Area Under The ROC Curve Research Articles

Diagnostic accuracy of Mycobacterium tuberculosis-specific triple-color FluoroSpot assay in differentiating tuberculosis infection status in febrile patients with suspected tuberculosis.

This study aims to evaluate the diagnostic accuracy of a Mycobacterium tuberculosis (MTB)-specific triple-color FluoroSpot assay (IFN-γ/IL-2/TNF-α) in the differentiation of tuberculosis (TB) infection status in febrile patients. Febrile patients with suspected active TB (ATB) were consecutively enrolled. The frequencies and proportions of MTB-specific T cells secreting IFN-γ, IL-2, and TNF-α were detected at the single-cell level by triple-color FluoroSpot assay. The diagnostic index was fitted with a binary logistic regression model, and the diagnostic accuracy was evaluated according to the receiver operating characteristic (ROC) curve. The sensitivity, specificity, predictive values (PV), and likelihood ratios (LR) were calculated. A total of 210 febrile patients were enrolled, 53 patients were diagnosed with ATB (28 pathogen-confirmed vs. 25 clinically diagnosed) and 157 patients were non-ATB (84 with latent tuberculosis infection (LTBI) vs. 73 uninfected with MTB). Additionally, 30 pathogen-confirmed ATB patients were assembled. When diagnosing ATB, the area under the ROC curve (AUROC) of the MTB-specific triple-color FluoroSpot assay was significantly better than that of T-SPOT.TB (0.882 vs. 0.811, p = 0.017). With the fitted diagnostic index at a cutoff value of 0.378, the sensitivity, specificity, LR+, and LR- were 74.7%, 93.0%, 10.66, and 0.27, respectively. When differentiating ATB from LTBI, the AUROC of the FluoroSpot assay and T-SPOT.TB was 0.878 and 0.692, respectively (p < 0.001). With a diagnostic index of 0.413, the sensitivity, specificity, LR+, and LR were 77.1%, 85.7%, 5.40, and 0.27, respectively. The MTB-specific triple-color FluoroSpot (IFN-γ/IL-2/TNF-α) might be helpful for the differentiation of TB infection status in febrile patients.

A Novel Point-of-Care Prediction Model for Steatotic Liver Disease: Expected Role of Mass Screening in the Global Obesity Crisis.

The incidence of steatotic liver disease (SLD) is increasing across all age groups as the incidence of obesity increases worldwide. The existing noninvasive prediction models for SLD require laboratory tests or imaging and perform poorly in the early diagnosis of infrequently screened populations such as young adults and individuals with healthcare disparities. We developed a machine learning-based point-of-care prediction model for SLD that is readily available to the broader population with the aim of facilitating early detection and timely intervention and ultimately reducing the burden of SLD. We retrospectively analyzed the clinical data of 28,506 adults who had routine health check-ups in South Korea from January to December 2022. A total of 229,162 individuals were included in the external validation study. Data were analyzed and predictions were made using a logistic regression model with machine learning algorithms. A total of 20,094 individuals were categorized into SLD and non-SLD groups on the basis of the presence of fatty liver disease. We developed three prediction models: SLD model 1, which included age and body mass index (BMI); SLD model 2, which included BMI and body fat per muscle mass; and SLD model 3, which included BMI and visceral fat per muscle mass. In the derivation cohort, the area under the receiver operating characteristic curve (AUROC) was 0.817 for model 1, 0.821 for model 2, and 0.820 for model 3. In the internal validation cohort, 86.9% of individuals were correctly classified by the SLD models. The external validation study revealed an AUROC above 0.84 for all the models. As our three novel SLD prediction models are cost-effective, noninvasive, and accessible, they could serve as validated clinical tools for mass screening of SLD.

A combined diagnostic model including middle rectal artery visualization for predicting lateral lymph node metastasis in rectal cancer.

This study attempted to establish a combined diagnostic model encompassing visualization of the middle rectal artery (MRA) and other imaging features to improve the diagnostic efficiency of lateral lymph node (LLN) metastasis, which is crucial for clinical decision-making in rectal cancer. One hundred eleven patients receiving bilateral or unilateral lymph node dissection were enrolled, and 140 cases of LLN status on a certain unilateral pelvic sidewall were selected. Enhanced computed tomography (CT) was used to determine whether MRA was visible. Multivariable regression was used to establish a diagnostic model combining MRA visualization with other imaging features to predict LLN metastasis. Receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) were used to test the diagnostic efficacy for LLN metastasis. Ten-fold cross-validation was completed to internally validate the diagnostic model. Of the 140 LLNs harvested from 111 patients, 76 were positive and 64 were negative for metastases, respectively. The diagnostic model combining the MRA visualization and lymph node short diameter showed a greater efficiency than a single scale (AUC = 0.945, 95% confidence interval = 0.893-0.976, P < 0.001). The mean cross-validated AUC was 0.869 (95% confidence interval = 0.835-0.903). Our results establish a combined diagnostic model with the help of MRA visualization to yield a high diagnostic efficiency of LLN metastasis in rectal cancer.

