Predictive Test Research Articles

Deep learning-based superconductivity prediction and experimental tests

The discovery of novel superconducting materials is a long-standing challenge in materials science, with a wealth of potential for applications in energy, transportation and computing. Recent advances in artificial intelligence (AI) have enabled expediting the search for new materials by efficiently utilizing vast materials databases. In this study, we developed an approach based on deep learning (DL) to predict new superconducting materials. We have synthesized a compound derived from our DL network and confirmed its superconducting properties in agreement with our prediction. Our approach is also compared to previous work based on random forests (RFs). In particular, RFs require knowledge of the chemical properties of the compound, while our neural net inputs depend solely on the chemical composition. With the help of hints from our network, we discover a new ternary compound Mo20Re6Si4, which becomes superconducting below 5.4 K. We further discuss the existing limitations and challenges associated with using AI to predict and, along with potential future research directions.

Fatigue Life Analysis of Titanium Torsion Spring Based on Continuous Damage Mechanics.

In this study, a titanium alloy torsional spring used in aviation was taken as the research subject. Aiming at the fatigue life prediction problem of this spring, the life analysis of the titanium alloy torsional spring was performed using a customized UMAT subroutine based on the theory of continuous damage mechanics. Several sets of life prediction models and tests were compared. The fatigue lives of the springs at 60, 80, 100, and 120 degrees were 45,070, 65,067, 99,677, and 181,322 cycles, respectively. Compared with other fatigue life prediction methods, the fatigue life calculated by the customized subroutine was the most consistent with the fatigue life of the titanium alloy torsion spring tests. The average relative error between the measured experimental life value and the predicted value was 2.04%, which is less than 5%, meeting engineering measurement requirements. The effectiveness and applicability of the proposed model and method were verified, and the time and economic cost caused by excessively long experimental cycles were reduced. This helps improve the accuracy of fatigue life prediction for this titanium alloy torsional spring and provides analysis support for subsequent structural optimization and improvement.

Quantification of Free Circulating DNA and Differential Methylation Profiling of Selected Genes as Novel Non-Invasive Biomarkers for Endometriosis Diagnosis.

Endometriosis is a chronic, estrogen-dependent disorder associated with the presence of endometrial cells mainly in the pelvic cavity, causing systemic immune inflammation, infertility, epigenetic dysregulation of differential DNA methylation, coelomic metaplasia, and pain. It affects approximately 10-12% of women. Despite decades of research, full pathophysiology, a diagnostic roadmap, and clinical management strategies for endometriosis are not yet fully elucidated. Cell-free DNA (Cf-DNA) in the peripheral blood of diseased and healthy individuals was discovered in the 1950s. Quantifying peripheral Cf-DNA and the specific differential methylation of a group of genes have been proposed as potential non-invasive diagnostic biomarkers for somatic and constitutional genetics and for various other pathological disorders. In this study, we investigated the Cf-DNA levels of 78 young women, 38 of whom had endometriosis confirmed via laparoscopy and 40 of whom were healthy. We found a significant difference between the two groups when Cf-DNA was quantified, with 3.9 times more Cf-DNA in the serum of women with endometriosis. We also identified nine target genes potentially involved in the pathogenesis of endometriosis, with a different methylation profile between the two groups. Our data suggest that the combination of cell-free DNA quantification and the assessment of the epigenetic signature of differential methylation of nine genes can be proposed as a non-invasive predictive and diagnostic test for endometriosis.

The skin allergy risk assessment-integrated chemical environment (SARA-ICE) defined approach to derive points of departure for skin sensitization.

Mechanistically based non-animal methods for assessing skin sensitization hazard have been developed, but are not considered sufficient, individually, to conclusively define the skin sensitization potential or potency of a chemical. This resulted in the development of defined approaches (DAs), as documented in OECD TG 497, for combining information sources in a prescriptive manner to provide a determination of risk or potency. However, there are currently no DAs within OECD TG 497 that can derive a point of departure (POD) for risk assessment. The Skin Allergy Risk Assessment - Integrated Chemical Environment (SARA-ICE) DA for skin sensitization is a Bayesian statistical model that estimates a human-relevant metric of sensitizer potency, the ED01, an estimate of the human predictive patch test dermal dose at which there is 1% chance of inducing sensitization, which can be used in a risk assessment paradigm. The model accounts for variability of input data and explicitly quantifies uncertainty. SARA-ICE derives the ED01 from a variety of in vitro and in vivo test method data and is built upon historical human, murine, and in vitro test data for 434 chemicals. In addition to the ED01 POD SARA-ICE DA also provides a Globally Harmonized System of Classification and Labelling of Chemicals (GHS) classification probability for GHS subcategories 1A, 1B and not classified (NC). Here we describe the SARA-ICE model and its evaluation, including performance versus benchmark PODs. In addition, via a case study with isothiazolinones (ITs), we demonstrate the utility of SARA-ICE for integrating different data inputs and compare the ED01 for six ITs to existing historical data.

