Machine Learning-Based Systems for the Anticipation of Adverse Events After Pediatric Cardiac Surgery
Ontology highlight
ABSTRACT:
SUBMITTER: Garcia-Canadilla P
PROVIDER: S-EPMC9271800 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Major adverse cardiac events (MACE) triggered by non-cardiac surgery are prognostically important perioperative complications. However, due to often asymptomatic presentation, the incidence and timing of postoperative MACE are incompletely understood. We conducted a prospective observational study implementing a perioperative screening for postoperative MACE [cardiovascular death (CVD), acute heart failure (AHF), haemodynamically relevant arrhythmias, spontaneous myocardial infarction (MI), and perioperative myocardial infarction/injury (PMI)] in patients at increased cardiovascular risk (≥65 years OR ≥45 years with history of cardiovascular disease) undergoing non-cardiac surgery at a tertiary hospital. All patients received serial measurements of cardiac troponin to detect asymptomatic MACE. Among 2265 patients (mean age 73 years, 43.4% women), the incidence of MACE was 15.2% within 30 days, and 20.6% within 365 days. CVD occurred in 1.2% [95% confidence interval (CI) 0.9-1.8] and in 3.7% (95% CI 3.0-4.5), haemodynamically relevant arrhythmias in 1.2% (95% CI 0.9-1.8) and in 2.1% (95% CI 1.6-2.8), AHF in 1.6% (95% CI 1.2-2.2) and in 4.2% (95% CI 3.4-5.1), spontaneous MI in 0.5% (95% CI 0.3-0.9) and in 1.6% (95% CI 1.2-2.2), and PMI in 13.2% (95% CI 11.9-14.7) and in 14.8% (95% CI 13.4-16.4) within 30 days and within 365 days, respectively. The MACE-incidence was increased above presumed baseline rate until Day 135 (95% CI 104-163), indicating a vulnerable period of 3-5 months. One out of five high-risk patients undergoing non-cardiac surgery will develop one or more MACE within 365 days. The risk for MACE remains increased for about 5 months after non-cardiac surgery. https://www.clinicaltrials.gov. Unique identifier: NCT02573532.
| S-EPMC8245139 | biostudies-literature
Project description:BackgroundAlthough there is a growing interest in prediction models based on electronic medical records (EMRs) to identify patients at risk of adverse cardiac events following invasive coronary treatment, robust models fully utilizing EMR data are limited.ObjectiveWe aimed to develop and validate machine learning (ML) models by using diverse fields of EMR to predict the risk of 30-day adverse cardiac events after percutaneous intervention or bypass surgery.MethodsEMR data of 5,184,565 records of 16,793 patients at a quaternary hospital between 2006 and 2016 were categorized into static basic (eg, demographics), dynamic time-series (eg, laboratory values), and cardiac-specific data (eg, coronary angiography). The data were randomly split into training, tuning, and testing sets in a ratio of 3:1:1. Each model was evaluated with 5-fold cross-validation and with an external EMR-based cohort at a tertiary hospital. Logistic regression (LR), random forest (RF), gradient boosting machine (GBM), and feedforward neural network (FNN) algorithms were applied. The primary outcome was 30-day mortality following invasive treatment.ResultsGBM showed the best performance with area under the receiver operating characteristic curve (AUROC) of 0.99; RF had a similar AUROC of 0.98. AUROCs of FNN and LR were 0.96 and 0.93, respectively. GBM had the highest area under the precision-recall curve (AUPRC) of 0.80, and the AUPRCs of RF, LR, and FNN were 0.73, 0.68, and 0.63, respectively. All models showed low Brier scores of <0.1 as well as highly fitted calibration plots, indicating a good fit of the ML-based models. On external validation, the GBM model demonstrated maximal performance with an AUROC of 0.90, while FNN had an AUROC of 0.85. The AUROCs of LR and RF were slightly lower at 0.80 and 0.79, respectively. The AUPRCs of GBM, LR, and FNN were similar at 0.47, 0.43, and 0.41, respectively, while that of RF was lower at 0.33. Among the categories in the GBM model, time-series dynamic data demonstrated a high AUROC of >0.95, contributing majorly to the excellent results.ConclusionsExploiting the diverse fields of the EMR data set, the ML-based 30-day adverse cardiac event prediction models demonstrated outstanding results, and the applied framework could be generalized for various health care prediction models.
