Area Under The ROC Curve Research Articles

Lymphocyte to high density lipoprotein ratio can predict the short-term prognosis of hepatitis B virus-related acute-on-chronic liver failure patients

BackgroundAcute-on-chronic liver failure (ACLF) is a syndrome characterized by systemic inflammation, leading to high short-term mortality. The lymphocyte to high-density lipoprotein ratio (LHR) has been introduced as a novel marker of inflammation. However, its role as a prognostic inflammatory biomarker in the context of hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) has received limited attention. MethodsWe retrospectively included 272 patients with HBV-ACLF who met the definition of APALC. Data on clinical features and laboratory tests were collected from medical records within 24 h. Logistic regression was used to identify risk factors for poor short-term prognosis, and LHR-based prediction (LHRB) models were constructed based on risk factors. Furthermore, the accuracy of the LHRB model was validated through rigorous testing. ResultsIn the survival and death groups, there were statistical differences in their CTP, MELD, MELD-Na, COSSH-ACLF II scores, and LHR. Multivariate logistic regression identified seven predictors significantly associated with 28-day mortality. Furthermore, statistically significant differences in short-term mortality and certain clinical laboratory tests for poor prognosis were observed between the high and low LHR groups. To assess the predictive performance of various models in terms of short-term mortality, the area under the receiver operating characteristic curve (AUROC) was calculated. The AUROC values for the CTP, MELD, MELD-Na, COSSH-ACLF II, and LHRB models were found to be 0.725, 0.788, 0.772, 0.871, and 0.877, respectively. The results in the validation group were similar to those in the training group, and the validation results suggested excellent performance of the LHRB model. ConclusionLHR levels have the potential to serve as indicators for the prognosis of HBV-ACLF. Additionally, the recently developed LHRB model offers an accessible risk assessment tool.

Prediction of early clinical response to neoadjuvant chemotherapy in Triple-negative breast cancer: Incorporating Radiomics through breast MRI

This study assessed pretreatment breast MRI coupled with machine learning for predicting early clinical responses to neoadjuvant chemotherapy (NAC) in triple-negative breast cancer (TNBC), focusing on identifying non-responders. A retrospective analysis of 135 TNBC patients (107 responders, 28 non-responders) treated with NAC from January 2015 to October 2022 was conducted. Non-responders were defined according to RECIST guidelines. Data included clinicopathologic factors and clinical MRI findings, with radiomics features from contrast-enhanced T1-weighted images, to train a stacking ensemble of 13 machine learning models. For subgroup analysis, propensity score matching was conducted to adjust for clinical disparities in NAC response. The efficacy of the models was evaluated using the area under the receiver-operating-characteristic curve (AUROC) before and after matching. The model combining clinicopathologic factors and clinical MRI findings achieved an AUROC of 0.752 (95% CI 0.644–0.860) for predicting non-responders, while radiomics-based models showed 0.749 (95% CI 0.614–0.884). An integrated model of radiomics, clinicopathologic factors, and clinical MRI findings reached an AUROC of 0.802 (95% CI 0.699–0.905). After propensity score matching, the hierarchical order of key radiomics features remained consistent. Our study demonstrated the potential of using machine learning models based on pretreatment MRI to non-invasively predict TNBC non-responders to NAC.

Predicting Pseudomonas aeruginosa drug resistance using artificial intelligence and clinical MALDI-TOF mass spectra.

Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is widely used in clinical microbiology laboratories for bacterial identification but its use for detection of antimicrobial resistance (AMR) remains limited. Here, we used MALDI-TOF MS with artificial intelligence (AI) approaches to successfully predict AMR in Pseudomonas aeruginosa, a priority pathogen with complex AMR mechanisms. The highest performance was achieved for modern β-lactam/β-lactamase inhibitor drugs, namely, ceftazidime/avibactam and ceftolozane/tazobactam. For these drugs, the model demonstrated area under the receiver operating characteristic curve (AUROC) of 0.869 and 0.856, specificity of 0.925 and 0.897, and sensitivity of 0.731 and 0.714, respectively. As part of this work, we developed dynamic binning, a feature engineering technique that effectively reduces the high-dimensional feature set and has wide-ranging applicability to MALDI-TOF MS data. Compared to conventional feature engineering approaches, the dynamic binning method yielded highest performance in 7 of 10 antimicrobials. Moreover, we showcased the efficacy of transfer learning in enhancing the AUROC performance for 8 of 11 antimicrobials. By assessing the contribution of features to the model's prediction, we identified proteins that may contribute to AMR mechanisms. Our findings demonstrate the potential of combining AI with MALDI-TOF MS as a rapid AMR diagnostic tool for Pseudomonas aeruginosa.IMPORTANCEPseudomonas aeruginosa is a key bacterial pathogen that causes significant global morbidity and mortality. Antimicrobial resistance (AMR) emerges rapidly in P. aeruginosa and is driven by complex mechanisms. Drug-resistant P. aeruginosa is a major challenge in clinical settings due to limited treatment options. Early detection of AMR can guide antibiotic choices, improve patient outcomes, and avoid unnecessary antibiotic use. Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is widely used for rapid species identification in clinical microbiology. In this study, we repurposed mass spectra generated by MALDI-TOF and used them as inputs for artificial intelligence approaches to successfully predict AMR in P. aeruginosa for multiple key antibiotic classes. This work represents an important advance toward using MALDI-TOF as a rapid AMR diagnostic for P. aeruginosa in clinical settings.

