Receiver Operating Characteristic Research Articles

A nomogram to predict “pure” vs. “mixed” uric acid urinary stones

PurposeUric acid stones (UAS) can be treated non-invasively by oral chemolysis. However, it is crucial to identify individuals who are most likely to benefit from this approach, specifically, patients with pure UAS. The aim of this study was to develop a nomogram that can differentiate between pure and mixed UAS.MethodsA retrospective analysis of demographic, clinical and stone composition data of patients with a predominant UAS composition (≥ 50%) treated between 2014 and 2022.ResultsA total of 135 patients were included in the analysis, 37.8% had mixed UAS (50–90% UA) and 62.2% had pure UAS (≥ 95% UA). The mean stone density and the percentage of radiopaque stones in the pure UAS group were significantly lower than those in the mixed UAS group (450 Hounsfield Units [HU] vs. 600 HU, and 24% vs. 58%, respectively). A stepwise multivariate logistic regression revealed that lower stone density, bigger size, decreased stone opacity and older age are predictive variables for pure UAS. Accordingly, a nomogram was generated with a receiver operating characteristic (ROC) curve that showed an area under the curve (AUC) of 0.78. A patient with a total score of 156 has a probability of > 95% for pure UAS.ConclusionImaging and demographic data can be used to identify patients with pure UAS. The nomogram may be useful for counseling patients regarding oral chemolysis. Future validation of the nomogram with a different data set is required to assess its efficacy.

The clinical significance of the Neutrophil-to-Lymphocyte Ratio as a novel inflammatory biomarker for assessing the severity of intervertebral disc degeneration.

Inflammation is integral to the pathogenesis of intervertebral disc degeneration, yet the role of systemic inflammatory markers in this process remains underexplored. This study aims to explore the association between the Neutrophil-to-Lymphocyte Ratio (NLR) and the severity of disc degeneration. A retrospective analysis was conducted on 375 patients diagnosed with lumbar disc degeneration between April 2018 and May 2021. All patients underwent a complete blood cell count examination. We applied the Pfirrmann grading system for cumulative disc grading, stratifying patients into two groups: a high-score group (cumulative grade > 17) and a low-score group (cumulative grade ≤ 17), based on the median cumulative grade. The association between the NLR and and the severity of disc degeneration was further analyzed using correlation analysis and logistic regression models. Furthermore, the predictive capacity of the NLR for lumbar disc degeneration was assessed using the Receiver Operating Characteristic (ROC) curve. We found a significant positive correlation between high NLR levels and severe disc degeneration. The high-score group exhibited a significantly higher NLR compared to the low-score group [2.63 (1.91-4.18) vs. 2.04 (1.38-2.74), respectively, p < 0.001]. Significant correlations were found between NLR and patient characteristics (including age, BMI, VAS, NSAIDs usage, hemoglobin) and the cumulative grading. Logistic regression analysis identified age and NLR as independent predictors of the severity of disc degeneration. The ROC curve analysis demonstrated a good predictive capability of NLR for lumbar disc degeneration. NLR could serve as a promising biomarker for assessing the severity of lumbar disc degeneration and offer potential benefits in both early diagnosis and treatment strategies.

Predicting new-onset persistent conduction disturbance following transcatheter aortic valve replacement: the usefulness of FEOPS finite element analysis

BackgroundDespite the frequency of persistent new-onset conduction disturbances after transcatheter aortic valve replacement (TAVR), few preoperative methods of prediction exist.MethodsPatients who underwent TAVR in the Department of Cardiology of the Second Affiliated Hospital of the Army Medical University from December 2020 to September 2021 and postoperative aortic root modeling via the FEOPS finite element analysis were included in this single-center case–control study, divided into persistent conduction disturbances (PCD) and non-PCD groups according to their pre- and postoperative electrocardiograms in the first month. Risk factors affecting PCD were identified by comparing the baseline data of these two groups, including echocardiograms, computed tomography angiography of the aortic root, surgical decision-making, and FEOPS data. Independent risk factors were screened using logistic regression modeling, and the receiver operating characteristic (ROC) curve was used to test the predictive ability.ResultsA total of 56 patients were included in this study, 37 with bicuspid aortic valve (BAV) and 19 with trileaflet aortic valve (TAV), with 17 cases of PCD. The contact pressure index (CPI) of FEOPS, valve oversize ratio, differences between membranous interventricular septum length and implantation depth (ΔMSID) and valve implantation depth were statistically different (P < 0.05). CPI could be used as an independent risk factor for PCD (P < 0.05), and the ROC curve comparison showed that the CPI was more predictive (AUC = 0.806, 95% CI: 0.684–0.928, P = 0.001).ConclusionsThe CPI of FEOPS has better predictive value for new-onset conduction disturbance after TAVR compared to other known predictors.

