Cardiopredict ML: machine learning-enhanced cardiovascular risk assessment for preconception care

Active Grant Awarded

Award ID
R21HL184290
Recipient
Awarding agency
Obligated
$232,719

Award facts

Assistance listing
93.837 Cardiovascular Diseases Research
Completion date 9% complete
Place of performance
Hershey, PA 17033, USA

Award details come from USAspending.gov.

About this award

The Department of Health and Human Services' National Institutes of Health, through the National Heart, Lung, and Blood Institute, awarded Pennsylvania State University, doing business as Penn State Milton S. Hershey Medical Center, $232,719 on September 17, 2026, under Cardiovascular Diseases Research (CFDA 93.837), Grant R21HL184290. The program fosters heart and vascular research across basic, translational, clinical, and population sciences.

The award funds CardioPredict ML, which will develop, test, and implement a machine learning-based predictive model for assessing the risk of pregnancy-related cardiovascular complications in women before conception. Under Aim 1A, the recipient will search TriNetX Global Health Research Network electronic health record data for women evaluated before conception who were subsequently diagnosed with pregnancy, delivered, and had follow-up data up to one year after delivery. The recipient will extract variables representing known cardiovascular risk factors identified from the literature, then build, train, and evaluate multiple machine learning models using a comprehensive set of metrics for predictive accuracy and clinical utility. Aim 1B evaluates the model's performance in predicting the occurrence of cardiovascular complications.

The period of performance runs from September 17, 2026, to an ultimate completion date of August 31, 2028. Place of performance is Hershey, Pennsylvania 17033.

Description

Abstract
Maternal mortality provides critical insight into womens' health and the overall responsiveness of healthcare systems globally. Among developed nations, the united states stands alone in having a steadily rising rate of pregnancy-related mortality since 1987. One of the main causes of such mortalities is cardiovascular disease. As more women with increased cardiovascular risk become pregnant, IT is becoming increasingly necessary to identify them in the preconception phase. Therefore, healthcare providers require an accurate and accessible method for risk assessment. Recent evidence has shown that machine learning-based risk- assessment models can have a higher predictive value than traditional risk-assessment tools. As such, our goal is to develop, test, and implement an advanced machine learning-based predictive model for assessing the risk of pregnancy-related cardiovascular complications in women prior to conception, with the following specific aims: aim 1A: to develop and train a machine learning algorithm for predicting the risk of cardiovascular complications of pregnancy. We will search the trinetx global health research network for electronic health record (EHR) data of women who were evaluated prior to conception, then diagnosed with pregnancy, delivered, and had follow-up data up to one year after delivery. We have identified variables representing known risk factors contributing to cardiovascular outcomes based on a review of current literature, and will extract these variables from the patient data then prepare the extracted data for training the model. We will then implement a systemic approach to build, train, and evaluate multiple machine learning models, then use a comprehensive set of metrics to ensure predictive accuracy and clinical utility. Aim 1B: to evaluate the performance of the model in predicting the occurrence of cardiovascular complications of pregnancy in an independent cohort. This will involve validation utilizing the observational medical outcomes partnership (OMOP) common data model at penn state health. Aim 2: to determine the feasibility of conducting a prospective evaluation of the preconception CVD risk assessment tool in a clinical setting. We will focus on optimizing study design, estimating potential participant numbers across the penn state health system and assessing workflow integration. Once the predictive capability of the model is established, healthcare professionals can intervene with personalized preconception care and women will be able to work with their healthcare team to make more informed reproductive decisions.

Quick answers

Who received award R21HL184290?

R21HL184290 was awarded to Pennsylvania State University by National Heart Lung and Blood Institute (National Institutes of Health) on September 17, 2026.

How much is award R21HL184290 worth?

$232,719 has been obligated on R21HL184290.

On GovTribe

See this award's full record

  • Every transaction and subaward
  • The recipient's other awards and teaming partners
  • Similar grant awards
  • Ask GovTribe AI about this award