NeoaPred: a deep-learning framework for predicting immunogenic neoantigen based on surface and structural features of peptide-human leukocyte antigen complexes.

Neoantigens, derived from somatic mutations in cancer cells, can elicit anti-tumor immune responses when presented to autologous T cells by human leukocyte antigen. Identifying immunogenic neoantigens is crucial for cancer immunotherapy development. However, the accuracy of current bioinformatic methods remains unsatisfactory. Surface and structural features of peptide-HLA class I (pHLA-I) complexes offer valuable insight into the immunogenicity of neoantigens. We present NeoaPred, a deep-learning framework for neoantigen prediction. NeoaPred accurately constructs pHLA-I complex structures, with 82.37% of the predicted structures showing an RMSD of < 1 Å. Using these structures, NeoaPred integrates differences in surface, structural, and atom group features between the mutant peptide and its wild-type counterpart to predict a foreignness score. This foreignness score is an effective factor for neoantigen prediction, achieving an AUROC (Area Under the Receiver Operating Characteristic Curve) of 0.81 and an AUPRC (Area Under the Precision-Recall Curve) of 0.54 in the test set, outperforming existing methods. The source code is released under an Apache v2.0 license and is available at the GitHub repository (https://github.com/Dulab2020/NeoaPred).

Genetic Risk, Lifestyle Adherence, and Risk of Developing Hyperuricaemia in a Japanese Population.

To investigate the inter-relationships among genetic risk, healthy lifestyle adherence, and hyperuricaemia susceptibility. This prospective cohort study was conducted with 7,241 hyperuricaemia-free individuals aged ≥ 20 years from the Tohoku Medical Megabank Community-based cohort study. A comprehensive lifestyle score included body mass index, smoking, drinking, and physical activity, and a polygenic risk score (PRS) was constructed based on uric acid loci from a previous genome-wide association study meta-analysis. A multiple logistic regression model was used to estimate the association between genetic risk, healthy lifestyle, and hyperuricaemia incidence and calculate the area under the receiver operating characteristic curve (AUROC). Hyperuricaemia was defined as a uric acid level ≥7.0 mg/dl or a self-reported history of hyperuricaemia. Of the 7,241 adults (80.7% females; mean [SD] age: 57.7 [12.6] years), 217 (3.0%) developed hyperuricaemia during 3.5 years of follow-up. Genetic risk correlated with hyperuricaemia development (P for interaction = 0.287), and lifestyle risks were independently associated. Those with a high genetic risk and poor lifestyle had the highest risk (odds ratio: 5.34; 95% confidence interval [CI]: 2.61-12.10). Although not statistically significant, incorporating the PRS in the model with lifestyle information improved predictive ability (AUROC = 0.771, 95% CI: 0.736-0.806 for lifestyle; AUROC = 0.785, 95% CI: 0.751-0.819 for lifestyle and PRS; p = 0.07). : A healthy lifestyle to prevent hyperuricaemia, irrespective of genetic risk, may mitigate the genetic risk. Genetic risk may complement lifestyle factors in identifying individuals at a heightened hyperuricaemia risk.

The performance relationship between the EQ-5D-5L composite "Anxiety/Depression" dimension and anxiety and depression symptoms in a large, general population sample.

This cross-sectional study aims to understand the relationship between responses on the Anxiety/Depression (A/D) dimension of the EQ-5D-5L and symptoms of anxiety and depression on the GAD-7 and PHQ-9 instruments. In doing so, we investigate the comparative performance of the dimension between diagnostic groups (i.e. anxiety (GAD-7); depression (PHQ-9); anxiety & depression versus none). We additionally investigate the discriminatory performance between sub-populations based on gender, age, education and self-reported chronic conditions. 19,902 general population participants completed a health survey in May/June 2020, from five European countries and the United States. Performance of A/D was calculated using the Area Under the Receiver Operating Characteristic curve (AUROC), and was compared to having anxiety (GAD-7 ≥ 8), depression (PHQ-9 ≥ 10) and both versus none for the total population and sub-populations. Several additional sensitivity analyses were conducted, including calculations of the optimal A/D cut-off. The performance in the total sample was good (AUROC > 0.8) and did not differ significantly between diagnostic groups. The performance differed significantly between the age groups, with worse performance in the younger groups, and differed between those with a singular chronic condition, with worse performance in those indicating having an anxiety or depression disorder. The performance did not differ significantly by gender, education, nor total chronic conditions. The A/D dimension captures symptoms of anxiety, depression or both equally well. Performance is worse in the younger population. Interpretation in those with a self-reported anxiety or depression disorder should be further investigated. This is the first-of-its-kind large population sample performance analysis, where we present evidence that the performance of the A/D dimension differs between ages, and thus intra-age comparative results may be flawed.

