Predictive Power Research Articles

Volatile Organic Compounds in Exhaled Breath: A Promising Approach for Accurate Differentiation of Lung Adenocarcinoma and Squamous Cell Carcinoma.

Lung cancer subtyping, particularly differentiating adenocarcinoma (ADC) from squamous cell carcinoma (SCC), is paramount for clinicians to develop effective treatment strategies. In this study, we aimed: (i) to discover VOC biomarkers for precise diagnosis of ADC and SCC, (ii) to investigated the impact of risk factors on ADC and SCC prediction, and (iii) to explore the metabolic pathways of VOC biomarkers. Exhaled breath samples from patients with ADC (n=149) and SCC (n=94) were analyzed by gas chromatography-mass spectrometry (GC-MS). Both multivariate and univariate statistical analysis method were employed to identify VOC biomarkers. Support vector machine (SVM) prediction models were developed and validated based on these VOC biomarkers. The impact of risk factors on ADC and SCC prediction was investigated. A panel of 13 VOCs was found to differ significantly between ADC and SCC. Utilizing the SVM algorithm, the VOC biomarkers achieved a specificity of 90.48%, a sensitivity of 83.50%, and an AUC value of 0.958 on the training set. On the validation set, these VOC biomarkers attained a predictive power of 85.71% for sensitivity and 73.08% for specificity, along with an AUC value of 0.875. Clinical risk factors exhibit certain predictive power on ADC and SCC prediction. Integrating these risk factors into the prediction model based on VOC biomarkers can enhance its predictive accuracy. This work indicates that exhaled breath holds the potential to precisely detect ADCs and SCCs. Considering clinical risk factors is essential when differentiating between these two subtypes.

The U-shape relationship between insulin resistance-related indexes and chronic kidney disease: a retrospective cohort study from National Health and Nutrition Examination Survey 2007–2016

BackgroundThere is ongoing debate on the correlation between chronic kidney disease (CKD) and insulin resistance (IR)-related indices. Our objective was to explore the prognostic ability of IR-related indexes for the prevalence of CKD, as well as the mortality from all causes and cardiovascular disease (CVD) in CKD patients.MethodsThe data used in this study came from the National Health and Nutrition Examination Survey (NHANES). Binary logistic regression analysis, Cox proportional hazards model, and restricted cubic spline (RCS) were used to analyze the relationship between IR-related indexes, including metabolic score of IR (METS-IR), homeostatic model assessment for IR (HOMA-IR), triglyceride glucose index (TyG), triglyceride glucose-waist-to-height ratio (TyG-WHtR), triglyceride glucose-body mass index (TyG-BMI), with CKD and its all-cause mortality and CVD mortality. Subgroup analysis was performed to test the stability of the results. Finally, the predictive power of IR-related indexes for CKD was tested by the receiver operating characteristic (ROC) curve.ResultsAmong the recruited 10,660 participants, 15.42% were CKD patients. All IR-related indexes were found to be nonlinearly correlated to the prevalence of CKD in the study. When the TyG index was higher than 9.05, it was positively associated with CKD (OR: 1.77, 95% CI 1.44–2.18). Moreover, increased TyG-WHtR level was correlated with a greater prevalence of CKD when it was higher than 4.3 (OR: 1.31, 95% CI 1.19–1.45). Other IR-related indexes (METS-IR, HOMA-IR, and TyG-BMI) showed fewer notable correlations with CKD. The association of IR-related indexes and the prevalence of CKD remained consistent in most subgroups (P for interactions > 0.05). TyG-WHtR was also the predictor of all-cause mortality in CKD patients (HR: 1.34, 95% CI 1.14–1.58), while other IR-related indexes were not correlated with the all-cause mortality or CVD mortality in CKD patients (P > 0.05). Otherwise, ROC curves showed that TyG-WHtR had more robust diagnostic efficacy than other IR-related indexes (METS-IR, HOMA-IR, TyG, and TyG-BMI) in predicting CKD (area under the curve: 0.630, 95% CI 0.615–0.644).ConclusionsIR-related biomarkers (METS-IR, HOMA-IR, TyG, and TyG-BMI) were positively correlated with the prevalence of CKD. Moreover, TyG-WHtR enhanced CKD and its all-cause mortality prediction. In patients with elevated levels of IR-related indexes, the early detection and intervention of IR may reduce the occurrence of CKD and the prognosis of CKD patients.