| S-EPMC9133980 | biostudies-literature
Project description:BackgroundMachine learning (ML) models are promising tools for predicting adverse postoperative outcomes in cardiac surgery, yet have not translated to routine clinical use. We conducted a systematic review and meta-analysis to assess the predictive performance of ML approaches.MethodsWe conducted an electronic search to find studies assessing ML and traditional statistical models to predict postoperative outcomes. Our primary outcome was the concordance (C-) index of discriminative performance. Using a Bayesian meta-analytic approach we pooled the C-indices with the 95% credible interval (CrI) across multiple outcomes comparing ML methods to logistic regression (LR) and clinical scoring tools. Additionally, we performed critical difference and sensitivity analysis.ResultsWe identified 2792 references from the search of which 51 met inclusion criteria. Two postoperative outcomes were amenable for meta-analysis: 30-day mortality and in-hospital mortality. For 30-day mortality, the pooled C-index and 95% CrI were 0.82 (0.79-0.85), 0.80 (0.77-0.84), 0.78 (0.74-0.82) for ML models, LR, and scoring tools respectively. For in-hospital mortality, the pooled C-index was 0.81 (0.78-0.84) and 0.79 (0.73-0.84) for ML models and LR, respectively. There were no statistically significant results indicating ML superiority over LR.ConclusionIn cardiac surgery patients, for the prediction of mortality, current ML methods do not have greater discriminative power over LR as measured by the C-index.
| S-EPMC9804388 | biostudies-literature
Project description:BackgroundPostoperative delirium is a common complication that is distressing. This study aimed to demonstrate a prediction model for delirium.MethodsAmong 203,374undergoing non-cardiac surgery between January 2011 and June 2019 at Samsung Medical Center, 2,865 (1.4%) were diagnosed with postoperative delirium. After comparing performances of machine learning algorithms, we chose variables for a prediction model based on an extreme gradient boosting algorithm. Using the top five variables, we generated a prediction model for delirium and conducted an external validation. The Kaplan-Meier and Cox survival analyses were used to analyse the difference of delirium occurrence in patients classified as a prediction model.ResultsThe top five variables selected for the postoperative delirium prediction model were age, operation duration, physical status classification, male sex, and surgical risk. An optimal probability threshold in this model was estimated to be 0.02. The area under the receiver operating characteristic (AUROC) curve was 0.870 with a 95% confidence interval of 0.855-0.885, and the sensitivity and specificity of the model were 0.76 and 0.84, respectively. In an external validation, the AUROC was 0.867 (0.845-0.877). In the survival analysis, delirium occurred more frequently in the group of patients predicted as delirium using an internal validation dataset (p < 0.001).ConclusionBased on machine learning techniques, we analyzed a prediction model of delirium in patients who underwent non-cardiac surgery. Screening for delirium based on the prediction model could improve postoperative care. The working model is provided online and is available for further verification among other populations.Trial registrationKCT 0006363.
| S-EPMC10161528 | biostudies-literature
Project description:BackgroundPerioperative adverse cardiac events (PACE), a composite of myocardial infarction, coronary revascularization, congestive heart failure, arrhythmic attack, acute pulmonary embolism, cardiac arrest, and stroke during 30-day postoperative period, is associated with long-term mortality, but with limited clinical evidence. We compared long-term mortality with PACE using data from nationwide multicenter electronic health records.MethodsData from 7 hospitals, converted to Observational Medical Outcomes Partnership Common Data Model, were used. We extracted records of 277,787 adult patients over 18 years old undergoing non-cardiac surgery for the first time at the hospital and had medical records for more than 180 days before surgery. We performed propensity score matching and then an aggregated meta‑analysis.ResultsAfter 1:4 propensity score matching, 7,970 patients with PACE and 28,807 patients without PACE were matched. The meta‑analysis showed that PACE was associated with higher one-year mortality risk (hazard ratio [HR]: 1.33, 95% CI [1.10, 1.60], P = 0.005) and higher three-year mortality (HR: 1.18, 95% CI [1.01, 1.38], P = 0.038). In subgroup analysis, the risk of one-year mortality by PACE became greater with higher-risk surgical procedures (HR: 1.20, 95% CI [1.04, 1.39], P = 0.020 for low-risk surgery; HR: 1.69, 95% CI [1.45, 1.96], P < 0.001 for intermediate-risk; and HR: 2.38, 95% CI [1.47, 3.86], P = 0.034 for high-risk).ConclusionsA nationwide multicenter study showed that PACE was significantly associated with increased one-year mortality. This association was stronger in high-risk surgery, older, male, and chronic kidney disease subgroups. Further studies to improve mortality associated with PACE are needed.