Predicting major adverse cardiovascular events in angina patients using radiomic features of pericoronary adipose tissue based on CCTA.

This study aims to evaluate whether radiomic features of pericoronary adipose tissue (PCAT) derived from coronary computed tomography angiography (CCTA) can better predict major adverse cardiovascular events (MACE) in patients with angina pectoris. A single-center retrospective study included 239 patients with angina pectoris who underwent coronary CT examinations. Participants were divided into MACE (n = 46) and non-MACE (n = 193) groups based on the occurrence of MACE during follow-up, and further allocated into a training cohort (n = 167) and a validation cohort (n = 72) at a 7:3 ratio. Automatic segmentation of PCAT surrounding the proximal segments of the left anterior descending artery (LAD), left circumflex coronary artery (LCX), and right coronary artery (RCA) was performed for all patients. Radiomic features of the coronary arteries were extracted, screened, and integrated while quantifying the fat attenuation index (FAI) for the three vessels. Univariate and multivariate logistic regression analyses were utilized to select clinical predictors of adverse cardiovascular events. Subsequently, machine learning techniques were employed to construct models based on FAI, clinical features, and radiomic characteristics. The predictive performance of each model was assessed and compared using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis for clinical utility. The radiomics model demonstrated superior performance in predicting MACE in patients with angina pectoris within both the training and validation cohorts, yielding areas under the curve (AUC) of 0.83 and 0.71, respectively, which significantly outperformed the FAI model (AUC = 0.71, 0.54) and the clinical model (AUC = 0.81, 0.67), with statistically significant differences in AUC (p < 0.05). Calibration curves for all three predictive models exhibited good fit (all p > 0.05). Decision curve analysis indicated that the radiomics model provided higher clinical benefit than the traditional clinical and FAI models. The CCTA-based PCAT radiomics model is an effective tool for predicting MACE in patients with angina pectoris, assisting clinicians in optimizing risk stratification for individual patients. The CCTA-based radiomics model significantly surpasses traditional FAI and clinical models in predicting major adverse cardiovascular events in patients with angina pectoris.

Predicting Outcomes after Hospitalisation for COPD Exacerbation Using Machine Learning

BackgroundEarly readmission and death are critical adverse outcomes following hospitalisation due to exacerbation of chronic obstructive pulmonary disease (ECOPD). This study aimed to develop and validate machine learning models to enhance the prediction of these outcomes after ECOPD hospitalisation.MethodsUtilising a nationwide database, data from the index ECOPD hospitalisation and the preceding year were collected. Prediction models for 30-day readmission and death were developed using logistic lasso regression, random forest, extreme gradient boosting (XGBoost), and neural network, with the LACE index serving as a reference. Model performance was assessed with receiver operating characteristic (ROC) curves and calibration plots from the validation dataset. Key predictors were identified using SHapley Additive exPlanations.ResultsThe study included 101 011 hospitalisations in the development dataset and 17 565 in the validation dataset. The rates of 30-day readmission and death were 29.1% and 4.3%, respectively. AmoXGBoost outperformed other models, achieving an area under the ROC curve of 0.721 (95% CI, 0.713–0.729) for readmission and 0.809 (95% CI, 0.794–0.824) for death, both exceeding the corresponding values for the LACE index (0.651 and 0.641). All machine learning models demonstrated good calibration. The number of hospitalisations in the previous year and the lowest haemoglobin level during the index hospitalisation were the top predictors of readmission and death, respectively.ConclusionsApplying machine learning techniques to large-scale data effectively improves the prediction of early readmission and death following ECOPD hospitalisation. Identifying critical prognostic factors could enhance targeted post-discharge care for this high-risk patient group.