Risk factors analysis of hypokalemia after radical resection of esophageal cancer and establishment of a nomogram risk prediction model.

This study aimed to explore the risk factors of hypokalemia after radical resection of esophageal cancer (EC) and establish a nomogram risk prediction model to evaluate hypokalemia risk after esophagectomy. Thus, this study provides a reference for the clinical development of intervention measures. Clinical data of EC patients who underwent radical surgery from January 2020 to November 2022 in the First Affiliated Hospital of Guangxi Medical University were retrospectively collected. The relevant variables were screened using multivariate logistic regression analysis with IBM SPSS 25.0 and R 4.2.0 software, and a nomogram for predicting hypokalemia risk was established. The established nomogram was evaluated by receiver operating characteristic (ROC), calibration, and decision curves. The model was also internally validated by 1000 bootstrap resampling methods. After radical EC resection, the incidence rate of hypokalemia in 213 patients was 19.2% (41/213). The hemoglobin levels, total serum protein, serum albumin, calcium ion concentration, direct bilirubin, prothrombin time (PT), and activated partial thromboplastin time (APTT) were related (p < 0.05). The multivariate logistic analysis showed that the white blood cell count, serum albumin level, direct bilirubin, and operation time were risk factors for hypokalemia after radical EC resection (p < 0.05). The area under the ROC curve (AUC) was 0.764, demonstrating the good discriminative ability of the established nomogram for hypokalemia prediction. The calibration curve showed a good fit between the predicted and actual observed probabilities. The model maintained a high C-index in the internal validation (C-index = 0.758), supporting that the nomogram can be widely used for hypokalemia prediction. The prediction model for hypokalemia risk with individualized scores based on the patient's white blood cell count, serum albumin level, direct bilirubin, and operation time can screen out high-risk patients who might develop hypokalemia. It is of certain reference value for clinicians to screen and follow up with patients with emphasis and to formulate preoperative and postoperative intervention strategies.

Involvement of the posterior limb of the internal capsule independently predicts the prognosis of patients with basal ganglia and thalamic hemorrhage.

Intracerebral hemorrhage (ICH) is the most lethal and devastating subtype of stroke. Basal ganglia hemorrhage and thalamic hemorrhage are the most common types of ICH, accounting for 50-70% of all ICH cases, leading to disability and death, and it involves the posterior limb of the internal capsule to varying degrees. In this study, we investigated the impact of varying degrees of the involvement of the posterior limb of the internal capsule on the prognosis of patients with basal ganglia and thalamic ICH and assessed whether it improves the predictive accuracy of the max-ICH score, an existing scale for ICH functional outcome. This is a multicenter, retrospective, observational study. We graded the involvement of the posterior limb of the internal capsule according to the degree of compression and injury (called iICH, ranging from 0 to 4). An unfavorable outcome was defined as a 90-day modified Rankin Scale (mRS) of > 2. Multivariate logistic regression analysis was used to identify independent risk factors associated with unfavorable prognosis. The discrimination was verified using receiver operating characteristic curve (ROC) analysis, while the calibration was verified by the Hosmer-Lemeshow test. Of the 305 patients included, 188 from Nanfang Hospital were assigned to the development cohort, and 117 from Heyuan People's Hospital and Huadu District People's Hospital were assigned to the validation cohort. In the development cohort, iICH was identified as an independent factor of a 90-day unfavorable outcome, and the area under the ROC (AUC) was 0.774. When combined with the iICH, the AUC of max-ICH was significantly elevated from 0.816 to 0.866. Comparable results were found in the validation cohort. Increased involvement of the posterior limb of the internal capsule is associated with a worse outcome in patients with basal ganglia and thalamic ICH.

Plasma Epstein-Barr Virus DNA load for diagnostic and prognostic assessment in intestinal Epstein-Barr Virus infection.