Revolution in malaria detection: unveiling current breakthroughs and tomorrow’s possibilities in biomarker innovation

The ongoing battle against malaria has seen significant advancements in diagnostic methodologies, particularly through the discovery and application of novel biomarkers. Traditional diagnostic techniques, such as microscopy and rapid diagnostic tests (RDTs), have their limitations in terms of sensitivity, specificity, and the ability to detect low-level infections. Recent breakthroughs in biomarker research promise to overcome these challenges, providing more accurate, rapid, and non-invasive detection methods. These advancements are critical in enhancing early detection, guiding effective treatment, and ultimately reducing the global malaria burden. Innovative approaches in biomarker detection are leveraging cutting-edge technologies like next-generation sequencing (NGS), proteomics, and metabolomics. These techniques have led to the identification of new biomarkers that can be detected in blood, saliva, or urine, offering less invasive and more scalable options for widespread screening. For instance, the discovery of specific volatile organic compounds (VOCs) in the breath of infected individuals presents a revolutionary non-invasive diagnostic tool. Additionally, the integration of machine learning algorithms with biomarker data is enhancing the precision and predictive power of malaria diagnostics, making it possible to distinguish between different stages of infection and identify drug-resistant strains. Looking ahead, the future of malaria detection lies in the continued exploration of multi-biomarker panels and the development of portable, point-of-care diagnostic devices. The incorporation of smartphone-based technologies and wearable biosensors promises to bring real-time monitoring and remote diagnostics to even the most resource-limited settings.

AURKB promotes immunogenicity and immune infiltration in clear cell renal cell carcinoma

BackgroundChromatin regulators (CRs) are capable of causing epigenetic alterations, which are significant features of cancer. However, the function of CRs in controlling Clear Cell Renal Cell Carcinoma (ccRCC) is not well understood. This research aims to discover a CRs prognostic signature in ccRCC and to elucidate the roles of CRs-related genes in tumor microenvironment (TME).MethodsExpression profiles and relevant clinical annotations were retrieved from the Cancer Genome Atlas (TCGA) and UCSC Xena platform for progression-free survival (PFS) data. The R package "limma" was used to identify differentially expressed CRs. A predictive model based on five CRs was developed using LASSO-Cox analysis. The model's predictive power and applicability were validated using K-M curves, ROC curves, nomograms, comparisons with other models, stratified survival analyses, and validation with the ICGC cohort. GO and GSEA analyses were performed to investigate mechanisms differentiating low and high riskScore groups. Immunogenicity was assessed using Tumor Mutational Burden (TMB), immune cell infiltrations were inferred, and immunotherapy was evaluated using immunophenogram analysis and the expression patterns of human leukocyte antigen (HLA) and checkpoint genes. Differentially expressed CRs (DECRs) between low and high riskScore groups were identified using log2|FC|> 1 and FDR < 0.05. AURKB, one of the high-risk DECRs and a component of our prognostic model, was selected for further analysis.ResultsWe constructed a 5 CRs signature, which demonstrated a strong capacity to predict survival and greater applicability in ccRCC. Elevated immunogenicity and immune infiltration in the high riskScore group were associated with poor prognosis. Immunotherapy was more effective in the high riskScore group, and certain chemotherapy medications, including cisplatin, docetaxel, bleomycin, and axitinib, had lower IC50 values. Our research shows that AURKB is critical for the immunogenicity and immune infiltration of the high riskScore group.ConclusionOur study produced a reliable prognostic prediction model using only 5 CRs. We found that AURKB promotes immunogenicity and immune infiltration. This research provides crucial support for the development of prognostic biomarkers and treatment strategies for ccRCC.