Metabolic impact of low dose IL-2 therapy for primary Sjögren's Syndrome in a double-blind, randomized clinical trial.

Low-dose interleukin 2 (Ld-IL2) is increasingly being explored as an immune-modulating treatment for autoimmune diseases which mainly affect T cell subsets. This study investigates the metabolic effects of Ld-IL2 therapy in patients with primary Sjögren's syndrome (pSS). A total of 60 patients were recruited to conduct a double-blind, randomized clinical trial. Of these patients, 50% (30/60) received Ld-IL2 therapy along with standard treatment for 12 weeks, followed by 12 weeks of follow-up. The effectiveness was evaluated by Sjögren's Tool for Assessing Response (STAR). An untargeted analysis was performed to profile hydrophilic metabolites. Metabolic profiling revealed significant alterations post-treatment, notably in metabolites like acetyl-CoA, ascorbic acid, and glutathione, which are beneficial in managing autoimmune diseases. In addition, the levels of metabolite accumulation were correlated with variations in immune cell subsets (p< 0.05), particularly Tregs. Moreover, patients exhibiting a specific metabolic profile, including lower serum levels of isoleucine, ADP, Thymidine 5'-triphosphate, and other metabolites, had a high response rate (91.7%-98.6%), as indicated by the receiver operating characteristic (ROC) curve. These findings suggest that Ld-IL2 therapy influences metabolic pathways in pSS, offering insights into the systemic effects of Ld-IL2 therapy beyond immune modulation. ClinicalTrials.gov number, NCT02464319. Key Points •Metabolic alteration in pSS is significantly associated with Ld-IL2 therapy. •Metabolic changes correlate with variations in immune cell subsets, particularly Tregs. •Metabolic profiling could be a valuable tool in guiding Ld-IL2 therapy choices for pSS patients.

Predictive Factors for Diabetes Mellitus: Insights from Complete Blood Count Analysis

The 10th edition of the International Diabetes Federation reports that 537 million people worldwide had diabetes in 2021. In Southeast Asia, countries like Malaysia are facing a growing burden of diabetes. This highlights the urgent need for innovative and resourceful approaches to diabetes management. As the prevalence of diabetes continues to rise in these countries, tailored strategies are necessary. To identify and evaluate the potential prognostic indicators for diabetes mellitus, this study involved a dataset consisting of 500 entries, comprising demographic information and selected blood cells from the Complete Blood Count (CBC) test results obtained from the Clinical Laboratory Faculty of Medicine &amp; Health Sciences, Universiti Malaysia Sabah. Using univariate and multivariate logistic regression analysis, the prognostic predictors for diabetes mellitus were identified. In the univariate analysis, all variables are statistically significance at 5% level of significance. However, at multivariate analysis, only age, mean corpuscular hemoglobin concentration (MCHC), white blood cells (WBC) and hematocrit (HCT) emerged as significant predictors of diabetes mellitus. Notably, the abnormal level in WBC exhibited the greatest association with diabetes mellitus, reflecting a 114.7% increased risk compared to a normal WBC level. The statistic value obtained from Hosmer-Lemeshow was 0.944 indicating a well-fitting model. Additionally, the receiver operator characteristic (ROC) curve has a value of 0.7, indicating a strong performance of the model. In conclusion, CBC parameters can be accurate markers and useful in assisting clinical decision-making when properly applied and interpreted.

NEWS-2 Accuracy in Predicting Mortality and Severe Morbidity Among Hospitalized COVID-19 Patients: A Prospective Cohort Study.