The prospective application of plasma Epstein-Barr virus (EBV) DNA load as a noninvasive measure of intestinal EBV infection remains unexplored. This study aims to identify ideal threshold levels for plasma EBV DNA loads in the diagnosis and outcome prediction of intestinal EBV infection, particularly in cases of primary intestinal lymphoproliferative diseases and inflammatory bowel disease (IBD). Receiver operating characteristic (ROC) curves were examined to determine suitable thresholds for plasma EBV DNA load in diagnosing intestinal EBV infection and predicting its prognosis. 108 patients were retrospectively assigned to the test group, while 56 patients were included in the validation group. Plasma EBV DNA loads were significantly higher in the intestinal EBV infection group compared to the non-intestinal EBV infection group (Median: 2.02 × 102 copies/mL, interquartile range [IQR]: 5.49 × 101-6.34×103 copies/mL versus 4.2×101 copies/mL, IQR: 1.07 ×101-6.08×101 copies/mL; P < 0.0001). Plasma EBV DNA levels at 9.21×101 and 6.77×101 copies/mL proved beneficial for the identification and prognostication in intestinal EBV infection, respectively. Values of 0.82 and 0.71 were yielded by the area under the ROC curve (AUC) in the test cohort, corresponding to sensitivities of 84.38% (95% confidence interval [95%CI]: 68.25%-93.14%) and 87.5% (95%CI: 69%-95.66%), specificities of 83.33% (95%CI: 64.15%-93.32%) and 68.09% (95%CI: 53.83%-79.6%), positive predictive values (PPV) of 87.1% (95%CI: 71.15%-94.87%) and 58.33% (95%CI: 42.2%-72.86%), and positive likelihood ratios (LR+) of 5.06 and 2.74 in the validation cohort, respectively. Furthermore, a plasma EBV DNA load of 5.4×102 copies/mL helped differentiate IBD with intestinal EBV infection from primary intestinal EBV-positive lymphoproliferative disorders (PIEBV+LPDs), achieving an AUC of 0.85 within the test cohort, as well as 85% sensitivity (95%CI: 63.96%-94.76%), 91.67% specificity (95%CI: 64.61%-99.57%), 94.44% PPV (95%CI: 74.24%-99.72%), and an LR+ of 10.2 in the validation cohort. Plasma EBV DNA load demonstrates notable potential in distinguishing between different patient cohorts with intestinal EBV infection, although its sensitivity requires further optimization for clinical application.

Nomogram for predicting early shoulder joint dysfunction after breast cancer surgery

AbstractObjectiveThis study aims to construct a clinical risk profile nomogram model for predicting early shoulder joint dysfunction (SJD) after breast cancer surgery.MethodsUsing a convenience sampling method, the clinical data of 161 breast cancer patients between February 2022 and July 2023 at Affiliated Cancer Hospital of Nanjing Medical University were selected and analyzed retrospectively. Risk factors were identified using univariate and multivariate logistic regression analyses. The R software was used to construct the risk prediction model and to plot the nomogram for early SJD post‐breast cancer surgery. The model's predictive performance was evaluated using the receiver operating characteristic (ROC) curve and the Hosmer–Lemeshow test.ResultsAmong the 161 patients, 104 (64.6%) experienced SJD. Multivariate logistic regression analysis finally included the functional exercise compliance scale for postoperative breast cancer patients (FECSPBCP), the compliance scale of physical exercise, the body mass index of patients, involved side to‐hand dominance, and the operation mode of breast and lymph nodes into the model. The area under the ROC curve (AUC) was 0.785 (95% confidence interval: 0.711–0.860), indicating a good model fit as confirmed by the Hosmer–Lemeshow test (X2 = 3.2891, p = .9149). Internal validation using the Bootstrap resampling method (n = 1000) yielded an AUC of 0.731.ConclusionsThe incidence of early SJD was high among postoperative breast cancer patients. The constructed risk prediction model can assist medical professionals in the early identification of high‐risk individuals and provide targeted interventions to prevent long‐term disabilities.

Machine learning prediction of anxiety symptoms in social anxiety disorder: utilizing multimodal data from virtual reality sessions.

