Dataset Information

Machine Learning Patient-Specific Prediction of Heart Failure Hospitalization Using Cardiac MRI-Based Phenotype and Electronic Health Information.

ABSTRACT:

Background

Heart failure (HF) hospitalization is a dominant contributor of morbidity and healthcare expenditures in patients with systolic HF. Cardiovascular magnetic resonance (CMR) imaging is increasingly employed for the evaluation of HF given capacity to provide highly reproducible phenotypic markers of disease. The combined value of CMR phenotypic markers and patient health information to deliver predictions of future HF events has not been explored. We sought to develop and validate a novel risk model for the patient-specific prediction of time to HF hospitalization using routinely reported CMR variables, patient-reported health status, and electronic health information.

Methods

Standardized data capture was performed for 1,775 consecutive patients with chronic systolic HF referred for CMR imaging. Patient demographics, symptoms, Health-related Quality of Life, pharmacy, and routinely reported CMR features were provided to both machine learning (ML) and competing risk Fine-Gray-based models (FGM) for the prediction of time to HF hospitalization.

Results

The mean age was 59 years with a mean LVEF of 36 ± 11%. The population was evenly distributed between ischemic (52%) and idiopathic non-ischemic cardiomyopathy (48%). Over a median follow-up of 2.79 years (IQR: 1.59-4.04) 333 patients (19%) experienced HF related hospitalization. Both ML and competing risk FGM based models achieved robust performance for the prediction of time to HF hospitalization. Respective 90-day, 1 and 2-year AUC values were 0.87, 0.83, and 0.80 for the ML model, and 0.89, 0.84, and 0.80 for the competing risk FGM-based model in a holdout validation cohort. Patients classified as high-risk by the ML model experienced a 34-fold higher occurrence of HF hospitalization at 90 days vs. the low-risk group.

Conclusion

In this study we demonstrated capacity for routinely reported CMR phenotypic markers and patient health information to be combined for the delivery of patient-specific predictions of time to HF hospitalization. This work supports an evolving migration toward multi-domain data collection for the delivery of personalized risk prediction at time of diagnostic imaging.

SUBMITTER: Cornhill AK

PROVIDER: S-EPMC9245012 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesThis study sought to describe the patterns of heart failure (HF)-exacerbating medications used among older adults hospitalized for HF and to examine determinants of HF-exacerbating medication use.BackgroundHF-exacerbating medications can potentially contribute to adverse outcomes and could represent an important target for future strategies to improve post-hospitalization outcomes.MethodsMedicare beneficiaries ≥65 years of age with an adjudicated HF hospitalization between 2003 and 2014 were derived from the geographically diverse REGARDS (Reasons for Geographic and Racial Difference in Stroke) cohort study. Major HF-exacerbating medications, defined as those listed on the 2016 American Heart Association Scientific Statement listing medications that can precipitate or induce HF, were examined. Patterns of prescribing medications at hospital admission and at discharge were examined, as well as changes that occurred between admission and discharge; and a multivariable logistic regression analysis was conducted to identify determinants of harmful prescribing practices following HF hospitalization (defined as either the continuation of an HF-exacerbating medications or an increase in the number of HF-exacerbating medications between hospital admission and discharge).ResultsAmong 558 unique individuals, 18% experienced a decrease in the number of HF-exacerbating medications between admission and discharge, 19% remained at the same number, and 12% experienced an increase. Multivariable logistic regression analysis revealed that diabetes (odds ratio [OR]: 1.80; 95% confidence interval [CI]: 1.18 to 2.75]) and small hospital size (OR: 1.93; 95% CI: 1.18 to 3.16) were the strongest, independently associated determinants of harmful prescribing practices.ConclusionsHF-exacerbating medication regimens are often continued or started following an HF hospitalization. These findings highlight an ongoing need to develop strategies to improve safe prescribing practices in this vulnerable population.