Background: Early risk stratification tools for COVID-19 patients have been indicated yet there are few data about their ability to effectively detect clinical deterioration among COVID-19 patients. Objectives: To evaluate the NEWS-2 to predict severe morbidity and mortality for COVID-19 patients admitted to hospitals. Methods: We conducted a prospective cohort study among adult COVID-19 patients with a confirmed diagnosis who were admitted to the inpatient units at COVID-19 Martyrs Medical Military Complex Hospital, from 1 March 2022, until 29 February 2023. NEWS-2 scores were measured at admission and 6, 12, 24, and 48 h after their admission to the hospital using receiver operating characteristic (ROC) curves. Results: Overall, 192 adult COVID-19 patients aged 25-94 years (mean = 62.1, SD = 13.9) were enrolled. Of those, 49.0% were males, 47.4% were vaccinated, and 53.6% had diabetes. The 192 enrolled patients were classified into NEWS-2 score categories, with almost 13% (12.5%) falling into the high-risk category already upon admission. The mean NEWS-2 scores were excellent predictors of mechanical ventilation, admission to the ICU, and mortality, as indicated by an AUROC of 0.94 (95% CI: 0.88-1.00, p < 0.001), 0.91 (95% CI: 0.87-0.96, p < 0.001), and 0.96 (95% CI: 0.92-1.00, p < 0.001), respectively. Significant differences in mean NEWS-2 scores were found between the participating patients, both with and without comorbidity in the course of the patient's stay in the ICU, and mortality (p = 0.004, p = 0.043, respectively). Positive correlations of the high NEWS-2 scores were revealed using a multiple linear regression model, indicating the necessity of administering non-invasive ventilatory assistance (p = 0.013), hospitalization for a minimum of six days (p = 0.013), and admission to the ICU (p = 0.006). Nonetheless, there was a negative association between mortality and the NEWS-2 score (p < 0.001). Conclusions: The NEWS-2 had moderate sensitivity and specificity in predicting the deterioration of patients with COVID-19 whereas there was high sensitivity and specificity in predicting the mortality for patients with COVID-19, both with and without comorbidity. Our findings support the utility of NEWS-2 monitoring as a sensitive approach for initially assessing COVID-19 patients. It could be helpful to enhance the accuracy of predictive performance by supplementing the score parameters by adding biological parameters in addition to clinical judgment.

To Determine the Risk-Based Screening Interval for Diabetic Retinopathy: Development and Validation of Risk Algorithm from a Retrospective Cohort Study.

The optimal screening interval for diabetic retinopathy (DR) remains controversial. This study aimed to develop a risk algorithm to predict the individual risk of referable sight-threatening diabetic retinopathy (STDR) in a mainly Chinese population and to provide evidence for risk-based screening intervals. The retrospective cohort data from 117,418 subjects who received systematic DR screening in Hong Kong between 2010 and 2016 were included to develop and validate the risk algorithm using a parametric survival model. The risk algorithm can be used to predict the individual risk of STDR within a specific time interval, or the time to reach a specific risk margin and thus to allocate a screening interval. The calibration performance was assessed by comparing the cumulative STDR events versus predicted risk over 2 years, and discrimination by using receiver operative characteristics (ROC) curve. Duration of diabetes, glycosylated hemoglobin, systolic blood pressure, presence of chronic kidney disease, diabetes medication, and age were included in the risk algorithm. The validation of prediction performance showed that there was no significant difference between predicted and observed STDR risks in males (5.6% vs. 5.1%, P=0.724) or females (4.8% vs. 4.6%, P=0.099). The area under the receiver operating characteristic curve was 0.80 (95% confidence interval [CI], 0.78 to 0.81) for males and 0.81 (95% CI, 0.79 to 0.83) for females. The risk algorithm has good prediction performance for referable STDR. Using a risk-based screening interval allows us to allocate screening visits disproportionally more to those at higher risk, while reducing the frequency of screening of lower risk people.

TM9SF1 expression correlates with autoimmune disease activity and regulates antibody production through mTOR-dependent autophagy