Machine learning (ML) is an effective tool for predicting mental states and is a key technology in digital psychiatry. This study aimed to develop ML algorithms to predict the upper tertile group of various anxiety symptoms based on multimodal data from virtual reality (VR) therapy sessions for social anxiety disorder (SAD) patients and to evaluate their predictive performance across each data type. This study included 32 SAD-diagnosed individuals, and finalized a dataset of 132 samples from 25 participants. It utilized multimodal (physiological and acoustic) data from VR sessions to simulate social anxiety scenarios. This study employed extended Geneva minimalistic acoustic parameter set for acoustic feature extraction and extracted statistical attributes from time series-based physiological responses. We developed ML models that predict the upper tertile group for various anxiety symptoms in SAD using Random Forest, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost) models. The best parameters were explored through grid search or random search, and the models were validated using stratified cross-validation and leave-one-out cross-validation. The CatBoost, using multimodal features, exhibited high performance, particularly for the Social Phobia Scale with an area under the receiver operating characteristics curve (AUROC) of 0.852. It also showed strong performance in predicting cognitive symptoms, with the highest AUROC of 0.866 for the Post-Event Rumination Scale. For generalized anxiety, the LightGBM's prediction for the State-Trait Anxiety Inventory-trait led to an AUROC of 0.819. In the same analysis, models using only physiological features had AUROCs of 0.626, 0.744, and 0.671, whereas models using only acoustic features had AUROCs of 0.788, 0.823, and 0.754. This study showed that a ML algorithm using integrated multimodal data can predict upper tertile anxiety symptoms in patients with SAD with higher performance than acoustic or physiological data obtained during a VR session. The results of this study can be used as evidence for personalized VR sessions and to demonstrate the strength of the clinical use of multimodal data.

An interpretable machine learning model for predicting in-hospital mortality in ICU patients with ventilator-associated pneumonia.

Ventilator-associated pneumonia (VAP) is a common nosocomial infection in ICU, significantly associated with poor outcomes. However, there is currently a lack of reliable and interpretable tools for assessing the risk of in-hospital mortality in VAP patients. This study aims to develop an interpretable machine learning (ML) prediction model to enhance the assessment of in-hospital mortality risk in VAP patients. This study extracted VAP patient data from versions 2.2 and 3.1 of the MIMIC-IV database, using version 2.2 for model training and validation, and version 3.1 for external testing. Feature selection was conducted using the Boruta algorithm, and 14 ML models were constructed. The optimal model was identified based on the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, and specificity across both validation and test cohorts. SHapley Additive exPlanations (SHAP) analysis was applied for global and local interpretability. A total of 1,894 VAP patients were included, with 12 features ultimately selected for model construction: 24-hour urine output, blood urea nitrogen, age, diastolic blood pressure, platelet count, anion gap, body temperature, bicarbonate level, sodium level, body mass index, and whether combined with congestive heart failure and cerebrovascular disease. The random forest (RF) model showed the best performance, achieving an AUC of 0.780 in internal validation and 0.724 in external testing, outperforming other ML models and common clinical scoring systems. The RF model demonstrated robust and reliable performance in predicting in-hospital mortality risk for VAP patients. The developed online tool can assist clinicians in efficiently assessing VAP in-hospital mortality risk, supporting clinical decision-making.

External validation of an artificial intelligence model using clinical variables, including ICD-10 codes, for predicting in-hospital mortality among trauma patients: a multicenter retrospective cohort study

Artificial intelligence (AI) is being increasingly applied in healthcare to improve patient care and clinical outcomes. We previously developed an AI model using ICD-10 (International Classification of Diseases, Tenth Revision) codes with other clinical variables to predict in-hospital mortality among trauma patients from a nationwide database. This study aimed to externally validate the performance of the AI model. Validation was conducted using a multicenter retrospective cohort study design, analyzing patient data from January 2020 to December 2021. The study included trauma patients based on specific ICD-10 codes, with other clinical variables. The performance of the AI model was evaluated against conventional metrics, including the ISS, and the ICISS (ICD-based ISS), using sensitivity, specificity, accuracy, balanced accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUROC) analyses. Data from 4,439 patients were analyzed. The AI model demonstrated high overall performance, achieving an AUROC of 0.9448 and a balanced accuracy of 85.08%, thereby outperforming traditional scoring systems such as ISS, or ICISS. Furthermore, the model accurately predicted mortality across datasets from each hospital (AUROCs of 0.9234 and 0.9653, respectively) despite significant differences in hospital characteristics. In the subset of patients with ISS < 9, the model showed a robust AUROC of 0.9043, indicating its effectiveness in predicting mortality, even in cases with lower-severity injuries. For patients with ISSs ≥ 9, the model maintained high sensitivity (93.60%) and balanced accuracy (77.08%), proving its reliability in more severe injury cases. External validation demonstrated the AI model’s high predictive accuracy and reliability in assessing in-hospital mortality risk among trauma patients across different injury severities and heterogeneous cohorts. These findings support the model’s potential integration into emergency departments and offer a significant tool for enhancing patient triage and treatment protocols.

A novel nomogram for predicting mortality risk in young and middle-aged patients undergoing maintenance hemodialysis: a retrospective study.

