Weight-based multiple empirical kernel learning with neighbor discriminant constraint for heart failure mortality prediction

Web-Based Dynamic Nomogram for Predicting Risk of Mortality in Heart Failure with Mildly Reduced Ejection Fraction

Purpose

This study aimed to develop an integrative dynamic nomogram, including N-terminal pro-B type natural peptide (NT-proBNP) and estimated glomerular filtration rate (eGFR), for predicting the risk of all-cause mortality in HFmrEF patients.

Patients and Methods

790 HFmrEF patients were prospectively enrolled in the development cohort for the model. The least absolute shrinkage and selection operator (LASSO) regression and Random Survival Forest (RSF) were employed to select predictors for all-cause mortality. Develop a nomogram based on the Cox proportional hazard model for predicting long-term mortality (1-, 3-, and 5-year) in HFmrEF. Internal validation was conducted using Bootstrap, and the final model was validated in an external cohort of 338 consecutive adult patients. Discrimination and predictive performance were evaluated by calculating the time-dependent concordance index (C-index), area under the ROC curve (AUC), and calibration curve, with clinical value assessed via decision curve analysis (DCA). Integrated discrimination improvement (IDI) and net reclassification improvement (NRI) were used to assess the contributions of NT-proBNP and eGFR to the nomogram. Finally, develop a dynamic nomogram using the "Dynnom" package.

Results

The optimal independent predictors for all-cause mortality (APSELNH: A: angiotensin-converting enzyme inhibitors/angiotensin receptor blockers/angiotensin receptor-neprilysin inhibitor (ACEI/ARB/ARNI), P: percutaneous coronary intervention/coronary artery bypass graft (PCI/CABG), S: stroke, E: eGFR, L: lg of NT-proBNP, N: NYHA, H: healthcare) were incorporated into the dynamic nomogram. The C-index in the development cohort and validation cohort were 0.858 and 0.826, respectively, with AUCs exceeding 0.8, indicating good discrimination and predictive ability. DCA curves and calibration curves demonstrated clinical applicability and good consistency of the nomogram. NT-proBNP and eGFR provided significant net benefits to the nomogram.

Conclusion

In this study, the dynamic APSELNH nomogram developed serves as an accessible, functional, and effective clinical decision support calculator, offering accurate prognostic assessment for patients with HFmrEF.

Predicting the risk of 1-year mortality among patients hospitalized for acute heart failure in China

BACKGROUND

We aimed to develop and validate a model to predict 1-year mortality risk among patients hospitalized for acute heart failure (AHF), build a risk score and interpret its application in clinical decision making.

METHODS

By using data from China Patient-Centred Evaluative Assessment of Cardiac Events Prospective Heart Failure Study, which prospectively enrolled patients hospitalized for AHF in 52 hospitals across 20 provinces, we used multivariate Cox proportional hazard model to develop and validate a model to predict 1-year mortality.

RESULTS

There were 4,875 patients included in the study, 857 (17.58%) of them died within 1-year following discharge of index hospitalization. A total of 13 predictors were selected to establish the prediction model, including age, medical history of chronic obstructive pulmonary disease and hypertension, systolic blood pressure, Kansas City Cardiomyopathy Questionnaire-12 score, angiotensin converting enzyme inhibitor or angiotensin receptor blocker at discharge, discharge symptom, N-terminal pro-brain natriuretic peptide, high-sensitivity troponin T, serum creatine, albumin, blood urea nitrogen, and highly sensitive C-reactive protein. The model showed a high performance on discrimination (C-index was 0.759 [95% confidence interval: 0.739, 0.778] in development cohort and 0.761 [95% confidence interval: 0.731, 0.791] in validation cohort), accuracy, calibration, and outperformed than several existed risk scores. A point-based risk score was built to stratify low- (0-12), intermediate- (13-16), and high-risk group (≥17) among patients.

CONCLUSIONS

A prediction model using readily available predictors was developed and internal validated to predict 1-year mortality risk among patients hospitalized for AHF. It may serve as a useful tool for individual risk stratification and informing decision making to improve clinical care.

The Efficacy of Machine Learning Models for Predicting the Prognosis of Heart Failure: A Systematic Review and Meta-Analysis

INTRODUCTION

Heart failure (HF) is a major global public health concern. The application of machine learning (ML) to identify individuals at high risk and enable early intervention is a promising approach for improving HF prognosis. We aim to systematically evaluate the performance and value of ML models for predicting HF prognosis.

METHODS

PubMed, Web of Science, Scopus, and Embase online databases were searched up to April 30, 2023, to identify studies on the use of ML models to predict HF prognosis. HF prognosis primarily encompasses readmission and mortality. The meta-analysis was conducted by MedCalc software. Subgroup analyses include grouping based on types of ML models, time intervals, sample sizes, the number of predictive variables, validation methods, whether to conduct hyperparameter optimization and calibration, data set partitioning methods.

RESULTS

A total of 31 studies were included. The most common ML models were random forest, boosting, support vector machine, neural network. The area under the receiver operating characteristic curve (AUC) for predicting HF readmission was 0.675 (95% CI: 0.651-0.699, p < 0.001), and the AUC for predicting HF mortality was 0.790 (95% CI: 0.765-0.816, p < 0.001). Subgroup analyses revealed that models with the prediction time interval of 1 year, sample sizes ≥10,000, the number of predictive variables ≥100, external validation, hyperparameter tuning, calibration adjustment, and data set partitioning using 10-fold cross-validation exhibited favorable performance within their respective subgroups.