BackgroundTransmembrane 9 superfamily member 1 (TM9SF1) is involved in inflammation. Since both inflammatory and autoimmune diseases are linked to immune cells regulation, this study investigated the association between TM9SF1 expression and autoimmune disease activity. As B cell differentiation and autoantibody production exacerbate autoimmune disease, the signaling pathways involved in these processes were explored.MethodsTm9sf1−/− mouse rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) models were used to verify the relationship between gene expression and disease severity. Peripheral blood mononuclear cells (PBMCs) from 156 RA and 145 SLE patients were used to explore the relationship between TM9SF1 expression and disease activity. The effectiveness of TM9SF1 as a predictor of disease activity was assessed using multiple logistic regression and receiver operating characteristic (ROC) curves. The signaling pathways regulated by TM9SF1 in B cell maturation and antibody production were conducted by plasma cell induction experiment in vitro.ResultsThe Tm9sf1−/− RA and SLE model mice produced fewer autoantibodies and showed reduced disease severity relative to wild-type (WT) mice. TM9SF1 levels in PBMCs of patients were higher than those in healthy controls, and were reduced in patients with low disease activity relative to those with active RA and SLE. Furthermore, TM9SF1 levels were positively linked with autoantibody titers and pro-inflammatory cytokine levels in both diseases. ROC analyses indicated TM9SF1 outperformed several important clinical indicators in predicting disease activity (area under the curve (AUC) were 0.858 and 0.876 for RA and SLE, respectively). In vitro experiments demonstrated that Tm9sf1 knockout blocked differentiation of B cells into antibody-producing plasma cells by activating mTOR and inhibiting autophagy, and mTOR inhibitors such as rapamycin could reverse this effect.ConclusionsThe primary finding was the identification of the molecular mechanism underlying autophagy regulation in B cells, in which Tm9sf1 knockout was found to modulate mTOR-dependent autophagy to block B cell differentiation into antibody-secreting plasma cells. It was also found that TM9SF1 expression level in PBMCs was an accurate indicator of disease activity in patients with RA and SLE, suggesting its clinical potential for monitoring disease activity in these patients.

Correlation between circulating dephosphorylated uncarboxylated matrix Gla protein and vascular calcification in peritoneal dialysis patients.

To explore the association between serum Dephosphorylated uncarboxylated matrix Gla protein (dp-ucMGP) and abdominal aortic calcification (AAC) in peritoneal dialysis (PD) patients. A total 128 PD patients and 120 healthy controls were enrolled into the study. Serum dp-ucMGP was measured by enzyme-linked immunosorbent assay. Abdominal lateral plain radiography was used to evaluate the abdominal aortic calcification score (AACS). PD patients were divided into two groups according to the presence or absence of AAC. The relationships between dp-ucMGP levels and AACS were assessed by Spearman analysis and the value of dp-ucMGP in predicting AAC was evaluated by receiver operating characteristic (ROC). Serum dp-ucMGP in PD patients were significantly higher than controls (p < 0.05). And PD patients with AAC had higher serum dp-ucMGP than that of PD patients without AAC (p < 0.05). The serum dp-ucMGP levels was positively associated with AACS (r = 0.794, p < 0.0001) in PD patients. The multivariate logistic regression analyses showed that serum dp-ucMGP was independent factors of AAC in PD patients (OR = 2.555, 95% CI = 1.415-4.609). The area under ROC curve of dp-ucMGP was 0.9227, the corresponding sensitivity was 0.86, and the specificity was 0.92. Serum dp-ucMGP levels were positively associated with the AACS in PD patients. Higher serum dp-ucMGP level is independently associated with AAC in PD patients.

Automatic detection of temporomandibular joint osteoarthritis radiographic features using deep learning artificial intelligence. A Diagnostic accuracy study

ObjectiveThe purpose of this study was to investigate the diagnostic performance of a neural network Artificial Intelligence model for the radiographic confirmation of Temporomandibular Joint Osteoarthritis in reference to an experienced radiologist. Materials and MethodsThe diagnostic performance of an AI model in identifying radiographic features in patients with TMJ-OA was evaluated in a diagnostic accuracy cohort study. Adult patients elected for radiographic examination by the Diagnostic Criteria for Temporomandibular Disorders decision tree were included. Cone-beam computed Tomography images were evaluated by object detection YOLO deep learning model. The diagnostic performance was verified against examiner radiographic evaluation. ResultsThe differences between the AI model and examiner were non-significant statistically, except in the subcortical cyst (P=0.049*). AI model showed substantial to near-perfect levels of agreement when compared to those of the examiner data. Regarding each radiographic phenotype, the AI model reported favorable sensitivity, specificity, accuracy, and highly statistically significant Receiver Operating Characteristic (ROC) analysis (p < 0.001). Area Under Curve ranged from 0.872, for surface erosion, to 0.911 for subcortical cyst. ConclusionAI object detection model could open the horizon for a valid, automated, and convenient modality for TMJ-OA radiographic confirmation and radiomic features identification with a significant diagnostic power.