Engaging adolescents and young adults in decisions about return of genomic research results: study protocol for a mixed-methods longitudinal clinical trial protocol

BackgroundTo protect minors’ future autonomy, professional organizations have historically discouraged returning predictive adult-onset genetic test results and carrier status to children. Recent clinical guidance diverges from this norm, suggesting that when minors have genomic sequencing performed for clinical purposes, parents and children should have the opportunity to learn secondary findings, including for some adult-onset conditions. While parents can currently opt in or out of receiving their child’s secondary findings, the American Society of Human Genetics Workgroup on Pediatric Genetic and Genomic Testing suggests including adolescents in the decision-making process. However, it is not clear what factors young people consider when given the opportunity to learn genetic findings for themselves. In this manuscript, we report on the methods for a clinical trial that examines adolescents’, young adults’, and their parents’ decisions about learning genomic information for the adolescent or young adult.MethodsWe are enrolling assenting (ages 13–17) adolescents and consenting (ages 18–21) young adults in a prospective genomic screening study to assess the choices they make about receiving individual genomic results. Participants use an online tool to indicate whether they want to learn their personal genetic risk for specific preventable, treatable, and adult-onset conditions, as well as carrier status for autosomal recessive conditions. We are examining (1) how choices differ between adolescent and young adult cohorts (as well as between adolescents/young adults and parents) and (2) decisional conflict and stability across study timepoints. Results are returned based on participants’ choices. Qualitative interviews with a subset of participants explore decisional stability, adolescent/young adult engagement with parents in decision-making, and the impact of learning pathogenic/likely pathogenic and autosomal recessive carrier results.DiscussionThis study explores decision making and decision stability between adolescents and parents (where applicable), as well as the ethical implications and impact of return of clinical-grade genetic research results to adolescents and young adults. The results of this study will contribute empirical evidence to support best practices and guidance on engaging young people in genomic research studies and clinical care that offer return of results.Trial registrationClinicalTrials.gov Identifier: NCT04481061. Registered 22 July 2020.

Point-of-Care Tests for Preeclampsia: Systematic Review and Meta-Analysis of Diagnostic Test Accuracy Studies.

Preeclampsia is a common condition associated with significant maternal and foetal complications. The diagnosis of preeclampsia is based on clinical criteria, with recent introduction of angiogenic markers as an aid to clinical triage. Several biomarkers are now available in a point-of-care test format, which provide swift results to support clinical decision-making. To determine sensitivity and specificity of available point-of-care tests used for preeclampsia prediction. Systematic searches of databases (PubMed/Medline, Ovid Embase) from inception till May 2024 were done to identify relevant studies. we included studies reporting on the diagnostic test accuracy of any point-of-care test for short-term prediction of preeclampsia in the second or third trimester. Data were extracted using pre-designed template, and statistical analysis was performed to pool sensitivity and specificity and a hierarchal summary receiver operating characteristic curve model was used. Quality of included studies and risk of bias were assessed following the QUADAS-2 tool. We identified 17 studies that met the inclusion criteria, six of which described the use of glycosylated fibronectin, five with Congo red, two with placental growth factor (PlGF) and one study each investigated nephrin, misfolded protein, CD44/FKBPL ratio and inositol phosphoglycan-P. Meta-analysis was possible only for glycosylated fibronectin with a pooled sensitivity of 0.80 (95% CI: 0.51-0.94) and a specificity of 0.84 (95% CI: 0.71-0.93). Five studies were included in meta-analysis, all of which had a low risk of bias on the QUADAS assessment. These findings suggest that the point-of-care glycosylated fibronectin testing might be beneficial in clinical triage for preeclampsia in an at-risk population.