Quantitative assessment of early caries lesion activity using novel dye-enhanced fluorescence imaging

ObjectivesThis study aimed to evaluate the validity of the dye-enhanced quantitative light-induced fluorescence (DEQLF) method for assessing early enamel caries activity. MethodsSeventy extracted human teeth with early enamel caries on smooth surfaces were included. Two dentists evaluated the caries activity using the Nyvad system followed by the DEQLF method. The teeth were hydrated with distilled water for 60 s, dehydrated with compressed air for 10 s, and stained with 100 μM fluorescein sodium solution for 10 s. White and fluorescent images were captured using a QLF-D 2+ Billuminator. The change in fluorescence (ΔΔG) was calculated using image analysis software. Independent-sample t-tests were performed to evaluate the difference in ΔΔG between active and inactive lesions for both DEQLF and conventional quantitative light-induced fluorescence (QLF) methods. Receiver operating characteristic (ROC) curve analysis was used to assess the validity of ΔΔG for distinguishing lesion activity using the area under the ROC curve (AUROC). ResultsAmong the 70 caries lesions, 33 were active and 37 were inactive. Using the DEQLF method, the ΔΔG for active lesions (3.8 ± 5.6) was significantly higher than that for inactive lesions (1.0 ± 2.5) (P < 0.05). With the conventional QLF method, there was no significant difference in ΔΔG between active (-1.1 ± 1.7) and inactive (-1.3 ± 1.7) lesions. DEQLF-derived ΔΔG demonstrated an AUROC of 0.72, a sensitivity of 0.67, and a specificity of 0.76. ConclusionsApplying the DEQLF method to human teeth enabled the quantitative assessment of lesion activity based on dye penetration. DEQLF-derived ΔΔG values showed significant differences based on lesion activity status and demonstrated high validity in distinguishing lesion activity. Clinical significanceClinicians can use the DEQLF method, which involves applying a fluorescent dye for 10 s prior to conventional QLF, to objectively quantify and distinguish the activity status of early enamel caries, potentially replacing the traditional reliance on subjective visual assessments.

IL-2Rα is a potential biomarker for heart failure diagnosis of patients with end-stage renal disease and haemodialysis.

Heart failure (HF) is a leading cause of mortality in patients with end-stage renal disease (ESRD) undergoing haemodialysis. Identifying novel predictors of HF is essential for improving diagnostic precision and enhancing patient outcomes. This study included 68 participants from the Haemodialysis Centre at the Second Hospital of Tianjin Medical University. Clinical characteristics and echocardiographic data were collected and analysed. We measured the plasma of 44 cytokines to investigate their correlation with cardiac function and their potential as HF biomarkers. In the HF with reduced ejection fraction (HFrEF) group, the levels of several cytokines, including stem cell growth factor-β (SCGF-β), C-X-C motif chemokine 10 (CXCL10), interleukin-1α (IL-1α), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-16 (IL-16), interleukin-1 receptor antagonist protein (IL-1Ra), interferon-γ (IFN-γ), tumour necrosis factor-α (TNF-α), leukaemia inhibitory factor (LIF), C-C motif chemokine 3 (CCL3), interleukin-10 (IL-10), interleukin-2 receptor subunit alpha (IL-2Rα), tumour necrosis factor ligand superfamily member 10 (TNFSF10), macrophage colony-stimulating factor (M-CSF), granulocyte colony-stimulating factor (G-CSF) and stem cell factor (SCF), were significantly increased, while C-C motif chemokine 11 (CCL11)/eotaxin levels were decreased compared with those in the control group (P<0.05). Receiver operating characteristic (ROC) curve analysis highlighted TNF-α [area under the ROC curve (AUC)=0.85, odds ratio (OR)=1.080, 95% confidence interval (CI): 1.033-1.128, P=0.001], IFN-γ (AUC=0.84, OR=1.836, 95% CI: 1.289-2.615, P=0.003) and IL-2Rα (AUC=0.82, OR=1.022, 95% CI: 1.009-1.035, P=0.001) as excellent predictors for HFrEF in haemodialysis patients with ESRD, and they outperformed soluble suppression of tumourigenicity-2 (sST2) but slightly underperformed N-terminal pro-brain natriuretic peptide (NT-proBNP). IL-2Rα (AUC=0.77, OR=1.018, 95% CI: 1.007-1.030, P=0.001) demonstrated superior diagnostic capabilities when distinguishing patients with HF with left ventricular ejection fraction (LVEF) <50% from controls. IL-2Rα emerged as a robust biomarker for left ventricular HF, while TNF-α (AUC=0.89, OR=1.140, 95% CI: 1.039-1.250, P=0.005) showed promise in assessing HF severity in patients with ESRD. IL-2Rα (AUC=0.80, OR=1.017, 95% CI: 1.007-1.027, P=0.001) also significantly predicted right ventricular systolic dysfunction. During a median follow-up of 14months, 10 patients (14.7%) experienced all-cause mortality. Multivariate Cox regression analysis confirmed that plasma IL-2Rα was an independent predictor of all-cause death [hazard ratio (HR): 1.010, 95% CI: 1.001-1.020, P=0.039] after adjusting for other variables. This study underscores the potential of IL-2Rα as a valuable biomarker for HF diagnosis and management in haemodialysis patients with ESRD and contributes to our understanding of this high-risk population.