Identification of the m6A/m5C/m1A methylation modification genes in Alzheimer's disease based on bioinformatic analysis.

As a progressive neurodegenerative disease, the comprehensive understanding of the pathogenesis of Alzheimer's disease (AD) is yet to be clarified. Modifications in RNA, including m6A/m5C/m1A, affect the onset and progression of many diseases. Consequently, this study focuses on the role of methylation modification in the pathogenesis of AD. Three AD-related datasets, namely GSE33000, GSE122063, and GSE44770, were acquired from GEO. Differential analysis of m6A/m5C/m1A regulator genes was conducted. Applying a consensus clustering approach, distinct subtypes within AD were identified as per the expression patterns of relevant differentially expressed genes. Machine learning models were constructed to identify five significant genes from the best model. The analysis of hub gene-based drug regulatory networks and ceRNA regulatory networks was conducted by Cytoscape. In comparison to non-AD patients, 24 genes were identified as dysregulated in AD patients, and these genes were associated with various immunological characteristics. Two distinct clusters were successfully identified through consensus clustering, with cluster 2 demonstrating higher immune characteristics compared to cluster 1. The performance of four machine learning models was determined by conducting a receiver operating characteristic (ROC) analysis. The analysis revealed that the SVM model achieved the highest AUC value of 0.947. Five genes (YTHDF1, METTL3, DNMT1, DNMT3A, ALKBH1) were selected as the predicted genes. Finally, a hub gene-based Gene-Drug regulatory network and a ceRNA regulatory network were successfully developed. The findings offered fresh perspectives on the molecular patterns and immune mechanisms underlying AD, contributing valuable insights into our understanding of this complex neurodegenerative disorder.

Coronary computed tomographic angiography predicts crossing through chronic total occlusion within three guidewires

BackgroundCoronary computed tomography angiography (CCTA) is a valuable non-invasive method for the diagnosis and treatment of coronary artery disease, especially providing important clinical decision-making basis for percutaneous coronary intervention (PCI) of chronic total occlusion (CTO). However, there is no data to discuss whether CCTA can predict the success rate of crossing through occlusive lesions within 3 guidewires. ObjectiveThe aim of this study was to establish a scoring model based on CCTA images to predict the success rate of CTO-PCI within 3 guidewires for grading treatment difficulty. MethodsData including 267 CTO lesions in 252 patients with pre-procedural CCTA were analyzed. Successfully crossing through the lesion within 3 guidewires was set as a new endpoint to eliminate operator bias and estimate surgical costs. The Guidewires of CTO (GW-CTO) score was determined by assigning an appropriate integer point for each independent predictor of this endpoint and summing all points accrued. Results72.3% reached the new endpoint. We identified 5 predictors. Including blunt proximal stump, both proximal and distal branches, occluded segmental curvature>45°, length of occlusion≥20 mm, and calcified plaque burden≥0.125. With the GW-CTO score rising from 0 to 5, the success rate dropped from 98.83% to 12.50%. The CTO-PCI difficulty was stratified into three grades: easy (0–1), intermediate (2–3), and difficult (4–5). The GW-CTO score yielded good predictive results, with an area under receiver-operator characteristic (ROC) curve of 0.87. ConclusionThis model is very novel and presents a new perspective to predict the probability of successfully passing through CTO lesions within 3 guidewires. This may help grade the difficulty of CTO-PCI and determining the cost efficiency.

Remote patient-reported outcome measure triage score for monitoring disease activity and allocation of consultations in rheumatoid arthritis patients