CONCLUSION

The performance of ML models in predicting HF readmission is relatively poor, while its performance in predicting HF mortality is moderate. The quality of the relevant studies is generally low, it is essential to enhance the predictive capabilities of ML models through targeted improvements in practical applications.

Improving Performance of Outcome Prediction for In-patients with Acute Myocardial Infarction Based on Embedding Representation Learned from Electronic Medical Records: Development and Validation Study (Preprint)

Background

The widespread secondary use of electronic medical records (EMRs) promotes health care quality improvement. Representation learning that can automatically extract hidden information from EMR data has gained increasing attention.

Objective

We aimed to propose a patient representation with more feature associations and task-specific feature importance to improve the outcome prediction performance for inpatients with acute myocardial infarction (AMI).

Methods

Medical concepts, including patients’ age, gender, disease diagnoses, laboratory tests, structured radiological features, procedures, and medications, were first embedded into real-value vectors using the improved skip-gram algorithm, where concepts in the context windows were selected by feature association strengths measured by association rule confidence. Then, each patient was represented as the sum of the feature embeddings weighted by the task-specific feature importance, which was applied to facilitate predictive model prediction from global and local perspectives. We finally applied the proposed patient representation into mortality risk prediction for 3010 and 1671 AMI inpatients from a public data set and a private data set, respectively, and compared it with several reference representation methods in terms of the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), and F1-score.

Results

Compared with the reference methods, the proposed embedding-based representation showed consistently superior predictive performance on the 2 data sets, achieving mean AUROCs of 0.878 and 0.973, AUPRCs of 0.220 and 0.505, and F1-scores of 0.376 and 0.674 for the public and private data sets, respectively, while the greatest AUROCs, AUPRCs, and F1-scores among the reference methods were 0.847 and 0.939, 0.196 and 0.283, and 0.344 and 0.361 for the public and private data sets, respectively. Feature importance integrated in patient representation reflected features that were also critical in prediction tasks and clinical practice.

Conclusions

The introduction of feature associations and feature importance facilitated an effective patient representation and contributed to prediction performance improvement and model interpretation.

An integrated review on machine learning approaches for heart disease prediction: Direction towards future research gaps

Objectives

To make a clear literature review on state-of-the-art heart disease prediction models.

Methods

It reviews 61 research papers and states the significant analysis. Initially, the analysis addresses the contributions of each literature works and observes the simulation environment. Here, different types of machine learning algorithms deployed in each contribution. In addition, the utilized dataset for existing heart disease prediction models was observed.

Results

The performance measures computed in entire papers like prediction accuracy, prediction error, specificity, sensitivity, f-measure, etc., are learned. Further, the best performance is also checked to confirm the effectiveness of entire contributions.

Conclusions

The comprehensive research challenges and the gap are portrayed based on the development of intelligent methods concerning the unresolved challenges in heart disease prediction using data mining techniques.

Ensemble classification technique for heart disease prediction with meta-heuristic-enabled training system

Objectives

This research work exclusively aims to develop a novel heart disease prediction framework including three major phases, namely proposed feature extraction, dimensionality reduction, and proposed ensemble-based classification.

Methods

As the novelty, the training of NN is carried out by a new enhanced optimization algorithm referred to as Sea Lion with Canberra Distance (S-CDF) via tuning the optimal weights. The improved S-CDF algorithm is the extended version of the existing “Sea Lion Optimization (SLnO)”. Initially, the statistical and higher-order statistical features are extracted including central tendency, degree of dispersion, and qualitative variation, respectively. However, in this scenario, the “curse of dimensionality” seems to be the greatest issue, such that there is a necessity of dimensionality reduction in the extracted features. Hence, the principal component analysis (PCA)-based feature reduction approach is deployed here. Finally, the dimensional concentrated features are fed as the input to the proposed ensemble technique with “Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (KNN)” with optimized Neural Network (NN) as the final classifier.

Results

An elaborative analyses as well as discussion have been provided by concerning the parameters, like evaluation metrics, year of publication, accuracy, implementation tool, and utilized datasets obtained by various techniques.

Conclusions

From the experiment outcomes, it is proved that the accuracy of the proposed work with the proposed feature set is 5, 42.85, and 10% superior to the performance with other feature sets like central tendency + dispersion feature, central tendency qualitative variation, and dispersion qualitative variation, respectively.

Results

Finally, the comparative evaluation shows that the presented work is appropriate for heart disease prediction as it has high accuracy than the traditional works.

Heart disease prediction using hybrid fuzzy K-medoids attribute weighting method with DBN-KELM based regression model

Automated prediction can be offered for further treatment to make effective and relieve the difficulties in the diagnosis of heart condition of patient. In this paper, a hybrid method is proposed combining FKMAW and DBNKELM based ensemble method to enhance medical diagnosis process. Firstly, the input attributes are weighed using a fuzzy k-medoids clustering based attribute weighting (FKMAW) method. Subsequently, the medical data classification performance is improved by applying the weighing method and the linearly separable dataset is obtained with the transformation of non-linearly separable dataset. With the weighted attributes, a regression model based heart disease prediction scheme is proposed combining Deep belief Network and Extreme learning machine (DBNKELM), in which Extreme learning machine is the top layer of the deep belief network to work as a regression model. The results demonstrate that FKMAW + DBNKELM achieved good performance in rectifying the problems in medical data classification for all the six datasets.