| S-EPMC10834726 | biostudies-literature
Project description:Extreme events and the related anomalous statistics are ubiquitously observed in many natural systems, and the development of efficient methods to understand and accurately predict such representative features remains a grand challenge. Here, we investigate the skill of deep learning strategies in the prediction of extreme events in complex turbulent dynamical systems. Deep neural networks have been successfully applied to many imaging processing problems involving big data, and have recently shown potential for the study of dynamical systems. We propose to use a densely connected mixed-scale network model to capture the extreme events appearing in a truncated Korteweg-de Vries (tKdV) statistical framework, which creates anomalous skewed distributions consistent with recent laboratory experiments for shallow water waves across an abrupt depth change, where a remarkable statistical phase transition is generated by varying the inverse temperature parameter in the corresponding Gibbs invariant measures. The neural network is trained using data without knowing the explicit model dynamics, and the training data are only drawn from the near-Gaussian regime of the tKdV model solutions without the occurrence of large extreme values. A relative entropy loss function, together with empirical partition functions, is proposed for measuring the accuracy of the network output where the dominant structures in the turbulent field are emphasized. The optimized network is shown to gain uniformly high skill in accurately predicting the solutions in a wide variety of statistical regimes, including highly skewed extreme events. The technique is promising to be further applied to other complicated high-dimensional systems.
| S-EPMC6955342 | biostudies-literature
Project description:Background Carotid endarterectomy (CEA) is a major vascular operation for stroke prevention that carries significant perioperative risks; however, outcome prediction tools remain limited. The authors developed machine learning algorithms to predict outcomes following CEA. Methods and Results The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent CEA between 2011 and 2021. Input features included 36 preoperative demographic/clinical variables. The primary outcome was 30-day major adverse cardiovascular events (composite of stroke, myocardial infarction, or death). The data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 machine learning models were trained using preoperative features. The primary metric for evaluating model performance was area under the receiver operating characteristic curve. Model robustness was evaluated with calibration plot and Brier score. Overall, 38 853 patients underwent CEA during the study period. Thirty-day major adverse cardiovascular events occurred in 1683 (4.3%) patients. The best performing prediction model was XGBoost, achieving an area under the receiver operating characteristic curve of 0.91 (95% CI, 0.90-0.92). In comparison, logistic regression had an area under the receiver operating characteristic curve of 0.62 (95% CI, 0.60-0.64), and existing tools in the literature demonstrate area under the receiver operating characteristic curve values ranging from 0.58 to 0.74. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.02. The strongest predictive feature in our algorithm was carotid symptom status. Conclusions The machine learning models accurately predicted 30-day outcomes following CEA using preoperative data and performed better than existing tools. They have potential for important utility in guiding risk-mitigation strategies to improve outcomes for patients being considered for CEA.
