scholarly journals Prediction of short-term mortality in acute heart failure patients using minimal electronic health record data

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Ashwath Radhachandran ◽  
Anurag Garikipati ◽  
Nicole S. Zelin ◽  
Emily Pellegrini ◽  
Sina Ghandian ◽  
...  
Keyword(s):  
Heart Failure ◽  
Logistic Regression ◽  
Decision Tree ◽  
Risk Stratification ◽  
Acute Heart Failure ◽  
Model Performance ◽  
Machine Learning Techniques ◽  
Electronic Health Record Data ◽  
Test Set ◽  
Patient Disposition

Abstract Background Acute heart failure (AHF) is associated with significant morbidity and mortality. Effective patient risk stratification is essential to guiding hospitalization decisions and the clinical management of AHF. Clinical decision support systems can be used to improve predictions of mortality made in emergency care settings for the purpose of AHF risk stratification. In this study, several models for the prediction of seven-day mortality among AHF patients were developed by applying machine learning techniques to retrospective patient data from 236,275 total emergency department (ED) encounters, 1881 of which were considered positive for AHF and were used for model training and testing. The models used varying subsets of age, sex, vital signs, and laboratory values. Model performance was compared to the Emergency Heart Failure Mortality Risk Grade (EHMRG) model, a commonly used system for prediction of seven-day mortality in the ED with similar (or, in some cases, more extensive) inputs. Model performance was assessed in terms of area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity. Results When trained and tested on a large academic dataset, the best-performing model and EHMRG demonstrated test set AUROCs of 0.84 and 0.78, respectively, for prediction of seven-day mortality. Given only measurements of respiratory rate, temperature, mean arterial pressure, and FiO2, one model produced a test set AUROC of 0.83. Neither a logistic regression comparator nor a simple decision tree outperformed EHMRG. Conclusions A model using only the measurements of four clinical variables outperforms EHMRG in the prediction of seven-day mortality in AHF. With these inputs, the model could not be replaced by logistic regression or reduced to a simple decision tree without significant performance loss. In ED settings, this minimal-input risk stratification tool may assist clinicians in making critical decisions about patient disposition by providing early and accurate insights into individual patient’s risk profiles.

scholarly journals GRACE score vs GWTG-HF score: risk stratification in acute heart failure

2021 ◽  
Vol 10 (Supplement_1) ◽  
Author(s):  
M Santos ◽  
S Paula ◽  
I Almeida ◽  
H Santos ◽  
H Miranda ◽  
...  
Keyword(s):  
Heart Failure ◽  
Logistic Regression ◽  
Risk Stratification ◽  
Acute Heart Failure ◽  
Roc Curve ◽  
Curve Analysis ◽  
Invasive Ventilation ◽  
Roc Curve Analysis ◽  
Post Discharge ◽  
Grace Score

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Patients (P) with acute heart failure (AHF) are a heterogeneous population. Therefore, early risk stratification at admission is essential. The Get With The Guidelines Heart Failure score (GWTG-HF) predicts in-hospital mortality (IHM) of patients admitted with AHF. GRACE score estimates risk of death, including IHM and long-term mortality (M), in non-ST elevation acute coronary syndromes. Objective To validate GRACE score in AHF and to compare GRACE and GWTG-HF scores as predictors of IHM, post discharge early and late M [1-month mortality (1mM) and 1-year M (1yM)], 1-month readmission (1mRA) and 1-year readmission (1yRA), in our center population, using real-life data. Methods Based on a single-center retrospective study, data collected from P admitted in the Cardiology department with AHF between 2010 and 2017. P without data on previous cardiovascular history or uncompleted clinical data were excluded. Statistical analysis used chi-square, non-parametric tests, logistic regression analysis and ROC curve analysis. Results 35.3% were admitted in Killip-Kimball class (KKC) 4. Mean GRACE was 147.9 ± 30.2 and mean GWTG-HF was 41.7 ± 9.6. Inotropes’ usage was necessary in 32.7% of the P, 11.3% of the patients needed non-invasive ventilation, 8% needed invasive ventilation. IHM rate was 5%, 1mM was 8% and 1yM 27%. 6.3% of the patients were readmitted 1 month after discharge and 52.7% had at least one more admission in the year following discharge. Older age (p < 0.001), lower SBP (p = 0,005), higher urea (p = 0,001), lower sodium (p = 0.005), previous history of percutaneous coronary intervention (p = 0,017), lower GFR (p < 0.001) and need of inotropes (0.001) were predictors of 1yM after discharge in our population. As expected, patients presenting in KKC 4 had higher IHM (OR 8.13, p < 0.001), higher 1mM (OR 4.13, p = 0.001) and higher 1yM (OR 1.96, p = 0.011). On the other hand, KKC at admission did not predict readmission (either 1mRA or 1yRA, respectively p = 0.887 and p = 0.695). Logistic regression confirmed that GWTG-HF was a good predictor of IHM (OR 1.12, p < 0.001, CI 1.05-1.19) but also 1mM (OR 1.1, p = 0.001, CI 1.04-1.16) and 1yM (OR 1.08, p < 0.001, CI 1.04-1.11). GRACE also showed the ability to predict IHM (OR 1.06, p < 0.001, CI 1.03-1.10), 1mM (OR 1.04, p < 0.001, CI 1.02-1.06) and 1yM (OR 1.03, p < 0.001, CI 1.01-1.03). ROC curve analysis revealed that GRACE and GWTG-HF were accurate at predicting IHM (AUC 0.866 and 0.774, respectively), 1mM (AUC 0.779 and 0.727, respectively) and 1yM (AUC 0.676 and 0.672, respectively). Both scores failed at predicting 1mRA (GRACE p = 0.463; GWTG-HF p = 0.841) and 1yRA (GRACE p = 0.244; GWTG-HF p = 0.806). Conclusion This study confirms that, in our population, both scores were excellent at predicting IHM, with GRACE performing better. Although both scores were able to predict post-discharge mortality outcomes, their performance was poorer.