Snow Avalanche Susceptibility Mapping of Transportation Corridors Based on Coupled Certainty Factor and Geodetector Models

Avalanche susceptibility assessment is a core aspect of regional avalanche early warning and risk analysis and is of great significance for disaster prevention and mitigation on proposed highways. Using sky–ground integration investigation, 83 avalanche points within the G219 Wen Quan to Horgos transportation corridor were identified, and the avalanche hazard susceptibility of the transportation corridor was partitioned using the certainty factor (CF) model and the coupled coefficient of the certainty factor–Geodetector (CF-GD) model. The CF model analysis presented nine elements of natural conditions which influence avalanche development; then, by applying the Geodetector for each of the factors, a weighting coefficient was given depending on its importance for avalanche occurrence. The results demonstrate the following: (1) According to the receiver operating characteristic (ROC) curve used to verify the accuracy, the area under the ROC curve (AUC) value for the CF-GD coupled model is 0.889, which is better than the value of 0.836 of the CF model’s evaluation accuracy, and the coupled model improves the accuracy by about 6.34% compared with the single model, indicating that the coupled model is more accurate. The results provide avalanche prevention and control recommendations for the G219 Wen Quan to Horgos transportation corridor. (2) The slope orientation, slope gradient, and mean winter temperature gradient are the main factors for avalanche development in the study area. (3) The results were validated based on the AUC values. The AUCs of the CF-GD coupled model and the CF model were 0.889 and 0.836, respectively. The accuracy of the coupled model was improved by about 6.34% compared to the single model, and the coupled CF-GD model was more accurate. The results provide avalanche control recommendations for the G219 Wen Quan to Horgos transportation corridor.

The relationship between severity of periapical periodontitis and next-generation systemic inflammatory biomarkers in children with early childhood caries

IntroductionEarly childhood caries (ECC) is closely associated with poor oral hygiene and cariogenic diet. Untreated ECC results in recurrent odontogenic infections and local and systemic consequences. In this study, our goal is to assess the relationship between the intensity of odontogenic infection-associated periapical periodontitis and new generation of systemic inflammatory markers (SII, NLR, PLR) in ECC-affected children.Material and method95 healthy patients in early childhood and demonstrating periapical periodontitis who underwent dental treatment under general anesthesia (GA) in the last two years were included in the present study. Their periapical statuses were dichotomized as mild and severe. Periapical Index (PAI) scores of 2 and 3 were regarded as “mild” whereas 4 and 5 as “severe”. Of the complete blood test (CBC) parameters, systemic inflammatory index (SII), neutrophil-to-lymphocyte ratio (NLR), platelet–lymphocyte ratio (PLR), neutrophilic granulocyte (NEUT), lymphocyte (LYMPH) and platelet (PLT) were recorded. The relationship between the degree of periapical pathology and the evaluated markers was assessed using Receiver Operating Characteristic (ROC) analysis.ResultsResults of the present study revealed that mean NLR, SII and NEUT index scores of the patients having severe periapical periodontitis were statistically higher than those of the ones with mild pathology (p < 0.05). A positive, statistically significant interrelationship was found between the number of teeth demonstrating a PAI score of severe periapical periodontitis with the signs of exacerbation (PAI 5) and NLR and SII values (p < 0.05). Area under the ROC curve (AUC) values for NLR and SII were determined as 66.8% and 66.6% respectively, indicating that classification performance was sufficient and statistically significant (p < 0.05).ConclusionPostponing the management of odontogenic infections will induce some complications such as, infective endocarditis and cause the systemic inflammatory process to continue by aggravating the systemic effects of local lesions. Thus, underlying mechanism should be eliminated and oral hygiene should be maintained, also novel biomarkers may be recommended to be used for the decision-making process for the teeth with persistent periapical lesions unresponsive to treatment.

FP2.3 - Is the Oakland score a reliable tool for safe discharge in lower gastrointestinal bleed patients?