The annual growth in the population of maintenance hemodialysis (MHD) patients is accompanied by a trend towards younger age groups among new cases. Despite the escalating mortality risk observed in MHD patients, there remains a dearth of research focused on young and middle-aged individuals in this cohort, leading to a deficiency in specialized predictive instruments for this demographic. This research seeks to explore the critical determinants impacting mortality risk in young and middle-aged MHD patients and to construct a prediction model accordingly. This study involved 127 young and middle-aged patients undergoing MHD in the Blood Purification Center of Chaohu Hospital of Anhui Medical University from January 2019 to January 2022. The follow-up period for each patient ended either at the time of death or on January 31, 2024. Participants were monitored to determine their survival status and categorized into two groups: those who survived (98 patients) and those who deceased (29 patients). Clinical data were gathered for analysis. Logistic regression was utilized to pinpoint independent risk factors for mortality among these patients. Subsequently, a nomogram was established to predict mortality risk. The efficacy of this model was assessed through the area under the receiver operating characteristic curve (AUC-ROC), alongside a calibration curve and the Hosmer-Lemeshow test to examine its fit. Additionally, decision curve analysis (DCA) was conducted to ascertain the clinical relevance of the predictive model. The study incorporated 127 young and middle-aged patients undergoing MHD, with a mortality rate recorded at 22.83% (29 cases). A logistic regression analysis revealed that age, hemoglobin (HB), serum magnesium (Mg), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-albumin ratio (PAR) were significant independent predictors of mortality among these patients. Utilizing these variables, a nomogram was developed to predict mortality risk, achieving an AUC of 0.899 (95% CI: 0.833-0.966). The model exhibited a specificity of 83.67% and a sensitivity of 82.76%, demonstrating substantial discriminative ability. The model's robustness was confirmed through internal validation with 1,000 bootstrap samples, yielding an AUC of 0.894 (95% CI: 0.806-0.949). The calibration curve closely aligned with the ideal curve, and the Hosmer-Lemeshow goodness-of-fit test yielded a χ 2 value of 6.312 with a p-value of 0.612, verifying the model's high calibration accuracy. Additionally, the DCA indicated that the model provides a net benefit across a wide range of decision thresholds from 0 to 0.99, underscoring its clinical utility. The nomogram developed from variables including age, HB levels, serum Mg, NLR, and PAR exhibits high levels of discrimination and calibration. This model effectively predicts mortality risk among young and middle-aged patients undergoing MHD, proving its clinical relevance.

Major Adverse Kidney Events in Hospitalized Older Patients With Acute Kidney Injury: Machine Learning-Based Model Development and Validation Study.

Acute kidney injury (AKI) is a common complication in hospitalized older patients, associated with increased morbidity, mortality, and health care costs. Major adverse kidney events within 30 days (MAKE30), a composite of death, new renal replacement therapy, or persistent renal dysfunction, has been recommended as a patient-centered endpoint for clinical trials involving AKI. This study aimed to develop and validate a machine learning-based model to predict MAKE30 in hospitalized older patients with AKI. A total of 4266 older patients (aged ≥ 65 years) with AKI admitted to the Second Xiangya Hospital of Central South University from January 1, 2015, to December 31, 2020, were included and randomly divided into a training set and an internal test set in a ratio of 7:3. An additional cohort of 11,864 eligible patients from the Medical Information Mart for Intensive Care Ⅳ database served as an external test set. The Boruta algorithm was used to select the most important predictor variables from 53 candidate variables. The eXtreme Gradient Boosting algorithm was applied to establish a prediction model for MAKE30. Model discrimination was evaluated by the area under the receiver operating characteristic curve (AUROC). The SHapley Additive exPlanations method was used to interpret model predictions. The overall incidence of MAKE30 in the 2 study cohorts was 28.3% (95% CI 26.9%-29.7%) and 26.7% (95% CI 25.9%-27.5%), respectively. The prediction model for MAKE30 exhibited adequate predictive performance, with an AUROC of 0.868 (95% CI 0.852-0.881) in the training set and 0.823 (95% CI 0.798-0.846) in the internal test set. Its simplified version achieved an AUROC of 0.744 (95% CI 0.735-0.754) in the external test set. The SHapley Additive exPlanations method showed that the use of vasopressors, mechanical ventilation, blood urea nitrogen level, red blood cell distribution width-coefficient of variation, and serum albumin level were closely associated with MAKE30. An interpretable eXtreme Gradient Boosting model was developed and validated to predict MAKE30, which provides opportunities for risk stratification, clinical decision-making, and the conduct of clinical trials involving AKI. Chinese Clinical Trial Registry ChiCTR2200061610; https://tinyurl.com/3smf9nuw.

A machine learning model to predict neurological deterioration after mild traumatic brain injury in older adults.