scholarly journals Time-to-event prediction analysis of patients with chronic heart failure comorbid with atrial fibrillation: a LightGBM model

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Chu Zheng ◽  
Jing Tian ◽  
Ke Wang ◽  
Linai Han ◽  
Hong Yang ◽  
...  
Keyword(s):  
Heart Failure ◽  
Atrial Fibrillation ◽  
Logistic Regression ◽  
Chronic Heart Failure ◽  
Risk Stratification ◽  
Risk Group ◽  
Low Risk ◽  
Training Set ◽  
Test Set ◽  
All Cause Mortality

Abstract Background Chronic heart failure (CHF) comorbid with atrial fibrillation (AF) is a serious threat to human health and has become a major clinical burden. This prospective cohort study was performed to design a risk stratification system based on the light gradient boosting machine (LightGBM) model to accurately predict the 1- to 3-year all-cause mortality of patients with CHF comorbid with AF. Methods Electronic medical records of hospitalized patients with CHF comorbid with AF from January 2014 to April 2019 were collected. The data set was randomly divided into a training set and test set at a 3:1 ratio. In the training set, the synthetic minority over-sampling technique (SMOTE) algorithm and fivefold cross validation were used for LightGBM model training, and the model performance was performed on the test set and compared using the logistic regression method. The survival rate was presented on a Kaplan–Meier curve and compared by a log-rank test, and the hazard ratio was calculated by a Cox proportional hazard model. Results Of the included 1796 patients, the 1-, 2-, and 3-year cumulative mortality rates were 7.74%, 10.63%, and 12.43%, respectively. Compared with the logistic regression model, the LightGBM model showed better predictive performance, the area under the receiver operating characteristic curve for 1-, 2-, and 3-year all-cause mortality was 0.718 (95%CI, 0.710–0.727), 0.744(95%CI, 0.737–0.751), and 0.757 (95%CI, 0.751–0.763), respectively. The net reclassification index was 0.062 (95%CI, 0.044–0.079), 0.154 (95%CI, 0.138–0.172), and 0.148 (95%CI, 0.133–0.164), respectively. The differences between the two models were statistically significant (P < 0.05). Patients in the high-risk group had a significantly higher hazard of death than those in the low-risk group (hazard ratios: 12.68, 13.13, 14.82, P < 0.05). Conclusion Risk stratification based on the LightGBM model showed better discriminative ability than traditional model in predicting 1- to 3-year all-cause mortality of patients with CHF comorbid with AF. Individual patients’ prognosis could also be obtained, and the subgroup of patients with a higher risk of mortality could be identified. It can help clinicians identify and manage high- and low-risk patients and carry out more targeted intervention measures to realize precision medicine and the optimal allocation of health care resources.