The MAPE Analysis of Arima (p,d,q) on LQ45 Stock Price to Determine Training Data Period

Most of the research using the Arima (p,d,q) focused on the accuracy of prediction results. Unlike other research, this work examines the training data period suitable for modeling ARIMA (p,d,q) in stock prices. Due to the volatile movement of stocks, the number of training data is assumed to affect the LQ45 prediction results. This research used five kinds of training data, including daily data for up to 5 years. With these five types of data series, the Arima (p,d,q) was made for LQ45 stocks. The prediction was conducted for two months after obtaining the model 5 data series of LQ45 stocks. Two months of data were used for January and February 2021 prediction test data. The results of this prediction were compared with the test data to produce the MAPE value. Based on the observations and calculation results, the most suitable stock to use the Arima (p,d,q) was ASII. In 5 years, the stocks produced the lowest MAPE value of 0.05%. Relatively stable LQ45 stocks with no change in the Arima (p,d,q) using four consecutive data series were ACES, CTRA, INTP, MIKA, and TLKM. Based on the MAPE value analysis performed in this study, we concluded that the best period to use the Arima (p,d,q) for LQ45 stocks is two years, with a median error rate of only 6.0091%.

Evaluation of first trimester serum double marker test in prediction of fetal growth restriction

Evaluation of first trimester serum double marker test in prediction of fetal growth restriction

Prospective randomized controlled trial study of liver function tests in unstable angina risk prediction

Prospective randomized controlled trial study of liver function tests in unstable angina risk prediction

EE21 Towards Recommendations for Cost-Effectiveness Analysis of Predictive, Prognostic, and Serial Biomarker Tests in Oncology

EE21 Towards Recommendations for Cost-Effectiveness Analysis of Predictive, Prognostic, and Serial Biomarker Tests in Oncology

Automation of Drug Discovery through Cutting-edge In-silico Research in Pharmaceuticals: Challenges and Future Scope.

The rapidity and high-throughput nature of in silico technologies make them advantageous for predicting the properties of a large array of substances. In silico approaches can be used for compounds intended for synthesis at the beginning of drug development when there is either no or very little compound available. In silico approaches can be used for impurities or degradation products. Quantifying drugs and related substances (RS) with pharmaceutical drug analysis (PDA) can also improve drug discovery (DD) by providing additional avenues to pursue. Potential future applications of PDA include combining it with other methods to make insilico predictions about drugs and RS. One possible outcome of this is a determination of the drug potential of nontoxic RS. ADME estimation, QSAR research, molecular docking, bioactivity prediction, and toxicity testing all involve impurity profiling. Before committing to DD, RS with minimal toxicity can be utilised in silico. The efficacy of molecular docking in getting a medication to market is still debated despite its refinement and improvement. Biomedical labs and pharmaceutical companies were hesitant to adopt molecular docking algorithms for drug screening despite their decades of development and improvement. Despite the widespread use of "force fields" to represent the energy exerted within and between molecules, it has been impossible to reliably predict or compute the binding affinities between proteins and potential binding medications.

Alzheimer’s Disease Prediction and Implications for Insurance

AbstractBackgroundThe possibilities to determine the individual risk of persons to develop future Alzheimer’s Disease and dementia have greatly increased in recent years. Promising and little invasive methods, e.g. blood or urine tests, may soon be available as predictive or early‐detection measures even for the general population. However, the knowledge incurred may be of interest not only for the affected persons themselves, but also for third parties. Particularly, the results of predictive tests may prompt disclosure or notification obligations to insurers, in turn affecting availability or affordable premiums for life or health insurance plans. This can be a substantial obstacle and burden for individuals considering to undergo predictive diagnosis. As predictive methods become more readily available, the problem gains pressing practical importance.MethodWe analyze existing legislation and jurisprudence in a variety of jurisdictions to investigate what regulations are in place for dealing with individual risk probabilities, and to what extent these are applicable or transferable to the particular problematic aspects regarding the prediction of Alzheimer’s Disease and/or dementia.ResultExisting insurance law, where in place, strikes only a rudimentary balance, and for a limited scope, mostly genetic diagnostics only. At the same time, access to disease prediction by non‐genetic means is quickly growing, and legal protections, rights and duties are largely unregulated in spite of a similar conflict of interests. The associated uncertainties for the involved parties keep increasing, and there is a pressing need for rules bringing the conflicting interests into a fair balance.ConclusionWe present efforts and their shortcomings to reduce the tension between the interest of insurers in offering affordable and calculable premiums on the basis of individual risk factors, and the particular need of affected persons to protect the results of predictive examinations due to their high sensitivity. We show how this tension could be sensibly resolved by taking into account ethical and societal values, considering the complex interests at stake.