Weight-adjusted-waist index is positively associated with urinary incontinence: results from the National Health and Nutrition Examination Survey (NHANES) 2001–2018

BackgroundUrinary incontinence (UI) is closely related to obesity. The aim of this study is to evaluate the association of a novel anthropometric indicator weight-adjusted-waist index (WWI) with UI.MethodsThis cross-sectional study used the data from National Health and Nutrition Examination Survey (NHANES) 2001–2018. Weighted multivariable logistic regression was used to evaluate the relationship between WWI and three types of UI [stress UI (SUI), urgency UI (UUI), and mixed UI (MUI)]. The receiver operating characteristic (ROC) curve and Delong et al.’s test were utilized for comparison of the predictive capability for UI between WWI and body mass index (BMI), waist circumference (WC).ResultsA total of 41,614 participants were included in this study, of whom 23.57% had SUI, 19.24% had UUI, and 9.43% had MUI. In the fully adjusted model, WWI was positively associated with three types of UI [SUI: odds ratio (OR) = 1.19, 95%Confidence interval (CI) 1.13–1.25; UUI: OR = 1.18, 95%CI 1.13–1.24; MUI: OR = 1.19, 95%CI 1.11–1.27, all p < 0.001]. Compared to the lowest WWI interval, the positive correlation between WWI and UI still existed in the highest WWI group after converting WWI to a categorical variable by quartiles (SUI: OR = 1.52, 95%CI 1.35–1.71, p < 0.001; UUI: OR = 1.50, 95%CI 1.33–1.69, p < 0.001; MUI: OR = 1.55, 95%CI 1.32–1.83, p < 0.001). WWI had a stronger prediction for three types of UI than BMI and WC (all p < 0.001).ConclusionA higher WWI was linked with an increased likelihood of three types of UI (SUI, UUI, and MUI) in the United State population. Compared to BMI and WC, WWI had a stronger predictive power for UI. WWI may be a better adiposity parameter for evaluating UI.

The classification algorithms to support the management of the patient with femur fracture

Effectiveness in health care is a specific characteristic of each intervention and outcome evaluated. Especially with regard to surgical interventions, organization, structure and processes play a key role in determining this parameter. In addition, health care services by definition operate in a context of limited resources, so rationalization of service organization becomes the primary goal for health care management. This aspect becomes even more relevant for those surgical services for which there are high volumes. Therefore, in order to support and optimize the management of patients undergoing surgical procedures, the data analysis could play a significant role. To this end, in this study used different classification algorithms for characterizing the process of patients undergoing surgery for a femoral neck fracture. The models showed significant accuracy with values of 81%, and parameters such as Anaemia and Gender proved to be determined risk factors for the patient’s length of stay. The predictive power of the implemented model is assessed and discussed in view of its capability to support the management and optimisation of the hospitalisation process for femoral neck fracture, and is compared with different model in order to identify the most promising algorithms. In the end, the support of artificial intelligence algorithms laying the basis for building more accurate decision-support tools for healthcare practitioners.

Scientific Inference with Interpretable Machine Learning: Analyzing Models to Learn About Real-World Phenomena

To learn about real world phenomena, scientists have traditionally used models with clearly interpretable elements. However, modern machine learning (ML) models, while powerful predictors, lack this direct elementwise interpretability (e.g. neural network weights). Interpretable machine learning (IML) offers a solution by analyzing models holistically to derive interpretations. Yet, current IML research is focused on auditing ML models rather than leveraging them for scientific inference. Our work bridges this gap, presenting a framework for designing IML methods—termed ’property descriptors’—that illuminate not just the model, but also the phenomenon it represents. We demonstrate that property descriptors, grounded in statistical learning theory, can effectively reveal relevant properties of the joint probability distribution of the observational data. We identify existing IML methods suited for scientific inference and provide a guide for developing new descriptors with quantified epistemic uncertainty. Our framework empowers scientists to harness ML models for inference, and provides directions for future IML research to support scientific understanding.