Abstract Aim There are no NICE guidelines regarding safe discharge of patients with lower gastrointestinal bleeds (LGIB). The aim of this study was to investigate the effectiveness and safety of the Oakland score, as suggested by the British Society of Gastroenterology guidelines, in patients presenting with LGIB at William Harvey hospital. Methods Patients with LGIB who presented to A&amp;E or inpatient referral from January to December 2023 were included in this retrospective study. Data was extracted from patient’s Sunrise documentation. The Oakland score for each patient was calculated. Those with a score of ≤8 were deemed safe for discharge; those with a higher score were deemed unsuitable. Patients’ admission, safe discharges and adverse outcomes, such as representation, blood transfusion, or further intervention, were investigated. The area under the receiver-operating characteristic curve (AUROC) for the Oakland score and adverse outcome was calculated. Results 123 patients were included. A total of 144 presentations to hospital, 29 had an Oakland score ≤8; 17 (58.6%) cases were safely discharged, 8 (27.6%) admitted and 4 (13.8%) represented. 115 had a score &amp;gt;8; 37 (32.2%) were safely discharged, 72 (62.6%) admitted and 41 (35.7%) experienced at least one adverse outcome. The AUROC for safe discharge was 0.84. Conclusion The Oakland score seems to be a safe and reliable tool for identifying LGIB patients who could be safely discharged home without hospital intervention. However, further research is required to assess whether a score &amp;gt;8 could be used as many higher risk patients did not experience adverse outcomes.

An Artificial Neural Network Prediction Model of Depressive Symptoms among Women with Abnormal Papanicolaou Smear Results before and after Diagnostic Procedures.

(1) Background: Cervical screening and additional diagnostic procedures often lead to depression. This research aimed to develop a prediction model for depression in women who received an abnormal Papanicolaou screening test, prior to and following the diagnostic procedures. (2) Methods: The study included women who had a positive Papanicolaou screening test (N = 172) and attended the Clinical Center of Kragujevac in Serbia for additional diagnostic procedures (colposcopy/biopsy/endocervical curettage). Women filled out a sociodemographic survey and the Center for Epidemiologic Studies Depression questionnaire (CES-D scale) before and after diagnostic procedures. A prediction model was built with multilayer perceptron neural networks. (3) Results: A correlation-based filter method of feature selection indicated four variables that correlated with depression both prior to and following the diagnostic procedures-anxiety, depression, worry, and concern about health consequences. In addition, the use of sedatives and a history of both induced and spontaneous abortion correlated with pre-diagnostic depression. Important attributes for predicting post-diagnostic depression were scores for the domains 'Tension/discomfort' and 'Embarrassment' and depression in personal medical history. The accuracy of the pre-diagnostic procedures model was 70.6%, and the area under the receiver operating characteristic curve (AUROC) was 0.668. The model for post-diagnostic depression prediction showed an accuracy of 70.6%, and an AUROC = 0.836. (4) Conclusions: This study helps provide means to predict the occurrence of depression in women with an abnormal Papanicolaou screening result prior to and following diagnostic procedures, which can aid healthcare professionals in successfully providing timely psychological support to those women who are referred to further diagnostics.

Prognostic Value of the Lactate/Albumin Ratio for 28-Day Mortality in Pediatric Septic Shock: A Prospective Cohort Study.

Background Septic shock remains a leading cause of mortality in children. The lactate/albumin ratio (LAR) has emerged as a potential prognostic marker for mortality in septic shock, yet most existing research focuses on adults, with limited data available for pediatric populations, particularly in Vietnam. Objectives This study aims to evaluate the prognostic utility of the LAR in predicting 28-day mortality among children aged two months to 15 years with septic shock in Vietnam. Methods We conducted a prospective cohort study involving children diagnosed with septic shock at the largest pediatric intensive care unit (PICU) in the Mekong Delta, Vietnam, from July 2022 to June 2024. Clinical and laboratory parameters, including lactate and albumin levels, were measured at the time of septic shock diagnosis. Patients were followed for 28 days, with outcomes categorized as either survival or mortality. The prognostic performance of LAR was assessed through its discrimination and calibration capabilities. Results The 28-day mortality rate was 63.4%. LAR was significantly higher in non-survivors compared to survivors (p < 0.001). The area under the receiver operating characteristic curve (AUROC) for LAR was 0.91, indicating superior discriminatory power compared to lactate alone and comparable to albumin. Using a Youden index-derived cut-off of 1.84, LAR demonstrated a sensitivity of 84.6% and a specificity of 80%. Kaplan-Meier analysis and log-rank testing revealed significantly lower survival probabilities in children with LAR ≥1.84 (p < 0.05). The Hosmer-Lemeshow test confirmed good calibration of LAR in mortality prediction (p > 0.05). Conclusion The lactate/albumin ratio exhibits excellent discriminatory and calibration properties, making it a valuable tool for predicting 28-day mortality in pediatric septic shock. This ratio should be considered for routine use in clinical practice to improve prognostic assessments in this vulnerable population.