Voice Analysis as a Method for Preoperatively Predicting a Difficult Airway Based on Machine Learning Algorithms: Original Research Report.

An unanticipated difficult airway is one of the greatest challenges for anesthesiologists. Proper preoperative airway assessment is crucial to reducing complications. However, current screening tests based on anthropometric features are of uncertain benefit. Therefore, our study explores using voice analysis with machine learning algorithms to predict a difficult airway. Observational, multicenter study with N = 438 patients initially enrolled at Centro Medico Teknon and Institut Universitari Dexeus (2019-2022) for the research study. After excluding 125 patients, N = 313 were included. Ethics committee approval was obtained. Adults ASA I-III scheduled for elective procedures under general anesthesia with endotracheal intubation were selected. Patient clinical features and traditional predictive tests were collected. Vowels "A, E, I, O, U" were recorded in normal, flexion, and extension positions. Cormack grade was assessed, and data were analyzed using KNIME, resulting in multiple models based on demographics and voice data. ROC curves and other metrics were evaluated for each model. Among multiple models evaluated, two yielded the best performance to predict a difficult airway both exclusively analyzing Cormack I and IV cases which showed the most distinct differences. The variables included in each model were the following: Model 1; included demographic data, vowel "A" in all positions and harmonics of the voice achieving an AUC of 0.91. Model 2; Included demographic data, vowel "O" in normal positions and voice parameters (Shimmer, Jitter, HNR); achieving in an AUC of 0.90. In contrast, models which focused on analyzing all Cormack grades (I, II, III, IV) cases performed less effectively. Acoustic parameters of the voice together with the demographic data of the patients, when introduced into classification algorithms based on machine learning showed promising signs of predicting a difficult airway.

Deep Learning for Weather Forecasting: A CNN-LSTM Hybrid Model for Predicting Historical Temperature Data

As global climate change intensifies, accurate weather forecasting has become increasingly important, affecting agriculture, energy management, environmental protection, and daily life. This study introduces a hybrid model combining Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to predict historical temperature data. CNNs are utilized for spatial feature extraction, while LSTMs handle temporal dependencies, resulting in significantly improved prediction accuracy and stability. By using Mean Absolute Error (MAE) as the loss function, the model demonstrates excellent performance in processing complex meteorological data, addressing challenges such as missing data and high-dimensionality. The results show a strong alignment between the prediction curve and test data, validating the model's potential in climate prediction. This study offers valuable insights for fields such as agriculture, energy management, and urban planning, and lays the groundwork for future applications in weather forecasting under the context of global climate change.

Kinetic Modeling Enables Understanding of Off-Cycle Processes in Pd-Catalyzed Amination of Five-Membered Heteroaryl Halides.

The mechanism of Pd-catalyzed amination of five-membered heteroaryl halides was investigated by integrating experimental kinetic analysis with kinetic modeling through predictive testing and likelihood ratio analysis, revealing an atypical productive coupling pathway and multiple off-cycle events. The GPhos-supported Pd catalyst, along with the moderate-strength base NaOTMS, was previously found to promote efficient coupling between five-membered heteroaryl halides and secondary amines. However, slight deviations from the optimal concentration, temperature, and/or solvent resulted in significantly lower yields, contrary to typical reaction optimization trends. We found that the coupling of 4-bromothiazole with piperidine proceeds through an uncommon mechanism in which the NaOTMS base, rather than the amine, binds first to the oxidative addition complex; the resulting OTMS-bound Pd species is the resting state. Formation of the Pd-amido complex via base/amine exchange was identified as the turnover-limiting step, unlike other reported catalyst systems for which reductive elimination is turnover-limiting. We determined that the amine-bound Pd complex, usually an on-cycle intermediate, is instead a reversibly generated off-cycle species, and that base-mediated decomposition of 4-bromothiazole is the primary irreversible catalyst deactivation pathway. Predictive testing and kinetic modeling were key to the identification of these off-cycle processes, providing insight into minor mechanistic pathways that are difficult to observe experimentally. Collectively, this report reveals the unique enabling features of the Pd-GPhos/NaOTMS system, implementing mechanistic insights to improve the yields of particularly challenging coupling reactions. Moreover, these findings highlight the utility of applying predictive tests to kinetic models for the rapid evaluation of mechanistic possibilities in small-molecule catalytic systems.