Decoding depression: a comprehensive multi-cohort exploration of blood DNA methylation using machine learning and deep learning approaches

The causes of depression are complex, and the current diagnosis methods rely solely on psychiatric evaluations with no incorporation of laboratory biomarkers in clinical practices. We investigated the stability of blood DNA methylation depression signatures in six different populations using six public and two domestic cohorts (n = 1942) conducting mega-analysis and meta-analysis of the individual studies. We evaluated 12 machine learning and deep learning strategies for depression classification both in cross-validation (CV) and in hold-out tests using merged data from 8 separate batches, constructing models with both biased and unbiased feature selection. We found 1987 CpG sites related to depression in both mega- and meta-analysis at the nominal level, and the associated genes were nominally related to axon guidance and immune pathways based on enrichment analysis and eQTM data. Random forest classifiers achieved the highest performance (AUC 0.73 and 0.76) in CV and hold-out tests respectively on the batch-level processed data. In contrast, the methylation showed low predictive power (all AUCs < 0.57) for all classifiers in CV and no predictive power in hold-out tests when used with harmonized data. All models achieved significantly better performance (>14% gain in AUCs) with pre-selected features (selection bias), with some of the models (joint autoencoder-classifier) reaching AUCs of up to 0.91 in the final testing regardless of data preparation. Different algorithmic feature selection approaches may outperform limma, however, random forest models perform well regardless of the strategy. The results provide an overview over potential future biomarkers for depression and highlight many important methodological aspects for DNA methylation-based depression profiling including the use of machine learning strategies.

Neuron-Specific Enolase as a Predictor of Neurologic Outcomes in Extracorporeal Cardiopulmonary Resuscitation Patients

Background: Neuron-specific enolase (NSE) has traditionally been used as a biomarker to predict neurologic outcomes after cardiac arrest. This study aimed to evaluate the utility of NSE in predicting neurologic outcomes in patients undergoing extracorporeal cardiopulmonary resuscitation (ECPR). Methods: This observational cohort study included 47 consecutive adult ECPR patients (median age, 59.0 years; 74.5% males) treated between January 2018 and December 2021 at a tertiary extracorporeal life support center. The primary outcome was a poor neurologic outcome, defined as a Cerebral Performance Category score of 3–5 at hospital discharge. Results: Twelve (25.5%) patients had abnormal findings on computed tomography of the brain. A poor neurologic outcome was demonstrated in 22 (46.8%) patients. The NSE level at 72 h after ECPR showed the best prediction power for a poor neurologic outcome compared with NSE at 24 and 48 h. A cutoff value exceeding 61.9 μg/L for NSE at 72 h yielded an area under the curve (AUC) of 0.791 for predicting poor neurologic outcomes and exceeding 62.1 μg/L with an AUC of 0.838 for 30-day mortality. Conclusions: NSE levels at 72 h after ECPR appear to be a reliable biomarker for predicting poor neurologic outcomes and 30-day mortality in ECPR patients.

Developing a machine learning-based prognosis and immunotherapeutic response signature in colorectal cancer: insights from ferroptosis, fatty acid dynamics, and the tumor microenvironment