Prediction of chemical-induced acute toxicity using in vitro assay data and chemical structure

Exposure to various chemicals found in the environment and in the context of drug development can cause acute toxicity. To provide an alternative to in vivo animal toxicity testing, the U.S. Tox21 consortium developed in vitro assays to test a library of approximately 10,000 drugs and environmental chemicals (Tox21 10K compound library) in a quantitative high-throughput screening (qHTS) approach. In this study, we assessed the utility of Tox21 assay data in comparison with chemical structure information in predicting acute systemic toxicity. Prediction models were developed using four machine learning algorithms, namely Random Forest, Naïve Bayes, eXtreme Gradient Boosting, and Support Vector Machine, and their performance was assessed using the area under the receiver operating characteristic curve (AUC-ROC). The chemical structure-based models as well as the Tox21 assay data demonstrated good predictive power for acute toxicity, achieving AUC-ROC values ranging from 0.83 to 0.93 and 0.73 to 0.79, respectively. We applied the models to predict the acute toxicity potential of the compounds in the Tox21 10K compound library, most of which were found to be non-toxic. In addition, we identified the Tox21 assays that contributed the most to acute toxicity prediction, such as acetylcholinesterase (AChE) inhibition and p53 induction. Chemical features including organophosphates and carbamates were also identified to be significantly associated with acute toxicity. In conclusion, this study underscores the utility of in vitro assay data in predicting acute toxicity.

Machine-learning-derived online prediction models of outcomes for patients with cholelithiasis-induced acute cholangitis: development and validation in two retrospective cohorts

Cholelithiasis-induced acute cholangitis (CIAC) is an acute inflammatory disease with poor prognosis. This study aimed to create machine-learning (ML) models to predict the outcomes of patients with CIAC. In this retrospective cohort and ML study, patients who met the both diagnosis of 'cholangitis' and 'calculus of gallbladder or bile duct' according to the International Classification of Disease (ICD) 9th revision, or met the diagnosis of 'calculus of bile duct with acute cholangitis with or without obstruction' according to the ICD 10th revision during a single hospitalization were included from the Medical Information Mart for Intensive Care database, which records patient admissions to Beth Israel Deaconess Medical Center, MA, USA, spanning June 1, 2001 to November 16, 2022. Patients who were neither admitted in an emergency department nor underwent biliary drainage within 24h after admission, had an age of less than 18, or lost over 20% of the information were excluded. Nine ML methods, including the Logistic Regression, eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine, Adaptive Boosting, Decision Tree, Gradient Boosting Decision Tree, Gaussian Naive Bayes, Multi-Layer Perceptron, and Support Vector Machine were applied for prediction of in-hospital mortality, re-admission within 30 days after discharge, and mortality within 180 days after discharge. Patients from Zhongda Hospital affiliated to Southeast University in China between January 1, 2019 and July 30, 2023 were enrolled as an external validation set. The area under the receiver operating characteristic curve (AUROC) was the main index for model performance assessment. A total of 1156 patients were included to construct models. We performed stratified analyses on all patients, patients admitted to the intensive care unit (ICU) and those who underwent biliary drainage during ICU treatment. 13-16 features were selected from 186 variables for model training. The XGBoost method demonstrated the most optimal predictive efficacy, as evidenced by training set AUROC of 0.996 (95% CI NaN-NaN) for in-hospital mortality, 0.886 (0.862-0.910) for re-admission within 30 days after discharge, and 0.988 (0.982-0.995) for mortality within 180 days after discharge in all patients, 0.998 (NaN-NaN), 0.933 (0.909-0.957), and 0.988 (0.983-0.993) in patients admitted to the ICU, 0.987 (0.970-0.999), 0.908 (0.873-0.942), and 0.982 (0.971-0.993) in patients underwent biliary drainage during ICU treatment, respectively. Meanwhile, in the internal validation set, the AUROC reached 0.967 (0.933-0.998) for in-hospital mortality, 0.589 (0.502-0.677) for re-admission within 30 days after discharge, and 0.857 (0.782-0.933) for mortality within 180 days after discharge in all patients, 0.963 (NaN-NaN), 0.668 (0.486-0.851), and 0.864 (0.757-0.970) in patients admitted to the ICU, 0.961 (0.922-0.997), 0.669 (0.540-0.799), and 0.828 (0.730-0.925) in patients underwent biliary drainage during ICU treatment, respectively. The AUROC values of external validation set consisting of 61 patients were 0.741 (0.725-0.763), 0.812 (0.798-0.824), and 0.848 (0.841-0.859), respectively. The XGBoost models could be promising tools to predict outcomes in patients with CIAC, and had good clinical applicability. Multi-center validation with a larger sample size is warranted. The Technological Development Program of Nanjing Healthy Commission, and Zhongda Hospital Affiliated to Southeast University, Jiangsu Province High-Level Hospital Construction Funds.