Testing and Treatment Thresholds for Pediatric Pneumonia in the Emergency Department.

Clinical prediction models for pediatric community-acquired pneumonia (CAP) may standardize management. Understanding physician risk thresholds is important for model implementation. We aimed to elucidate physician-derived thresholds for chest radiograph performance and empirical antibiotic treatment of CAP among children presenting to the emergency department with respiratory illness before and after knowledge of results of a validated clinical prediction model. Pediatric emergency physicians were surveyed through the American Academy of Pediatrics Pediatric Emergency Medicine Collaborative Research Committee and provided 8 clinical vignettes for children with respiratory symptoms. Respondents were asked to indicate their probability of radiographic CAP and choose whether they would obtain a chest radiograph or give empirical antibiotics before and after being provided with the probability of radiographic CAP based on a validated prediction model. We used logistic regression to establish testing and treatment thresholds, defined as the disease probability at which half of physicians acted. Two-hundred and eight (44.3%) of 469 physicians completed the survey. Most were attending physicians (96.0%) practicing in a freestanding children's hospital (76.8%). Testing and treatment thresholds for CAP were 17.6% (95% confidence interval [CI] 16.4% to 18.8%) and 66.1% (95% CI 60.1% to 72.5%), respectively, before knowledge of the model-estimated probability. With knowledge of the prediction model, testing and treatment thresholds were 13.5% (95% CI 12.3% to 14.7%) and 58.0% (95% CI 53.2-62.8). We elucidated physician thresholds for testing and treatment of CAP, which may be integrated into future pneumonia risk models to improve acceptability and incorporation into practice.

Enhancing flood forecasting and warning precision through multi-task deep learning approaches

ABSTRACT This study proposes a multi-task deep learning model for simultaneous prediction of time-series water levels and flood risk thresholds, aiming to enhance flood forecasting precision. Using AutoKeras, single-task and multi-task models were optimised to predict water levels 10–360 min ahead based on 720 min of prior data. The multi-task model consistently outperformed the single-task model across multiple evaluation metrics, including correlation coefficients, RMSE, Nash–Sutcliffe efficiency, and Kling–Gupta efficiency scores. Real-time prediction tests on actual rainfall events further validated the multi-task model's improved accuracy and applicability in operational flood forecasting. The study demonstrates significant progress in flood prediction methodologies, offering a more comprehensive approach to forecasting and categorising flood incidents.

Standing balance test for fall prediction in older adults: a 6-month longitudinal study

BackgroundA core component of older adult health care assessment includes identifying fall risk, which also includes identifying those with subtle balance deficits.ObjectiveTo compare body displacement of the Center of Pressure (CoP) and time held during the balance test. Also, to examine whether balance tests at baseline can predict falls after 6 months.MethodsA longitudinal study with 153 community-dwelling older adults, between 60–89 years old. Anteroposterior (AP) and mediolateral (ML) amplitude and velocity CoP displacements were assessed in four upright positions using a force platform: double-leg, semi-tandem, tandem, and single-leg stances, with a maximum duration of 30 s each. Adjusted repeated measures ANOVA were used to compare the differences among the balance positions. Comparisons between males and females were also conducted. Logistic regression adjusted for confounders was performed to verify whether upright balance tests can predict future falls.ResultsAs the base of support narrows, body sway increases. A decrease in stance time was observed across the balance stages, i.e., double-leg/semi-tandem versus tandem versus single-leg stances. The mean duration held in the single-leg stance was 14.8 s and for tandem was 22.2 s. Similar stance durations were observed for double-leg and semi-tandem stances. Males were able to maintain balance positions longer than females even with greater CoP displacement. ML amplitude of CoP displacement and the time held during tandem and single-leg positions were able to predict falls after 6 months (p < 0.05).ConclusionIn clinical practice in which only stance time is recorded, it is possible to interchangeably use the double-leg or semi-tandem stance. To identify early signs of imbalance, we suggest setting a time limit for the balance test equal to or greater than 23 s, as 10 s appear to be insufficient to detect subtle balance deficits. The time maintenance on tandem and single-leg positions was able to predict future falls.