Neurological deterioration after mild traumatic brain injury (TBI) has been recognized as a poor prognostic factor. Early detection of neurological deterioration would allow appropriate monitoring and timely therapeutic interventions to improve patient outcomes. In this study, we developed a machine learning model to predict the occurrence of neurological deterioration after mild TBI using information obtained on admission. This was a retrospective cohort study of data from the Think FAST registry, a multicenter prospective observational study of elderly TBI patients in Japan. Patients with an admission Glasgow Coma Scale (GCS) score of 12 or below or who underwent surgical treatment immediately upon admission were excluded. Neurological deterioration was defined as a decrease of 2 or more points from a GCS score of 13 or more within 24 h of hospital admission. The model predictive accuracy was judged with the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC), and the Youden index was used to determine the cutoff value. A total of 421 of 721 patients registered in the Think FAST registry between December 2019 and May 2021 were included in our study, among whom 25 demonstrated neurological deterioration. Among several machine learning algorithms, eXtreme Gradient Boosting (XGBoost) demonstrated the highest predictive accuracy in cross-validation, with an AUROC of 0.81 (±0.07) and an AUPRC of 0.33 (±0.08). Through SHapley Additive exPlanations (SHAP) analysis, five important features (D-dimer, fibrinogen, acute subdural hematoma thickness, cerebral contusion size, and systolic blood pressure) were identified and used to construct a better performing model (cross-validation AUROC of 0.84 and AUPRC of 0.34; testing data AUROC of 0.77 and AUPRC of 0.19). At the cutoff value from the Youden index, the model showed a sensitivity, specificity, and positive predictive value of 60, 96, and 38%, respectively. When neurosurgeons attempted to predict neurological deterioration using the same testing data, their values were 20, 94, and 19%, respectively. In this study, our predictive model showed an acceptable performance in detecting neurological deterioration after mild TBI. Further validation through prospective studies is necessary to confirm these results.

Identifying psychosocial predictors and developing a risk score for preterm birth among Kenyan pregnant women

BackgroundPreterm birth (PTB) is a leading cause of neonatal mortality, particularly in sub-Saharan Africa where 40% of global neonatal deaths occur. We identified and combined demographic, clinical, and psychosocial correlates of PTB among Kenyan women to develop a risk score.MethodsWe used data from a prospective study enrolling HIV-negative women from 20 antenatal clinics in Western Kenya (NCT03070600). Depressive symptoms were assessed by study nurses using the Center for Epidemiologic Studies Depression Scale (CESD-10), intimate partner violence (IPV) with the Hurt, Insult, Threaten, Scream scale (HITS), and social support using the Medical Outcomes Survey scale (MOS-SSS). Predictors of PTB (birth < 37 weeks gestation) were identified using multivariable Cox proportional hazards models, clustered by facility. We used stratified k-fold cross-validation methods for risk score derivation and validation. Area under the receiver operating characteristic curve (AUROC) was used to evaluate discrimination of the risk score and Brier score for calibration.ResultsAmong 4084 women, 19% had PTB (incidence rate: 70.9 PTB per 100 fetus-years (f-yrs)). Predictors of PTB included being unmarried (HR:1.29, 95% CI:1.08–1.54), lower education (years) (HR:0.97, 95% CI:0.94–0.99), IPV (HITS score ≥ 5, HR:1.28, 95% CI:0.98–1.68), higher CESD-10 score (HR:1.02, 95% CI:0.99–1.04), lower social support score (HR:0.99, 95% CI:0.97–1.01), and mild-to-severe depressive symptoms (CESD-10 score ≥ 5, HR:1.46, 95% CI:1.07–1.99). The final risk score included being unmarried, social support score, IPV, and MSD. The risk score had modest discrimination between PTB and term deliveries (AUROC:0.56, 95% CI:0.54–0.58), and Brier Score was 0.4672. Women considered “high risk” for PTB (optimal risk score cut-point) had 40% higher risk of PTB (83.6 cases per 100 f-yrs) than “low risk” women (59.6 cases per 100 f-ys; HR:1.6, 95% CI:1.2–1.7, p < 0.001).ConclusionA fifth of pregnancies were PTB in this large multi-site cohort; PTB was associated with several social factors amenable to intervention. Combining these factors in a risk score did not predict PTB, reflecting the multifactorial nature of PTB and need to include other unmeasured factors. However, our findings suggest PTB risk could be better understood by integrating mental health and support services into routine antenatal care.

