress professions like nursing. However, a notable gap exists in understanding how workplace fun is perceived and measured among nurses in Persian-speaking countries. This study addresses this gap by translating the Workplace Fun Scale and assessing its psychometric properties among nurses. The findings will provide insights into the scale’s applicability in these contexts and pave the way for healthcare organizations to significantly explore ways to enhance nurses’ enjoyment of work in Persian-speaking countries. This survey, which involved 321 nurses from medical education centers in Ardabil, was conducted with a meticulous and rigorous methodology. Standard questionnaires collected the data, including a demographic form and the translated Workplace Fun Scale. The study examined the three aspects of workplace fun: fun activities, coworker socializing, and manager support for fun. Confirmatory factor analysis was used to validate the structure, and reliability was assessed through retest coefficients, Cronbach’s alpha coefficients, and composite reliability coefficients. The data were analyzed using SPSS version 14 and LISREL version 8.8, ensuring the highest research standards. The validity of both form and content was confirmed through translation and reverse translation. The Workplace Fun Scale showed high internal consistency and reliability, with significant Cronbach’s alpha coefficients, composite reliability, and two-week retest coefficients of 0.859, 0.885, and 0.459, respectively (all at the p < 0.01 level). Fit indices, including GFI (0.97), AGFI (0.94), CFI (0.99), NFI (0.98), TLI (0.97), and SRMR (0.04), indicated a good fit of the measurement model, confirming the validity of the scale in assessing workplace fun. The confirmatory factor analysis results indicated that the translated version of the workplace fun scale, adapted from Tews, exhibited a robust factor structure and internal homogeneity within the Iranian sample. Furthermore, the scale demonstrated positive internal validity and reliability in Persian translation. These findings suggest that the scale possesses acceptable psychometric properties, making it a valuable tool for assessing workplace fun among nurses in Persian-speaking countries.

Deep Learning Predicts Lymphovascular Invasion Status in Muscle Invasive Bladder Cancer Histopathology.

Lymphovascular invasion (LVI) is linked to poor prognosis in patients with muscle-invasive bladder cancer (MIBC). Accurately identifying the LVI status in MIBC patients is crucial for effective risk stratification and precision treatment. We aim to develop a deep learning model to identify the LVI status in whole-slide images (WSIs) of MIBC patients. A cohort from The Cancer Genome Atlas (TCGA) database was used to train a deep learning model, slide-based lymphovascular invasion predictor (SBLVIP), based on multiple-instance learning. This model was externally validated using the Renmin Hospital of Wuhan University (RHWU) and People's Hospital of Hanchuan City (PHHC) cohorts. Kaplan-Meier curves, along with univariate and multivariate Cox models, were employed to evaluate the association between the LVI status predicted by SBLVIP and the survival outcomes of MIBC patients. In the TCGA cohort, the SBLVIP model achieved an average accuracy of 0.804 [95% confidence interval (CI) 0.712-0.895] and an area under the receiver operating characteristic curve (AUC) of 0.77 (95% CI 0.63-0.84) in the training set. In the internal validation set, the model's average accuracy and AUC were 0.774 (95% CI, 0.701-0.846) and 0.76 (95% CI, 0.60-0.83), respectively. In the RHWU cohort, the SBLVIP model achieved an average accuracy of 0.807 (95% CI 0.734-0.880) and an AUC of 0.74 (95% CI 0.55-0.83). In the PHHC cohort, SBLVIP demonstrated an average accuracy of 0.821 (95% CI 0.737-0.909) and an AUC of 0.74 (95% CI 0.58-0.89). Moreover, the LVI status predicted by SBLVIP showed significant independent prognostic value (P = 1 × 10-6). We developed a deep learning model named SBLVIP to predict the LVI status in routine WSIs of MIBC patients.

ABLE-SCORE: a simplified risk score for major adverse cardiovascular outcomes in patients with left ventricular noncompaction

Abstract Background Left ventricular noncompaction (LVNC) is a heterogeneous entity with life-threatening complications and variable prognosis [1,2]. However, there are limited prediction models available to identify individuals at high risk of adverse outcomes, and the current risk score in LVNC is comparatively complex for clinical practice [3,4]. Purpose To develop and validate a simplified risk score to predict major adverse cardiovascular events (MACE) in LVNC. Methods This multicenter longitudinal cohort study consecutively enrolled morphologically diagnosed LVNC patients between January 2009 and December 2020 at a single national-level medical center (derivation cohort n=300; internal validation cohort n=129), and between January 2014 and December 2022 at two national-level medical centers (external validation cohort n=95). The derivation/internal validation cohorts and the multicenter external validation cohort were followed annually until December 2022 and December 2023, respectively. MACE was defined as a composite of all-cause mortality, heart transplantation/left ventricular assist device implantation, cardiac resynchronization therapy, malignant ventricular arrhythmia, and thromboembolism. A simplified risk score, the ABLE-SCORE, was developed based on independent risk factors for MACE in the multivariable Cox regression predictive model, and underwent both internal and external validation to confirm its discrimination (Harrell’s C-index), calibration (calibration plots), and clinical applicability (decision curve analysis). Results A total of 524 LVNC patients (43.5 ± 16.6 years, 65.8% male) were included in the study. At the end of the follow-up, 97 (32.3%), 38 (29.5%), and 20 (21.1%) individuals in the derivation, internal validation, and external validation cohort experienced at least one endpoint of MACE. The ABLE-SCORE was established using four easily accessible clinical variables: age at diagnosis, N-terminal pro-brain natriuretic peptide levels, left atrium enlargement and left ventricular ejection fraction≤40% measured by echocardiography. The risk score showed excellent performance in discrimination, with Harrell’s C-index of 0.821 (95%CI 0.772-0.869), 0.786 (95%CI 0.703-0.869), and 0.750 (95%CI 0.644-0.856) in the derivation, internal validation, and external validation cohort, respectively. Calibration plots of the three datasets suggested accurate agreement between the predicted and observed 5-year risk of MACE in LVNC. According to decision curve analysis, the ABLE-SCORE displayed greater net benefit than the currently existing risk score for MACE in LVNC [3], indicating its strength in clinical applicability. Conclusions Derived from readily available laboratory and echocardiographic variables, a simplified and efficient risk score for MACE was developed and validated using a large LVNC cohort, making it a reliable and convenient tool for the risk stratification and clinical management of patients with LVNC.Development of the ABLE-SCOREValidation of the ABLE-SCORE