Decrease of haemoconcentration reliably detects hydrostatic pulmonary oedema in dyspnoeic patients in the emergency department – a machine learning approach

BackgroundHaemoglobin variation (ΔHb) induced by fluid transfer through the intestitium has been proposed as a useful tool for detecting hydrostatic pulmonary oedema (HPO). However, its use in the emergency department (ED) setting still needs to be determined.MethodsIn this observational retrospective monocentric study, ED patients admitted for acute dyspnoea were enrolled. Hb values were recorded both at ED presentation (T0) and after 4 to 8 h (T1). ΔHb between T1 and T0 (ΔHbT1-T0) was calculated as absolute and relative value. Two investigators, unaware of Hb values, defined the cause of dyspnoea as HPO and non-HPO. ΔHbT1-T0 ability to detect HPO was evaluated. A machine learning approach was used to develop a predictive tool for HPO, by considering the ability of ΔHb as covariate, together with baseline patient characteristics.ResultsSeven-hundred-and-six dyspnoeic patients (203 HPO and 503 non-HPO) were enrolled over 19 months. Hb levels were significantly different between HPO and non-HPO patients both at T0 and T1 (p < 0.001). ΔHbT1-T0 were more pronounced in HPO than non-HPO patients, both as relative (-8.2 [-11.2 to -5.6] vs. 0.6 [-2.1 to 3.3] %) and absolute (-1.0 [-1.4 to -0.8] vs. 0.1 [-0.3 to 0.4] g/dL) values (p < 0.001). A relative ΔHbT1-T0 of -5% detected HPO with an area under the receiver operating characteristic curve (AUROC) of 0.901 [0.896–0.906]. Among the considered models, Gradient Boosting Machine showed excellent predictive ability in identifying HPO patients and was used to create a web-based application. ΔHbT1-T0 was confirmed as the most important covariate for HPO prediction.ConclusionsΔHbT1-T0 in patients admitted for acute dyspnoea reliably identifies HPO in the ED setting. The machine learning predictive tool may represent a performing and clinically handy tool for confirming HPO.

Sphingolipid metabolites as potential circulating biomarkers for sarcopenia in men.

Sarcopenia is an age-related progressive loss of muscle mass and function. Sarcopenia is a multifactorial disorder, including metabolic disturbance; therefore, metabolites may be used as circulating biomarkers for sarcopenia. We aimed to investigate potential biomarkers of sarcopenia using metabolomics. After non-targeted metabolome profiling of plasma from mice of an aging mouse model of sarcopenia, sphingolipid metabolites and muscle cells from the animal model were evaluated using targeted metabolome profiling. The associations between sphingolipid metabolites identified from mouse and cell studies and sarcopenia status were assessed in men in an age-matched discovery (72 cases and 72 controls) and validation (36 cases and 128 controls) cohort; women with sarcopenia (36 cases and 36 controls) were also included as a discovery cohort. Both non-targeted and targeted metabolome profiling in the experimental studies showed an association between sphingolipid metabolites, including ceramides (CERs) and sphingomyelins (SMs), and sarcopenia. Plasma SM (16:0), CER (24:1), and SM (24:1) levels in men with sarcopenia were significantly higher in the discovery cohort than in the controls (all P<0.05). There were no significant differences in plasma sphingolipid levels for women with or without sarcopenia. In men in the discovery cohort, an area under the receiver-operating characteristic curve (AUROC) of SM (16:0) for low muscle strength and low muscle mass was 0.600 (95% confidence interval [CI]: 0.501-0.699) and 0.647 (95% CI: 0.557-0.737). The AUROC (95% CI) of CER (24:1) and SM (24:1) for low muscle mass in men was 0.669 (95% CI: 0.581-0.757) and 0.670 (95% CI: 0.582-0.759), respectively. Using a regression equation combining CER (24:1) and SM (16:0) levels, a sphingolipid (SphL) score was calculated; an AUROC of the SphL score for sarcopenia was 0.712 (95% CI: 0.626-0.798). The addition of the SphL score to HGS significantly improved the AUC from 0.646 (95% CI: 0.575-0.717; HGS only) to 0.751 (95% CI: 0.671-0.831, P=0.002; HGS+SphL) in the discovery cohort. The predictive ability of the SphL score for sarcopenia was confirmed in the validation cohort (AUROC=0.695, 95% CI: 0.591-0.799). SM (16:0), reflecting low muscle strength, and CER (24:1) and SM (16:0), reflecting low muscle mass, are potential circulating biomarkers for sarcopenia in men. Further research on sphingolipid metabolites is required to confirm these results and provide additional insights into the metabolomic changes relevant to the pathogenesis and diagnosis of sarcopenia.

Clinical value of fibroblast growth factor 19 in predicting gastrointestinal dysfunction in patients with sepsis.

To assess the potential value of fibroblast growth factor 19 (FGF19) as a predictor of gastrointestinal (GI) dysfunction in patients with sepsis. A prospective study was conducted, and 209 patients who were diagnosed with sepsis and admitted to the intensive care unit (ICU) at teaching hospitals in China were enrolled from June 2023 to December 2023. The serum FGF19 level was determined at ICU admission. The differences in serum FGF19 levels between the two groups were compared via the Mann-Whitney U test, and Spearman's correlation coefficient was used to identify the correlations of the FGF19 concentration with other clinical variables and biomarkers. Receiver operating characteristic (ROC) analysis was used to determine the value of FGF19 in predicting GI dysfunction in patients with sepsis. The total ICU mortality rate was 13.3% (24/180). There was a tendency toward increased ICU mortality in patients with sepsis-associated GI dysfunction compared with patients without GI dysfunction with statistical significance (21.9% vs. 8.6%, p = 0.031). Serum FGF19 levels were significantly higher in patients with sepsis-associated GI dysfunction than in patients without GI dysfunction [355.1 (37.2, 2315.4) μg/mL vs. 127.4 (5.7, 944.2) μg/mL, p = 0.003]. The results of receiver operating characteristic (ROC) curve analysis revealed that the area under the ROC curve (AUC) for the ability of FGF19 to predict GI dysfunction in patients with sepsis was 0.773 (95% CI 0.712 ~ 0.827), which was greater than the predictive capacity of PCT [AUC = 0.632 (95% CI 0.562 ~ 0.804)]. Serum FGF19 could be considered as a novel predictor or biomarker of GI dysfunction in patients with sepsis.