Acute heart failure: predicting early in-hospital outcomes

2021 ◽  
Vol 28 (Supplement_1) ◽  
Author(s):  
M Santos ◽  
S Paula ◽  
I Almeida ◽  
H Santos ◽  
H Miranda ◽  
...  
Keyword(s):  
Heart Failure ◽  
Logistic Regression ◽  
Acute Heart Failure ◽  
Roc Curve ◽  
Real Life ◽  
Study Data ◽  
Curve Analysis ◽  
Older Age ◽  
Invasive Ventilation ◽  
Roc Curve Analysis

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Patients (P) with acute heart failure (AHF) are a heterogeneous population. Risk stratification at admission may help predict in-hospital complications and needs. The Get With The Guidelines Heart Failure score (GWTG-HF) predicts in-hospital mortality (M) of P admitted with AHF. ACTION ICU score is validated to estimate the risk of complications requiring ICU care in non-ST elevation acute coronary syndromes. Objective To validate ACTION-ICU score in AHF and to compare ACTION-ICU to GWTG-HF as predictors of in-hospital M (IHM), early M [1-month mortality (1mM)] and 1-month readmission (1mRA), using real-life data. Methods Based on a single-center retrospective study, data collected from P admitted in the Cardiology department with AHF between 2010 and 2017. P without data on previous cardiovascular history or uncompleted clinical data were excluded. Statistical analysis used chi-square, non-parametric tests, logistic regression analysis and ROC curve analysis. Results Among the 300 P admitted with AHF included, mean age was 67.4 ± 12.6 years old and 72.7% were male. Systolic blood pressure (SBP) was 131.2 ± 37.0mmHg, glomerular filtration rate (GFR) was 57.1 ± 23.5ml/min. 35.3% were admitted in Killip-Kimball class (KKC) 4. ACTION-ICU score was 10.4 ± 2.3 and GWTG-HF was 41.7 ± 9.6. Inotropes’ usage was necessary in 32.7% of the P, 11.3% of the P needed non-invasive ventilation (NIV), 8% needed invasive ventilation (IV). IHM rate was 5% and 1mM was 8%. 6.3% of the P were readmitted 1 month after discharge. Older age (p &lt; 0.001), lower SBP (p = 0,035) and need of inotropes (p &lt; 0.001) were predictors of IHM in our population. As expected, patients presenting in KKC 4 had higher IHM (OR 8.13, p &lt; 0.001). Older age (OR 1.06, p = 0.002, CI 1.02-1.10), lower SBP (OR 1.01, p = 0.05, CI 1.00-1.02) and lower left ventricle ejection fraction (LVEF) (OR 1.06, p &lt; 0.001, CI 1.03-1.09) were predictors of need of NIV. None of the variables were predictive of IV. LVEF (OR 0.924, p &lt; 0.001, CI 0.899-0.949), lower SBP (OR 0.80, p &lt; 0.001, CI 0.971-0.988), higher urea (OR 1.01, p &lt; 0.001, CI 1.005-1.018) and lower sodium (OR 0.92, p = 0.002, CI 0.873-0.971) were predictors of inotropes’ usage. Logistic regression showed that GWTG-HF predicted IHM (OR 1.12, p &lt; 0.001, CI 1.05-1.19), 1mM (OR 1.10, p = 1.10, CI 1.04-1.16) and inotropes’s usage (OR 1.06, p &lt; 0.001, CI 1.03-1.10), however it was not predictive of 1mRA, need of IV or NIV. Similarly, ACTION-ICU predicted IHM (OR 1.51, p = 0.02, CI 1.158-1.977), 1mM (OR 1.45, p = 0.002, CI 1.15-1.81) and inotropes’ usage (OR 1.22, p = 0.002, CI 1.08-1.39), but not 1mRA, the need of IV or NIV. ROC curve analysis revealed that GWTG-HF score performed better than ACTION-ICU regarding IHM (AUC 0.774, CI 0.46-0-90 vs AUC 0.731, CI 0.59-0.88) and 1mM (AUC 0.727, CI 0.60-0.85 vs AUC 0.707, CI 0.58-0.84). Conclusion In our population, both scores were able to predict IHM, 1mM and inotropes’s usage.