Correlation analysis of the triglyceride–glucose index and related parameters in metabolic dysfunction-associated fatty liver disease

This study aimed to investigate the correlation and predictive value of TyG and related parameters with metabolic dysfunction-associated fatty liver disease (MAFLD) MAFLD. This study retrospectively included individuals who underwent health examinations and abdominal ultrasound from July 2021 to June 2024 at the Affiliated Hospital of Southwest Medical University, Sichuan Province, China. A total of 71,299 subjects’ clinical and laboratory data were extracted, the correlation between TyG and related parameters and MAFLD was analyzed via univariate and multivariate logistic regression methods, and the nonlinear relationship between the TyG index and the risk of MAFLD was explored via restricted cubic spline (RCS) analysis. The predictive value of TyG and related parameters for MAFLD was assessed using the receiver operating characteristic (ROC) curves and the area under the ROC curve (AUC). TyG and related parameters were positively correlated with MAFLD, and the results remained unchanged after adjustment for the corresponding parameters. RCS analysis revealed a significant dose‒response relationship between TyG and related parameters and MAFLD. ROC curve analysis revealed AUC values of 0.83 (0.82–0.83), 0.92 (0.91–0.92), 0.90 (0.90–0.91), and 0.87 (0.87–0.88) for TyG, TyG-BMI, TyG-WC, and TyG-WHR, respectively. Subgroup analyses revealed that the TyG index and related parameters had greater predictive value in the female, younger, and BMI < 23.7 populations.

Association between the triglyceride-glucose index and liver fibrosis in adults with metabolism-related fatty liver disease in the United States: a cross-sectional study of NHANES 2017–2020

ObjectiveThis study aimed to examine the association between the triglyceride-glucose (TyG) index and liver fibrosis (LF) in U.S. adults with Metabolic Dysfunction-Associated Steatotic Liver Disease (MAFLD).MethodsUsing data from the 2017 to 2020 National Health and Nutrition Examination Survey (NHANES) database, we conducted a population-based cross-sectional study with 1,324 participants. MAFLD was defined as a controlled attenuation parameter (CAP) score ≥ 248 dB/m accompanied by metabolic dysfunction. A median liver stiffness measurement ≥ 8.2 kPa was used to identify significant fibrosis (≥ F2). Multivariable logistic regression was employed to assess the impact of the TyG index on LF outcomes. A restricted cubic spline (RCS) model was used to explore nonlinear effects, and receiver operating characteristic (ROC) curves were applied to evaluate the effectiveness in predicting.ResultsAmong the participants, 716 were men and 608 were women, aged 20 to 80 years, representing various racial groups. Significant fibrosis was observed in 137 out of 1,324 participants. After adjusting for confounding factors, a higher TyG index was significantly associated with an increased incidence of MAFLD-related LF (OR = 2.18, 95% CI, 1.14–4.18; p < 0.05). Elevated TyG levels showed a positive correlation with significant fibrosis, with an odds ratio (OR) exceeding 1 when the TyG index was above 8.054. Subgroup analyses stratified by sex, age, and body mass index (BMI) revealed differences after adjusting for confounders. The association was stronger in women (OR = 2.53, 95% CI, 1.16–5.53) than in men (OR = 1.95, 95% CI, 0.81–4.72). A significant correlation was also found between TyG levels and obesity status (overweight: OR = 4.80, 95% CI, 1.27–18.2; obese: OR = 2.26, 95% CI, 1.20–5.53). In MAFLD patients aged 40–59, TyG was strongly associated with LF (OR = 2.85, 95% CI, 1.16–6.79). Furthermore, the area under the ROC curve (AUC) for the TyG index in predicting significant fibrosis in MAFLD patients was 0.73 (95% CI, 0.68–0.78), indicating moderate predictive ability.ConclusionsIn the general U.S. population, elevated TyG index levels were positively associated with an increased risk of LF in MAFLD patients.

Integration of metabolomics methodologies for the development of predictive models for mortality risk in elderly patients with severe COVID-19