Objective Currently, expedited by the coronavirus disease 2019 pandemic, there is high demand for allocating patients in a state of low disease activity to telehealth, ideally based on remote measurements. This cross-sectional study assesses the discriminative accuracy of the Rheumatoid Arthritis Impact of Disease (RAID) questionnaire regarding high and low disease activity. Furthermore, we aimed to optimize this classification, developing a remote triage score based on RAID and other patient-reported outcome measures (PROMs). Method Data were acquired from an outpatient clinic cohort of chronic rheumatoid arthritis patients at a large trainee hospital in the Netherlands. Patients were divided into high and low disease categories, based on 28-joint Disease Activity Score–C-reactive protein. Least absolute shrinkage and selection operator logistic regression were performed, including RAID item scores and other PROMs. Receiver operating characteristics curves and areas under the curve (AUCs) were obtained, and cut-off scores were based on predefined criteria of 90% and 95% sensitivity. Results In total, 278 patients were analysed, of whom 77.2% were identified as having low disease activity. RAID results correlated with DAS28-CRP, showing good performance. The regression model included the RAID items pain and functional disability assessment, and the self-reported swollen joint count (SR-SJC). With an AUC of 0.88 (95% confidence interval 0.84–0.92), this model performed better than the RAID total score. Conclusion A remote triage score based on a composite score of pain, functional disability assessment, and SR-SJC can detect a sufficient proportion of patients with low disease activity who can be allocated to remote consultations.

A modified analysis protocol for the PhNR test.

Several studies have reported that glaucoma patients have abnormal photopic negative response (PhNR) results compared to reference control subjects. The International Society for Clinical Electrophysiology of Vision (ISCEV) released an extended protocol for PhNR (I-PhNR) in 2018. The purpose of this study was to compare the I-PhNR protocol to a similar protocol modified (M-PhNR) to enhance the performance of the method in detecting glaucomatous damage. Thirty subjects were enrolled in this study (12 glaucoma patients, 10 glaucoma suspects, 8 normal controls). PhNR tests were conducted with a Diagnosys E3 mobile system (Diagnosys LLC, Lowell, MA). I-PhNR tests utilized all parameters specified by the ISCEV requirement. M-PhNR tests used the same parameters as the ISCEV tests with the exceptions of a 5-45Hz bandpass filter and a novel, objective sweep-selection parameter. According to the ISCEV protocol, the PhNR relative to baseline (i.e., BT), a-wave and b-wave response amplitudes and BT/b-wave amplitude ratios were measured. Coefficients of variation, receiver operating characteristic (ROC) curves, and t-tests were used to assess the data from one randomly chosen eye per subject. The M-PhNR protocol resulted in a decrease in the intra-subject repeat test coefficient of variation and a decrease in the average inter-subject coefficient of variation for the glaucoma subjects. The ROC curves demonstrated an increase in the area under the curve (AUC) for the M-PhNR compared to the I-PhNR protocol. The sensitivity and specificity were also greater for the M-PhNR protocol. The M-PhNR protocol resulted in a decrease in intra-subject and inter-subject data variability which resulted in a significant increase in the ROC AUC, sensitivity, and specificity for glaucoma. Thus, the M-PhNR protocol shows promise as a better diagnostic tool than the I-PhNR protocol for detecting glaucoma.

Exploring common biomarkers of ischemic stroke and obstructive sleep apnea through bioinformatics analysis.

Clinical observations have shown that many patients with ischemic stroke (IS) have a history of obstructive sleep apnea (OSA) both before and after the stroke's onset, suggesting potential underlying connections and shared comorbid mechanisms between the two conditions. The aim of this study is to identify the genetic characteristics of OSA patients who develop IS and to establish a reliable disease diagnostic model to assess the risk of IS in OSA patients. We selected IS and OSA datasets from the Gene Expression Omnibus (GEO) database as training sets. Core genes were identified using the Limma package, Weighted Gene Co-expression Network Analysis (WGCNA), and machine learning algorithms. Gene Set Variation Analysis (GSVA) was conducted for pathway enrichment analysis, while single-sample gene set enrichment analysis (ssGSEA) was employed for immune infiltration analysis. Finally, a diagnostic model was developed using Least Absolute Shrinkage and Selection Operator (LASSO) regression, with its diagnostic efficacy validated using receiver operating characteristic (ROC) curves across two independent validation sets. The results revealed that differential analysis and machine learning identified two common genes, TM9SF2 and CCL8, shared between IS and OSA. Additionally, seven signaling pathways were found to be commonly upregulated in both conditions. Immune infiltration analysis demonstrated a significant decrease in monocyte levels, with TM9SF2 showing a negative correlation and CCL8 showing a positive correlation with monocytes. The diagnostic model we developed exhibited excellent predictive value in the validation set. In summary, two immune-related core genes, TM9SF2 and CCL8, were identified as common to both IS and OSA. The diagnostic model developed based on these genes may be used to predict the risk of IS in OSA patients.