Predicting 30-day Hospital Readmission with Publicly Available Administrative Database

2015 ◽  
Vol 54 (06) ◽  
pp. 560-567 ◽  
Author(s):  
K. Zhu ◽  
Z. Lou ◽  
J. Zhou ◽  
N. Ballester ◽  
P. Parikh ◽  
...  
Keyword(s):  
Heart Failure ◽  
Logistic Regression ◽  
Decision Tree ◽  
Ad Hoc ◽  
Prediction Models ◽  
Conditional Logistic Regression ◽  
Hospital Readmissions ◽  
Decision Rules ◽  
Classification Models ◽  
Standard Classification

SummaryIntroduction: This article is part of the Focus Theme of Methods of Information in Medicine on “Big Data and Analytics in Healthcare”.Background: Hospital readmissions raise healthcare costs and cause significant distress to providers and patients. It is, therefore, of great interest to healthcare organizations to predict what patients are at risk to be readmitted to their hospitals. However, current logistic regression based risk prediction models have limited prediction power when applied to hospital administrative data. Meanwhile, although decision trees and random forests have been applied, they tend to be too complex to understand among the hospital practitioners.Objectives: Explore the use of conditional logistic regression to increase the prediction accuracy.Methods: We analyzed an HCUP statewide in-patient discharge record dataset, which includes patient demographics, clinical and care utilization data from California. We extracted records of heart failure Medicare beneficiaries who had inpatient experience during an 11-month period. We corrected the data imbalance issue with under-sampling. In our study, we first applied standard logistic regression and decision tree to obtain influential variables and derive practically meaning decision rules. We then stratified the original data set accordingly and applied logistic regression on each data stratum. We further explored the effect of interacting variables in the logistic regression modeling. We conducted cross validation to assess the overall prediction performance of conditional logistic regression (CLR) and compared it with standard classification models.Results: The developed CLR models outperformed several standard classification models (e.g., straightforward logistic regression, stepwise logistic regression, random forest, support vector machine). For example, the best CLR model improved the classification accuracy by nearly 20% over the straightforward logistic regression model. Furthermore, the developed CLR models tend to achieve better sensitivity of more than 10% over the standard classification models, which can be translated to correct labeling of additional 400 – 500 readmissions for heart failure patients in the state of California over a year. Lastly, several key predictor identified from the HCUP data include the disposition location from discharge, the number of chronic conditions, and the number of acute procedures.Conclusions: It would be beneficial to apply simple decision rules obtained from the decision tree in an ad-hoc manner to guide the cohort stratification. It could be potentially beneficial to explore the effect of pairwise interactions between influential predictors when building the logistic regression models for different data strata. Judicious use of the ad-hoc CLR models developed offers insights into future development of prediction models for hospital readmissions, which can lead to better intuition in identifying high-risk patients and developing effective post-discharge care strategies. Lastly, this paper is expected to raise the awareness of collecting data on additional markers and developing necessary database infrastructure for larger-scale exploratory studies on readmission risk prediction.

scholarly journals Predicting Metabolic Syndrome With Machine Learning Models Using a Decision Tree Algorithm: Retrospective Cohort Study (Preprint)

2019 ◽  
Author(s):  
Cheng-Sheng Yu ◽  
Yu-Jiun Lin ◽  
Chang-Hsien Lin ◽  
Sen-Te Wang ◽  
Shiyng-Yu Lin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Metabolic Syndrome ◽  
Logistic Regression ◽  
Decision Tree ◽  
Characteristic Curve ◽  
Significant Risk ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Health Examination ◽  
Multivariate Logistic Regression

BACKGROUND Metabolic syndrome is a cluster of disorders that significantly influence the development and deterioration of numerous diseases. FibroScan is an ultrasound device that was recently shown to predict metabolic syndrome with moderate accuracy. However, previous research regarding prediction of metabolic syndrome in subjects examined with FibroScan has been mainly based on conventional statistical models. Alternatively, machine learning, whereby a computer algorithm learns from prior experience, has better predictive performance over conventional statistical modeling. OBJECTIVE We aimed to evaluate the accuracy of different decision tree machine learning algorithms to predict the state of metabolic syndrome in self-paid health examination subjects who were examined with FibroScan. METHODS Multivariate logistic regression was conducted for every known risk factor of metabolic syndrome. Principal components analysis was used to visualize the distribution of metabolic syndrome patients. We further applied various statistical machine learning techniques to visualize and investigate the pattern and relationship between metabolic syndrome and several risk variables. RESULTS Obesity, serum glutamic-oxalocetic transaminase, serum glutamic pyruvic transaminase, controlled attenuation parameter score, and glycated hemoglobin emerged as significant risk factors in multivariate logistic regression. The area under the receiver operating characteristic curve values for classification and regression trees and for the random forest were 0.831 and 0.904, respectively. CONCLUSIONS Machine learning technology facilitates the identification of metabolic syndrome in self-paid health examination subjects with high accuracy.