InstructionColorectal cancer (CRC) poses a challenge to public health and is characterized by a high incidence rate. This study explored the relationship between ferroptosis and fatty acid metabolism in the tumor microenvironment (TME) of patients with CRC to identify how these interactions impact the prognosis and effectiveness of immunotherapy, focusing on patient outcomes and the potential for predicting treatment response.MethodsUsing datasets from multiple cohorts, including The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), we conducted an in-depth multi-omics study to uncover the relationship between ferroptosis regulators and fatty acid metabolism in CRC. Through unsupervised clustering, we discovered unique patterns that link ferroptosis and fatty acid metabolism, and further investigated them in the context of immune cell infiltration and pathway analysis. We developed the FeFAMscore, a prognostic model created using a combination of machine learning algorithms, and assessed its predictive power for patient outcomes and responsiveness to treatment. The FeFAMscore signature expression level was confirmed using RT-PCR, and ACAA2 progression in cancer was further verified.ResultsThis study revealed significant correlations between ferroptosis regulators and fatty acid metabolism-related genes with respect to tumor progression. Three distinct patient clusters with varied prognoses and immune cell infiltration were identified. The FeFAMscore demonstrated superior prognostic accuracy over existing models, with a C-index of 0.689 in the training cohort and values ranging from 0.648 to 0.720 in four independent validation cohorts. It also responses to immunotherapy and chemotherapy, indicating a sensitive response of special therapies (e.g., anti-PD-1, anti-CTLA4, osimertinib) in high FeFAMscore patients.ConclusionFerroptosis regulators and fatty acid metabolism-related genes not only enhance immune activation, but also contribute to immune escape. Thus, the FeFAMscore, a novel prognostic tool, is promising for predicting both the prognosis and efficacy of immunotherapeutic strategies in patients with CRC.

Improving the Accuracy of Permeability Data to Gain Predictive Power: Assessing Sources of Variability in Assays Using Cell Monolayers

Abstract: The ability to predict the rate of permeation of new compounds across biological membranes is of high importance for their success as drugs, as it determines their efficacy, pharmacokinetics, and safety profile. In vitro permeability assays using Caco-2 monolayers are commonly employed to assess permeability across the intestinal epithelium, with an extensive number of apparent permeability coefficient (Papp) values available in the literature and a significant fraction collected in databases. The compilation of these Papp values for large datasets allows for the application of artificial intelligence tools for establishing quantitative structure–permeability relationships (QSPRs) to predict the permeability of new compounds from their structural properties. One of the main challenges that hinders the development of accurate predictions is the existence of multiple Papp values for the same compound, mostly caused by differences in the experimental protocols employed. This review addresses the magnitude of the variability within and between laboratories to interpret its impact on QSPR modelling, systematically and quantitatively assessing the most common sources of variability. This review emphasizes the importance of compiling consistent Papp data and suggests strategies that may be used to obtain such data, contributing to the establishment of robust QSPRs with enhanced predictive power.

A thermal fluid dynamic model for the melt region during the laser powder bed fusion of polyamide 11 (PA11)

Studying the physical characteristics of the melt region in polymer-based laser powder bed fusion (PBF-LB/P) offers good comprehension that is essential for quality control, process optimization, understanding of material behavior, and preserving process consistency. Among the few numerical modelling investigations of PBF-LB/P, this study focuses on 3D multiphysics simulation of melt pool morphologies, beginning with powder bed preparation incorporating the actual particle size distribution using the discrete element method (DEM). A thermal fluid dynamic numerical model is then developed based on the finite volume method (FVM) using Flow-3D Weld to simulate PBF-LB while using polyamide 11 (PA11) as the primary material. The cooling after sintering plays a crucial role for the formation of the melt region due to the high viscosity of the polymer material effectively resulting in creeping flow. Here, we propose a model for both sintering as well as the subsequent cooling step, a two-step model capable of simulating the finished structure accurately with relatively low computational resources. The model's accuracy is verified through a mesh dependency analysis and its predictive power via validation against internal single-track experiments providing dimension data on the melt region width and depth. The simulation results allow for a detailed exploration of the effects of laser settings on melt region morphologies and the occurrence of defects. The model's scope is further extended to multi-track simulation, revealing the connection between surface roughness and laser settings. In general, this study provides significant contributions to the comprehension of PBF-LB/P methodologies which in turn can be used for further process improvement.