| S-EPMC10757546 | biostudies-literature
Project description:The differentiation of induced pluripotent stem cells (iPSCs) into diverse functional cell types provides a promising solution to support drug discovery, disease modeling, and regenerative medicine. However, functional cell differentiation is currently limited by the substantial line-to-line and batch-to-batch variability, which severely impede the progress of scientific research and the manufacturing of cell products. For instance, iPSC-to-cardiomyocyte (CM) differentiation is vulnerable with great sensitivity to inappropriate doses of CHIR99021 (CHIR) that are applied in the initial stage of mesoderm differentiation. Here, by harnessing live-cell bright-field imaging and machine learning (ML), we realize real-time cell recognition in the entire differentiation process, e.g., CMs, cardiac progenitor cells (CPCs), iPSC clones, and even misdifferentiated cells. This enables non-invasive prediction of differentiation efficiency, purification of ML-recognized CMs and CPCs for reducing cell contamination, early assessment of the CHIR dose for correcting the misdifferentiation trajectory, and evaluation of initial iPSC colonies for controlling the start point of differentiation, all of which provide a more invulnerable differentiation method with resistance to variability. Moreover, with the established ML models as a readout for the chemical screen, we identify a CDK8 inhibitor to further improve the cell resistance to the overdose of CHIR. Together, this study indicates that artificial intelligence is able to guide and iteratively optimize iPSC differentiation to achieve consistently high efficiency across cell lines and batches, providing a better understanding and rational modulation of the differentiation process for functional cell manufacturing in biomedical applications. With the assistance of RNA sequencing, we transcriptomically characterize the image-recognized CPCs (IR-CPCs), the cells treated with different CHIR doses at stage I, iPSC-CM treated with or without BI-1347, and cells treated with or without BI-1347 at stage I.
2023-04-01 | GSE226159 | GEO
Project description:The quality of treatment and prognosis after pediatric congenital heart surgery remains unsatisfactory. A reliable prediction model for postoperative complications of congenital heart surgery patients is essential to enable prompt initiation of therapy and improve the quality of prognosis. Here, we develop an interpretable machine-learning-based model that integrates patient demographics, surgery-specific features and intraoperative blood pressure data for accurately predicting complications after pediatric congenital heart surgery. We used blood pressure variability and the k-means algorithm combined with a smoothed formulation of dynamic time wrapping to extract features from time-series data. In addition, SHAP framework was used to provide explanations of the prediction. Our model achieved the best performance both in binary and multi-label classification compared with other consensus-based risk models. In addition, this explainable model explains why a prediction was made to help improve the clinical understanding of complication risk and generate actionable knowledge in practice. The combination of model performance and interpretability is easy for clinicians to trust and provide insight into how they should respond before the condition worsens after pediatric congenital heart surgery.
| S-EPMC8390484 | biostudies-literature
Project description:BackgroundAt present, the prediction of adverse events (AE) had practical significance in clinic and the accuracy of AE prediction model after left atrial appendage closure (LAAC) needed to be improved. To identify a good prediction model based on machine learning for short- and long-term AE after LAAC.MethodsIn this study, 869 patients were included from the Department of Cardiovascular Medicine of Shanghai University of Medicine and Health Sciences Affiliated Zhoupu Hospital during 2017 and 2021. Univariate and multivariate analyses were conducted for short-term AE after LAAC to determine possible risk factors related with AE. We compared 8 machine learning algorithms for prediction short-term AE, and XGBoost was found to have the best performance. In addition, Cox-regression was used for long-term AE to find out the risk factors and establish a prediction model.ResultsIn univariate and multivariate analysis, body mass index (BMI) [odds ratio (OR) =0.91], congestive heart failure, hypertension, age ≥75 years, diabetes, stroke2 attack (CHADS2) score (OR =0.49) and bleeding history or predisposition, labile international normalized ratio (INR), elderly, drug/alcohol usage (BLED) score (OR =1.71) were shown to be significant risk factors for short-term AE. The XGbosst algorithm was used to predict short-term AE based on 15 possible risk factors. For long-term AE, Cox regression was used for the prediction. The CHADS2 score [hazard ratio (HR) =1.43], hypertension (HR =2.18), age more than 75 (HR =0.49), diabetes (HR =0.57), BLED score (HR=0.28), stroke (HR =19.8), hepatopathy (HR =3.97), nephropathy (HR =2.93), INR instability (HR =4.18), drinking (HR =2.67), and drugs (HR =2.36) were significant risk factors for long-term AE. The XGBoost had a good receiver operating characteristic (ROC) curve and area under the curve (AUC) was 0.85. The accuracy of the XGBoost model stayed at nearly 0.95.ConclusionsIn short- and long-term AE, CHADS2 score and BLED score were the most obvious risk factors. Several other risk factors also played roles in AE of LAAC. The incidence of long-term AE is under 15% and LAAC is effective and safe. The XGBoost model had good prediction accuracy and ROC curve.
| S-EPMC9264065 | biostudies-literature