scholarly journals ACUTE Heart Failure Risk Stratification

Circulation ◽  
2019 ◽  
Vol 139 (9) ◽  
pp. 1157-1161 ◽  
Author(s):  
Sean P. Collins ◽  
Peter S. Pang
Keyword(s):  
Heart Failure ◽  
Risk Stratification ◽  
Acute Heart Failure ◽  
Failure Risk

scholarly journals Estimating real-world performance of a predictive model: a case-study in predicting mortality

JAMIA Open ◽  
2020 ◽  
Vol 3 (2) ◽  
pp. 243-251
Author(s):  
Vincent J Major ◽  
Neil Jethani ◽  
Yindalon Aphinyanaphongs
Keyword(s):  
Experimental Design ◽  
Real World ◽  
Model Performance ◽  
Assistive Technologies ◽  
Training Data ◽  
Electronic Health Record Data ◽  
Test Set ◽  
Model Composite ◽  
Temporal Validation ◽  
Cohort Selection

Abstract Objective One primary consideration when developing predictive models is downstream effects on future model performance. We conduct experiments to quantify the effects of experimental design choices, namely cohort selection and internal validation methods, on (estimated) real-world model performance. Materials and Methods Four years of hospitalizations are used to develop a 1-year mortality prediction model (composite of death or initiation of hospice care). Two common methods to select appropriate patient visits from their encounter history (backwards-from-outcome and forwards-from-admission) are combined with 2 testing cohorts (random and temporal validation). Two models are trained under otherwise identical conditions, and their performances compared. Operating thresholds are selected in each test set and applied to a “real-world” cohort of labeled admissions from another, unused year. Results Backwards-from-outcome cohort selection retains 25% of candidate admissions (n = 23 579), whereas forwards-from-admission selection includes many more (n = 92 148). Both selection methods produce similar performances when applied to a random test set. However, when applied to the temporally defined “real-world” set, forwards-from-admission yields higher areas under the ROC and precision recall curves (88.3% and 56.5% vs. 83.2% and 41.6%). Discussion A backwards-from-outcome experiment manipulates raw training data, simplifying the experiment. This manipulated data no longer resembles real-world data, resulting in optimistic estimates of test set performance, especially at high precision. In contrast, a forwards-from-admission experiment with a temporally separated test set consistently and conservatively estimates real-world performance. Conclusion Experimental design choices impose bias upon selected cohorts. A forwards-from-admission experiment, validated temporally, can conservatively estimate real-world performance. LAY SUMMARY The routine care of patients stands to benefit greatly from assistive technologies, including data-driven risk assessment. Already, many different machine learning and artificial intelligence applications are being developed from complex electronic health record data. To overcome challenges that arise from such data, researchers often start with simple experimental approaches to test their work. One key component is how patients (and their healthcare visits) are selected for the study from the pool of all patients seen. Another is how the group of patients used to create the risk estimator differs from the group used to evaluate how well it works. These choices complicate how the experimental setting compares to the real-world application to patients. For example, different selection approaches that depend on each patient’s future outcome can simplify the experiment but are impractical upon implementation as these data are unavailable. We show that this kind of “backwards” experiment optimistically estimates how well the model performs. Instead, our results advocate for experiments that select patients in a “forwards” manner and “temporal” validation that approximates training on past data and implementing on future data. More robust results help gauge the clinical utility of recent works and aid decision-making before implementation into practice.

Machine Learning Techniques Applied to Profile Mobile Banking Users in India

2013 ◽  
Vol 5 (1) ◽  
pp. 82-92 ◽  
Author(s):  
M. Carr ◽  
V. Ravi ◽  
G. Sridharan Reddy ◽  
D. Veranna
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Decision Tree ◽  
Decision Trees ◽  
Multilayer Perceptron ◽  
Machine Learning Techniques ◽  
Mobile Banking ◽  
Classification Rules ◽  
Learning Techniques ◽  
Potential Customers

This paper profiles mobile banking users using machine learning techniques viz. Decision Tree, Logistic Regression, Multilayer Perceptron, and SVM to test a research model with fourteen independent variables and a dependent variable (adoption). A survey was conducted and the results were analysed using these techniques. Using Decision Trees the profile of the mobile banking adopter’s profile was identified. Comparing different machine learning techniques it was found that Decision Trees outperformed the Logistic Regression and Multilayer Perceptron and SVM. Out of all the techniques, Decision Tree is recommended for profiling studies because apart from obtaining high accurate results, it also yields ‘if–then’ classification rules. The classification rules provided here can be used to target potential customers to adopt mobile banking by offering them appropriate incentives.

Sign in / Sign up

Export Citation Format

Share Document