A general prediction model for compound-protein interactions based on deep learning.

The identification of compound-protein interactions (CPIs) is crucial for drug discovery and understanding mechanisms of action. Accurate CPI prediction can elucidate drug-target-disease interactions, aiding in the discovery of candidate compounds and effective synergistic drugs, particularly from traditional Chinese medicine (TCM). Existing in silico methods face challenges in prediction accuracy and generalization due to compound and target diversity and the lack of largescale interaction datasets and negative datasets for model learning. To address these issues, we developed a computational model for CPI prediction by integrating the constructed large-scale bioactivity benchmark dataset with a deep learning (DL) algorithm. To verify the accuracy of our CPI model, we applied it to predict the targets of compounds in TCM. An herb pair of Astragalus membranaceus and Hedyotis diffusaas was used as a model, and the active compounds in this herb pair were collected from various public databases and the literature. The complete targets of these active compounds were predicted by the CPI model, resulting in an expanded target dataset. This dataset was next used for the prediction of synergistic antitumor compound combinations. The predicted multi-compound combinations were subsequently examined through in vitro cellular experiments. Our CPI model demonstrated superior performance over other machine learning models, achieving an area under the Receiver Operating Characteristic curve (AUROC) of 0.98, an area under the precision-recall curve (AUPR) of 0.98, and an accuracy (ACC) of 93.31% on the test set. The model's generalization capability and applicability were further confirmed using external databases. Utilizing this model, we predicted the targets of compounds in the herb pair of Astragalus membranaceus and Hedyotis diffusaas, yielding an expanded target dataset. Then, we integrated this expanded target dataset to predict effective drug combinations using our drug synergy prediction model DeepMDS. Experimental assay on breast cancer cell line MDA-MB-231 proved the efficacy of the best predicted multi-compound combinations: Combination I (Epicatechin, Ursolic acid, Quercetin, Aesculetin and Astragaloside IV) exhibited a half-maximal inhibitory concentration (IC50) value of 19.41μM, and a combination index (CI) value of 0.682; and Combination II (Epicatechin, Ursolic acid, Quercetin, Vanillic acid and Astragaloside IV) displayed a IC50 value of 23.83μM and a CI value of 0.805. These results validated the ability of our model to make accurate predictions for novel CPI data outside the training dataset and evaluated the reliability of the predictions, showing good applicability potential in drug discovery and in the elucidation of the bioactive compounds in TCM. Our CPI prediction model can serve as a useful tool for accurately identifying potential CPI for a wide range of proteins, and is expected to facilitate drug research, repurposing and support the understanding of TCM.

Diagnostic value of artificial intelligence-assisted CTA for the assessment of atherosclerosis plaque: a systematic review and meta-analysis.

Artificial intelligence (AI) has increasingly been applied to computed tomography angiography (CTA) images to aid in the assessment of atherosclerotic plaque. Our aim was to explore the diagnostic accuracy of AI-assisted CTA for plaque diagnosis and classification through a systematic review and meta-analysis. A systematic literature review was performed by searching PubMed, EMBASE, and the Cochrane Library according to PRISMA guidelines. Original studies evaluating the diagnostic accuracy of radiomics, machine-learning, or deep-learning techniques applied to CTA images for detecting stenosis, calcification, or plaque vulnerability were included. The quality and risk of bias of the included studies were evaluated using the QUADAS-2 tool. The meta-analysis was conducted using STATA software (version 17.0) to pool sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) to determine the overall diagnostic performance. A total of 11 studies comprising 1,484 patients were included. There was low risk of bias and substantial heterogeneity. The overall pooled AUROC for atherosclerotic plaque assessment was 0.96 [95% confidence interval (CI) 0.94-0.97] across 21 trials. Of these, for ≥50% stenosis detection, the AUROC was 0.95 (95% CI 0.93-0.96) in five studies. For identifying ≥70% stenosis, the AUROC was 0.96 (95% CI 0.94-0.97) in six studies. For calcium detection, the AUROC was 0.92 (95% CI 0.90-0.94) in six studies. Our meta-analysis demonstrates that AI-assisted CTA has high diagnostic accuracy for detecting stenosis and characterizing plaque composition, with optimal performance in detecting ≥70% stenosis. https://www.crd.york.ac.uk/, PROSPERO, identifier (CRD42023431410).