A review of cell-free DNA and epigenetics for non-invasive diagnosis in solid organ transplantation.

Circulating cell-free DNA (cfDNA) is emerging as a non-invasive biomarker in solid organ transplantation (SOT) monitoring and data on its diagnostic potential have been increasing in recent years. This review aims to summarize the main advances in technologies, clinical applications and future perspectives of cfDNA for transplantation, and to approach the contribution of epigenetics to improve the specific detection of rejection. Published literature investigating cfDNA as a biomarker for the diagnosis of transplant rejection was systematically reviewed, specifically clinical trials evaluating the test performance of algorithms predicting rejection based on cfDNA fraction. Literature highlighting epigenetic features in transplant rejection was also reviewed to outline the potential contribution of the epigenomic analysis to the needs of rejection-specific diagnosis. 40 articles were reviewed, and results were extracted and summarized. 16 met the inclusion criteria by evaluating the diagnostic performance of a predictive test for the discrimination of rejection vs. non-rejection patients (2 heart, 3 liver, 4 kidney, and 7 lung transplantations). The recurring conclusion is the kinetics of dd-cfDNA levels, strongly increasing immediately after transplantation and reaching basal levels after days to weeks and remaining stable in non-rejection patients. On the other hand, rejection is characterized by an increase in dd-cfDNA levels, depending on the transplanted organs. In addition, the epigenetic signature can help improve the specificity of the diagnosis of rejection by searching for specific epigenetic features that are by the clinical status of patients. Cell-free DNA is a promising non-invasive biomarker but still needs standardization of technologies and protocols to be used for diagnostic purposes. Moreover, the lack of specificity of this marker can be compensated by the contribution of epigenetic analysis for which data are growing, although progress is still needed for its use in a clinical context.

Early mNGS testing for diagnose and prognostic prediction of early onset pneumonia among in-hospital cardiac arrest patients undergoing extracorporeal cardiopulmonary resuscitation.

Metagenomic next-generation sequencing (mNGS) is emerging as a novel diagnostic technology for various infectious diseases; however, limited studies have investigated its application in etiological diagnosis of early onset pneumonia (EOP) among patients undergoing extracorporeal cardiopulmonary resuscitation (ECPR) following in-hospital cardiac arrest (IHCA), The clinical significance of early mNGS in predicting short-term prognosis of IHCA patients after ECPR remains unclear. This retrospective study included 76 patients with IHCA who underwent ECPR at the First Affiliated Hospital of Zhengzhou University from January 2018 to December 2022. Baseline characteristics and etiological data of all patients during their hospitalization were collected and statistically analyzed. The primary outcome of this study was the diagnosis of EOP, while the secondary outcomes included successful extracorporeal membrane oxygenation (ECMO) weaning and survival at discharge. Additionally, the characteristics of bronchoalveolar lavage fluid (BALF) flora in these patients were compared by analyzing both mNGS results and culture results. Multivariate logistic regression were employed to analyze the predictors of ECMO weaning failure, mortality at discharge, and the incidence of EOP. Ultimately, patients with lower SOFA scores on admission [OR (95%CI): 1.447 (1.107-1.890), p=0.007] and those who underwent early mNGS testing within 48 hours after ECPR [OR (95%CI): 0.273 (0.086-0.865), p=0.027] demonstrated a higher probability of successful weaning from ECMO. Patients with higher SOFA scores on admission [OR (95%CI): 2.404 (1.422-4.064), p=0.001], and elevated lactate levels [OR (95%CI): 1.176 (1.017-1.361), p=0.029] exhibited an increased likelihood of mortality at discharge. Furthermore, early mNGS detection [OR (95%CI): 0.186 (0.035-0.979), p=0.047], and lower CRP levels (48h-7d after ECMO) [OR (95%CI):1.011 (1.003-1.019), p=0.006] were associated with a reduced incidence of EOP. In addition, the pathogens detected by mNGS within 48 hours after ECPR were mainly oral colonizing bacteria and viruses, and viruses were in the majority, while all BALF cultures were negative. In contrast, between 48 hours and 7 days after ECPR, BALF cultures were positive in all EOP patients. Early mNGS testing to identify microbial flora facilitates timely adjustment of antibiotic regimens, thereby reducing the incidence of EOP and improving short-term prognosis in patients undergoing ECPR following IHCA.