Deep learning model based on endoscopic images predicting treatment response in locally advanced rectal cancer undergo neoadjuvant chemoradiotherapy: a multicenter study

PurposeNeoadjuvant chemoradiotherapy has been the standard practice for patients with locally advanced rectal cancer. However, the treatment response varies greatly among individuals, how to select the optimal candidates for neoadjuvant chemoradiotherapy is crucial. This study aimed to develop an endoscopic image-based deep learning model for predicting the response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer.MethodsIn this multicenter observational study, pre-treatment endoscopic images of patients from two Chinese medical centers were retrospectively obtained and a deep learning-based tumor regression model was constructed. Treatment response was evaluated based on the tumor regression grade and was defined as good response and non-good response. The prediction performance of the deep learning model was evaluated in the internal and external test sets. The main outcome was the accuracy of the treatment prediction model, measured by the AUC and accuracy.ResultsThis deep learning model achieved favorable prediction performance. In the internal test set, the AUC and accuracy were 0.867 (95% CI: 0.847–0.941) and 0.836 (95% CI: 0.818–0.896), respectively. The prediction performance was fully validated in the external test set, and the model had an AUC of 0.758 (95% CI: 0.724–0.834) and an accuracy of 0.807 (95% CI: 0.774–0.843).ConclusionThe deep learning model based on endoscopic images demonstrated exceptional predictive power for neoadjuvant treatment response, highlighting its potential for guiding personalized therapy.

Time Changes of Survival and Cardiovascular Determinants in a Cohort of Middle-Aged Men Followed Up for 61 Years until Extinction

Objective: To study possible determinants of longevity in a cohort of middle-aged men followed for 61 years until extinction using measurements taken at baseline and at years 31 or 61 of follow-up. Material and Methods: In 1960, two rural cohorts including a total of 1712 men aged 40–59 years were enrolled within the Italian section of the Seven Countries Study of Cardiovascular Diseases, and measurements related to mainly cardiovascular risk factors, lifestyle behaviors, and chronic diseases were taken at year 0 and year 31 of follow-up (when only 390 could be examined). Multiple linear regression models were computed to relate personal characteristics with the length of survival in both dead men and survivors. Results: Baseline cardiovascular risk factors, smoking and dietary habits, and chronic diseases (taken at year 0 with men aged 40–59 years) were significant predictors of the length of survival both from year 0 to year 31 and from year 0 to year 61, but only chronic diseases were independent predictors for the period of 31 to 61 years. Significant predictors of survival using measurements taken at year 31 (age range 71 to 90 years) were only smoking and dietary habits and chronic diseases. Conclusions: During a lifetime of follow-up, the personal characteristics with continuous predictive power of survival were only lifestyle behaviors and major chronic diseases.

Development and validation of nomogram for predicting the risk of transferring to the ICU for children with influenza.

Development of a nomogram model for predicting the magnitude of risk of transferring hospitalized children with influenza to the ICU. In a single-center retrospective study, 318 children with influenza who were hospitalized in our hospital from January 2018 to August 2023 were collected as study subjects. Children with influenza were randomly assigned to the training set and validation set in a ratio of 4:1. In the training set, risk factors were identified using univariate and multivariate logistic regression analyses, and a nomogram model was created on this basis. The validation set was used to evaluate the predictive power of the model. Multifactorial logistic regression analysis revealed six independent risk factors for transfer to the ICU in hospitalized children with influenza, including elevated peripheral white blood cell counts, elevated large platelet ratios, reduced mean platelet width, reduced complement C3, elevated serum globulin levels, and reduced total immunoglobulin M levels. Using these six metrics as predictors to construct a nomogram graphical model, the C-index was 0.970 (95% Cl: 0.953-0.988). The areas under the curve for the training and validation sets were 0.966 (95%Cl 0.947-0.985) and 0.919 (95%Cl 0.851-0.986), respectively. A nomogram for predicting the risk of transferring to the ICU for children with influenza was developed and validated, which demonstrates good calibration and clinical benefits.