BackgroundThe rapid evolution of the COVID-19 pandemic and subsequent global immunization efforts have rendered early metabolomics studies potentially outdated, as they primarily involved non-exposed, non-vaccinated populations. This paper presents a predictive model developed from up-to-date metabolomics data integrated with clinical data to estimate early mortality risk in critically ill COVID-19 patients. Our study addresses the critical gap in current research by utilizing current patient samples, providing fresh insights into the pathophysiology of the disease in a partially immunized global population.MethodsOne hundred elderly patients with severe COVID-19 infection, including 46 survivors and 54 non-survivors, were recruited in January-February 2023 at the Second Hospital affiliated with Harbin Medical University. A predictive model within 24 h of admission was developed using blood metabolomics and clinical data. Differential metabolite analysis and other techniques were used to identify relevant characteristics. Model performance was assessed by comparing the area under the receiver operating characteristic curve (AUROC). The final prediction model was externally validated in a cohort of 50 COVID-19 elderly critically ill patients at the First Hospital affiliated with Harbin Medical University during the same period.ResultsSignificant disparities in blood metabolomics and laboratory parameters were noted between individuals who survived and those who did not. One metabolite indicator, Itaconic acid, and four laboratory tests (LYM, IL-6, PCT, and CRP), were identified as the five variables in all four models. The external validation set demonstrated that the KNN model exhibited the highest AUC of 0.952 among the four models. When considering a 50% risk of mortality threshold, the validation set displayed a sensitivity of 0.963 and a specificity of 0.957.ConclusionsThe prognostic outcome of COVID-19 elderly patients is significantly influenced by the levels of Itaconic acid, LYM, IL-6, PCT, and CRP upon admission. These five indicators can be utilized to assess the mortality risk in affected individuals.

Comparison of cardiovascular risk prediction models developed using machine learning based on data from a Sri Lankan cohort with World Health Organization risk charts for predicting cardiovascular risk among Sri Lankans: a cohort study

IntroductionModels derived from non-Sri Lankan cohorts are used for cardiovascular (CV) risk stratification of Sri Lankans.ObjectiveTo develop a CV risk prediction model using machine learning (ML) based on data from...

Risk Factors Analysis of Cutaneous Adverse Drug Reactions Caused by Targeted Therapy and Immunotherapy Drugs for Oncology and Establishment of a Prediction Model.

Targeted therapy and immunotherapy drugs for oncology have greater efficacy and tolerability than cytotoxic chemotherapeutic drugs. However, the cutaneous adverse drug reactions associated with these newer therapies are more common and remain poorly predicted. An effective prediction model is urgently needed and essential. This retrospective study included 1052 patients, divided into train set, test set, and external validation set. As a data-driven study, a total of 76 variables were collected. Univariate logistic analysis, least absolute shrinkage and selection operator regression, and stepwise logistic regression were utilized for feature screening. Finally, nine machine-learning models were constructed and compared, and grid search was performed to adjust the parameters. Model performance was evaluated using calibration curve and the area under the receiver operating characteristic curve (AUROC). Nine risk factors were eventually identified: age, treatment modality, cancer types, history of allergies, age-corrected Charlson comorbidity index, percentage of eosinophils, absolute number of monocytes, Eastern Cooperative Oncology Group Performance Status, and C-reactive protein. Among the models, the logistic model performed best, demonstrating strong performance in test set (AUROC = 0.734) and external validation set (AUROC = 0.817). This study identified nine significant risk factors and developed a nomogram prediction model. These findings have important implications for optimizing therapeutic efficacy and maintaining the quality of life of patients from the perspective of managing cutaneous adverse drug reactions. Trial Registration: ChiCTR2400088422.

Diagnostic and Predictive Recurrence Value of Plasma Fibrinogen in Patients With Adrenocortical Carcinoma.

The correlation between fibrinogen levels and adrenocortical carcinoma (ACC) remains unclear. This study aimed to explore the value of preoperative plasma fibrinogen as a biomarker for ACC. We identified 40 patients with ACC and 170 patients with adrenal adenoma (AA) who underwent surgery at our institution between 2015 and 2022. Plasma fibrinogen levels and postoperative tumor recurrence information of the patients were also recorded. For intergroup comparisons, data obtained from the AA and ACC groups were evaluated using a t-test. The cutoff value of fibrinogen level was determined using a receiver operating characteristic (ROC) curve. Mean fibrinogen levels in the AA and ACC groups were 2.81 ± 0.59 g/L and 3.88 ± 1.75 g/L, respectively (P < .001). Fibrinogen level, which can help distinguish between AA and ACC, was evaluated using the ROC curve. The cutoff fibrinogen level was estimated as 3.87 g/L according to the Youden index. With this value, the sensitivity was 62.5%, specificity was 95.7%, and the area under the ROC curve (AUC) was 0.74 (P < .001). Fibrinogen level, which can help distinguish between recurrence and non-recurrence, was evaluated using the ROC curve. The cutoff fibrinogen level was estimated as 3.96 g/L according to the Youden index. The sensitivity, specificity, and AUC were 90%, 71.4%, and 0.85, respectively (P < .001). According to the data in this study, plasma fibrinogen could be used to distinguish ACC from AA. Most importantly, plasma fibrinogen may be used to identify recurrence of postoperative ACC.