Predictive Models for Sustained, Uncontrolled Hypertension and Hypertensive Crisis Based on Electronic Health Record Data: Algorithm Development and Validation.

Assessing disease progression among patients with uncontrolled hypertension is important for identifying opportunities for intervention. We aim to develop and validate 2 models, one to predict sustained, uncontrolled hypertension (≥2 blood pressure [BP] readings ≥140/90 mm Hg or ≥1 BP reading ≥180/120 mm Hg) and one to predict hypertensive crisis (≥1 BP reading ≥180/120 mm Hg) within 1 year of an index visit (outpatient or ambulatory encounter in which an uncontrolled BP reading was recorded). Data from 142,897 patients with uncontrolled hypertension within Atrium Health Greater Charlotte in 2018 were used. Electronic health record-based predictors were based on the 1-year period before a patient's index visit. The dataset was randomly split (80:20) into a training set and a validation set. In total, 4 machine learning frameworks were considered: L2-regularized logistic regression, multilayer perceptron, gradient boosting machines, and random forest. Model selection was performed with 10-fold cross-validation. The final models were assessed on discrimination (C-statistic), calibration (eg, integrated calibration index), and net benefit (with decision curve analysis). Additionally, internal-external cross-validation was performed at the county level to assess performance with new populations and summarized using random-effect meta-analyses. In internal validation, the C-statistic and integrated calibration index were 0.72 (95% CI 0.71-0.72) and 0.015 (95% CI 0.012-0.020) for the sustained, uncontrolled hypertension model, and 0.81 (95% CI 0.79-0.82) and 0.009 (95% CI 0.007-0.011) for the hypertensive crisis model. The models had higher net benefit than the default policies (ie, treat-all and treat-none) across different decision thresholds. In internal-external cross-validation, the pooled performance was consistent with internal validation results; in particular, the pooled C-statistics were 0.70 (95% CI 0.69-0.71) and 0.79 (95% CI 0.78-0.81) for the sustained, uncontrolled hypertension model and hypertensive crisis model, respectively. An electronic health record-based model predicted hypertensive crisis reasonably well in internal and internal-external validations. The model can potentially be used to support population health surveillance and hypertension management. Further studies are needed to improve the ability to predict sustained, uncontrolled hypertension.

Interethnic validation of an artificial intelligence algorithm for prediction of atrial fibrillation using sinus rhythm electrocardiogram

Abstract Background Previous research has demonstrated acceptable diagnostic accuracy of artificial intelligence (AI)-enabled sinus rhythm (SR) electrocardiogram (ECG) interpretation for identifying paroxysmal atrial fibrillation (AF). However, interethnic validations of the AI algorithms have not been widely conducted. Purpose We aimed to develop our own AI model for the identification of paroxysmal AF based on SR ECGs in the Korean population and to validate its diagnostic performance in Brazilian citizens. Methods We trained a Transformer-based vision network on 90% of a dataset comprising 808,194 ECGs from 121,282 patients at Seoul National University Bundang Hospital (2003-2020). The remaining 10% of the dataset was used for internal validation. The model was trained to compute a risk score for paroxysmal AF or new-onset AF within 2 years. External validation was conducted using non-AF ECGs from the CODE 15% dataset provided by the Telehealth Network of Minas Gerais, Brazil. Results In internal validation, our AI model achieved an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.911 (95% CI: 0.902 - 0.921), with a sensitivity of 78.1% and a specificity of 89.0%. Subgroup analyses showed an AUROC of 0.893 (95% CI: 0.876 - 0.909) for patients in routine health checkups or outpatient settings, and 0.854 for patients with "Normal ECG" interpretations. In external validation with the CODE 15% dataset, the AI model exhibited an AUROC of 0.884 (95% CI: 0.869-0.900), which increased to 0.906 (95% CI: 0.893-0.919) when adjusted for age and sex (Figure 1). In the subset of patients with "Normal ECG" interpretations, the AUROC was 0.826 (95% CI: 0.769-0.883), increasing to 0.861 (95% CI: 0.814-0.908) after applying the same adjustments. Conclusions Our AI-powered SR ECG interpretation model demonstrated excellent diagnostic performance in predicting paroxysmal AF, with valid performance in the Brazilian population as well. This suggests that the model has potential for broad application across different ethnic groups.Figure 1