Non-insulin-based insulin resistance indices predict early neurological deterioration in elderly and middle-aged acute ischemic stroke patients in Northeast China

Insulin resistance (IR) has a strong association with acute ischemic stroke (AIS) occurrence and poor prognosis of afflicted patients. However, the relation between early neurological deterioration (END) risk and IR in elderly and middle-aged patients remains to be thoroughly studied. Here, we investigated the relationship between four indicators of IR and the risk of END in middle-aged patients patients with AIS. The study retrospectively analyzed 1696 elderly and middle-aged patients having AIS between January 2019 and June 2023. Within 7 days of admission, the patients were then stratified relying upon alternations in the National Institutes of Health Stroke Scale. Subsequently, we employed logistic regression analyses for assessing each index correlation with END on the basis of the tertiles of TyG index (TyGI), triglyceride to high-density lipoprotein ratio (TG/HDL), TyG-BMI, alongside IR metabolic score (METS-IR). These four indicators were significantly heightened in the END group (n = 680) in comparison to the non-END group (n = 1016). When grouping using tertiles, the four aforementioned indicators emerged as independent risk factors for END occurrence, whether or not adjusted for confounding factors. The results revealed a progressive elevation in END occurrence risk with the rise in the tertile of each indicator. Finally, we utilized receiver operating characteristic (ROC) curves for assessing the indicators' predictive power. TyG-BMI, TyGI, TG/HDL, and METS-IRs’ area under the curve (AUC) were, respectively, 0.736 (95% CI: 0.712–0.761; P < 0.001), 0. 694 (95% CI: 0.668–0.721; P < 0.001), 0.684 (95% CI: 0.658–0.711; P < 0.001), and 0.722 (95% CI: 0.697–0.747; P < 0.001). IR is associated with END risk in middle-aged AIS patients. TyG-BMI, TyGI, TG/HDL, and METS-IR are independent risk factors of END in elderly and middle-aged AIS patients. Simultaneously, these four IR indicators have significant predictive power for END.

Power of prediction: spatiotemporal Gaussian process modeling for predictive control in slope-based wavefront sensing

Power of prediction: spatiotemporal Gaussian process modeling for predictive control in slope-based wavefront sensing

Predictors of left atrial appendage thrombus in atrial fibrillation patients undergoing cardioversion.

Atrial fibrillation and atrial flutter represent the most prevalent clinically significant cardiac arrhythmias. While the CHA2DS2-VASc score is commonly used to inform anticoagulation therapy decisions for patients with these conditions, its predictive power is limited. Therefore, we sought to improve risk prediction for left atrial appendage thrombus (LAAT), a known risk factor for stroke in these patients. We developed and validated an explainable machine learning model using the eXtreme Gradient Boosting algorithm with 5 × 5 nested cross-validation. The primary outcome was to predict the probability of LAAT in patients with atrial fibrillation and atrial flutter who underwent transesophageal echocardiogram prior to cardioversion. Our algorithm used 37 demographic, comorbid, and transthoracic echocardiographic variables. A total of 795 patients were included in our analysis. LAAT was present in 11.3% of the patients. The average age of patients was 63.3 years and 34.7% were women. Patients with LAAT had significantly lower left ventricular ejection fraction (29.9% vs 43.5%; p < 0.001), lower E' lateral velocity (5.7cm vs. 7.9cm; p < 0.001) and higher E/A ratio (2.6 vs 1.8; p = 0.002). Our machine learning model achieved a high AUC of 0.79, with a high specificity of 0.82, and modest sensitivity of 0.57. Left ventricular ejection fraction was the most important variable in predicting LAAT. Patients were split into 10 buckets based on the percentile of their predicted probability of having thrombus. The lower the percentile (e.g., 10%), the lower the probability of having thrombus. Using a cutoff point of 0.16 which includes 10.0% of the patients, we can rule out thrombus with 100% confidence. Using machine learning, we refined the predictive power of predicting LAAT and explained the model. These results show promise in providing better guidance for anticoagulation therapy and cardioversion in AF and AFL patients.