HOLISMOKES. XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys

While supervised neural networks have become state of the art for identifying the rare strong gravitational lenses from large imaging data sets, their selection remains significantly affected by the large number and diversity of non-lens contaminants. This work evaluates and compares systematically the performance of neural networks in order to move towards a rapid selection of galaxy-scale strong lenses with minimal human input in the era of deep, wide-scale surveys. We used multiband images from PDR2 of the Hyper-Suprime Cam (HSC) Wide survey to build test sets mimicking an actual classification experiment, with 189 securely-identified strong lenses from the literature over the HSC footprint and 70\,910 non-lens galaxies in COSMOS covering representative lens-like morphologies. Multiple networks were trained on different sets of realistic strong-lens simulations and non-lens galaxies, with various architectures and data preprocessing, mainly using the deepest $gri$-bands. Most networks reached excellent area under the Receiver Operating Characteristic (ROC) curves on the test set of 71,099 objects, and we determined the ingredients to optimize the true positive rate for a total number of false positives equal to zero or 10 $ and TPR$_ $). The overall performances strongly depend on the construction of the ground-truth training data and they typically, but not systematically, improve using our baseline residual network architecture presented in HOLISMOKES VI. TPR$_ $ tends to be higher for ResNets (simeq 10--40&lt;!PCT!&gt;) compared to AlexNet-like networks or G-CNNs. Improvements are found when (1) applying random shifts to the image centroids, (2) using square-root scaled images to enhance faint arcs, (3) adding $z$-band to the otherwise used $gri$-bands, or (4) using random viewpoints of the original images. In contrast, we find no improvement when adding $g - i$ difference images (where alpha is a tuned constant) to subtract emission from the central galaxy. The most significant gain is obtained with committees of networks trained on different data sets, with a moderate overlap between populations of false positives. Nearly-perfect invariance to image quality can be achieved by using realistic PSF models in our lens simulation pipeline, and by training networks either with large number of bands, or jointly with the PSF and science frames. Overall, we show the possibility to reach a $ as high as 60&lt;!PCT!&gt; for the test sets under consideration, which opens promising perspectives for pure selection of strong lenses without human input using the Rubin Observatory and other forthcoming ground-based surveys.

Identification of m5C-Related gene diagnostic biomarkers for sepsis: a machine learning study.

Sepsis is a serious condition that occurs when the body's response to infection becomes uncontrolled, resulting in a high risk of death. Despite improvements in healthcare, identifying sepsis early is difficult because of its diverse nature and the absence of distinct biomarkers. Recent studies suggest that 5-methylcytosine (m5C)-related genes play a significant role in immune responses, yet their diagnostic potential in sepsis remains unexplored. This research combined and examined four sepsis-related datasets (GSE95233, GSE57065, GSE100159, and GSE65682) sourced from the Gene Expression Omnibus (GEO)database to discover m5C-related genes with differential expression. Various machine learning methods, such as decision tree, random forest, and XGBoost, were utilized in identifying crucial hub genes. Receiver Operating Characteristic (ROC) curve analysis was used to assess the diagnostic accuracy of these genetic markers. Additionally, single-gene enrichment and immune infiltration analyses were conducted to investigate the underlying mechanisms involving these hub genes in sepsis. Three hub genes, DNA Methyltransferase 1 (DNMT1), tumor protein P53 (TP53), and toll-like receptor 8 (TLR8), were identified and validated for their diagnostic efficacy, showing area under the curve (AUC) values above 0.7 in both test and validation sets. Enrichment analyses revealed that these genes are involved in key pathways such as p53 signaling and Toll-like receptor signaling. Immune infiltration analysis indicated significant correlations between hub genes and various immune cell types, suggesting their roles in modulating immune responses during sepsis. The study highlights the diagnostic potential of m5C-related genes in sepsis and their involvement in immune regulation. These findings offer new insights into sepsis pathogenesis and suggest that DNMT1, TP53, and TLR8 could serve as valuable biomarkers for early diagnosis. Further studies should prioritize validating these biomarkers in clinical settings and investigating their potential for therapy.

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.