The associations between the Geriatric Nutritional Risk Index and all-cause, cancer-specific, and cardiovascular mortality in the U.S. population: a large-scale pooled survey

BackgroundPrevious studies have reported a close association between the Geriatric Nutritional Risk Index (GNRI) and various conditions. However, the association between the GNRI and mortality remains unclear. To examine the correlation between the GNRI and all-cause, cancer-specific, and cardiovascular mortality, this study was performed.MethodsWe analyzed elderly participants in the National Health and Nutrition Examination Survey from 2005 to 2016. The GNRI was calculated using body mass index and serum albumin. Kaplan-Meier survival curves were drawn to compare the survival probability between the normal and decreased GNRI groups. Weighted multivariate Cox regression and restricted cubic spline (RCS) models were employed to determine the linear and non-linear associations of the GNRI with all-cause, cancer-specific, and cardiovascular mortality.ResultsA total of 3,276 participants were included in the analysis. The Kaplan-Meier survival curve showed that the decreased GNRI group had a lower survival probability for all-cause mortality and cancer-specific mortality (P < 0.001) but not for cardiovascular mortality (P > 0.05). In the full regression models, the decreased group had a higher risk of all-cause mortality (HR = 1.67, 95% CI = 1.21–2.30, P = 0.002), and cancer-specific mortality (HR = 2.20, 95% CI = 1.32–3.67, P = 0.003) than the normal group. For cardiovascular mortality, no significant association with GNRI (HR = 1.39, 95% CI = 0.60–3.22, P = 0.436) was detected. Notably, the RCS analysis identified a linear downward trend between the GNRI and all-cause, alongside cancer-specific mortalities (all P for overall < 0.05). The time-dependent Receiver Operating Characteristic (ROC) analysis unveiled the predictive power of the GNRI for 5-year all-cause mortality, cancer mortality, and cardiovascular mortality was 0.754, 0.757, and 0.836, respectively, after adjusting for covariates.ConclusionsIndividuals with a decreased GNRI had increased risks of all-cause, and cancer-specific mortality. There were linear associations of the GNRI with all-cause, and cancer-specific mortality. Nutritional status should be carefully monitored, which may improve the overall prognosis for the general population.

Association of volatile organic compound levels with chronic obstructive pulmonary diseases in NHANES 2013–2016

Volatile organic compounds (VOCs) represent a significant component of air pollution. However, studies evaluating the impact of VOC exposure on chronic obstructive pulmonary disease (COPD) have predominantly focused on single pollutant models. This study aims to comprehensively assess the relationship between multiple VOC exposures and COPD. A large cross-sectional study was conducted on 4983 participants from the National Health and Nutrition Examination Survey. Four models, including weighted logistic regression, restricted cubic splines (RCS), weighted quantile sum regression (WQS), and the dual-pollution model, were used to explore the association between blood VOC levels and the prevalence of COPD in the U.S. general population. Additionally, six machine learning algorithms were employed to develop a predictive model for COPD risk, with the model’s predictive capacity assessed using the area under the curve (AUC) indices. Elevated blood concentrations of benzene, toluene, ortho-xylene, and para-xylene were significantly associated with the incidence of COPD. RCS analysis further revealed a non-linear and non-monotonic relationship between blood levels of toluene and m-p-xylene with COPD prevalence. WQS regression indicated that different VOCs had varying effects on COPD, with benzene and ortho-xylene having the greatest weights. Among the six models, the Extreme Gradient Boosting (XGBoost) model demonstrated the strongest predictive power, with an AUC value of 0.781. Increased blood concentrations of benzene and toluene are significantly correlated with a higher prevalence of COPD in the U.S. population, demonstrating a non-linear relationship. Exposure to environmental VOCs may represent a new risk factor in the etiology of COPD.