Skip to Main Content

Using Machine Learning Algorithms to Manage Diabetes and ASCVD Risk

February 03, 2022
by Elisabeth Reitman

Using machine learning algorithms, Yale researchers created a tool to determine when to prescribe canagliflozin to manage ASCVD in Type 2 diabetes patients.

Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of morbidity and mortality among diabetes patients. One method of care is a common class of Type 2 diabetes medications called sodium-glucose cotransporter-2 (SGLT2) inhibitors. In addition to lowering blood sugar, these drugs also provide cardiovascular health benefits. However, because of their high cost, they are underused.

Now, a novel online tool called INSIGHT© offers a machine learning-based approach when prescribing SGLT2 inhibitors to maximize the benefit for patients with Type 2 diabetes.

The study focusing on this support tool was led by Yale researcher Rohan Khera, MD, MS, and was published in February 2022 in the journal Diabetes Care. The study’s first author is Yale Clinical Fellow, Evangelos Oikonomou, MD, DPhil, who works with Khera at the CarDS Lab.

INSIGHT Identifies Diabetes Patients Most Likely to Benefit

Evidence from the CANagliflozin cardioVascular Assessment Study (CANVAS) and Canagliflozin on Renal and Cardiovascular Outcomes in Subjects with Type 2 Diabetes Mellitus and Diabetic Neuropathy (CREDENCE) trials suggests that SGLT2 inhibitors, such as canagliflozin, reduce the risk of hospitalizations and death from ASCVD in adults with Type 2 diabetes. However, canagliflozin prescriptions are expensive and widely underutilized.

The study describes an individualized approach that addresses a common clinical question for ASCVD risk management: Which patients with Type 2 diabetes and an elevated risk for cardiovascular disease are most likely to benefit from canagliflozin? INSIGHT’s machine learning algorithms use distinct patient phenotypes such as the duration of Type 2 diabetes, elevated blood pressure (hypertension), smoking habits, and cholesterol levels to identify such patients.

Machine Learning Algorithms Uncover Vital Clues

Rather than relying on observations or assumptions, machine learning enables researchers to analyze vast amounts of clinical data and discover patterns or clues that might otherwise have been overlooked.

In this instance, machine learning algorithms help isolate the characteristics that best determine the effects of canagliflozin for cardiovascular risk reduction. The study identified one-third of patients as being those who would benefit the most from this therapy, which may help more targeted implementation studies. The authors collected data from over 10,000 patients using an open science project known as the Yale University Open Data Access Project

Yale Researchers Are Recognized for Excellence

The findings were first presented at the 2021 Northwestern Cardiovascular Young Investigators’ Forum (NCYIF), where Oikonomou received an award for his presentation, “A Machine Learning Approach to Individualize the Cardiovascular Benefits of Canagliflozin Based on Participant-level Analyses of the CANVAS Trials.”

During the NCYIF, Khera received the Jeremiah Stamler Distinguished Young Investigator Research Award for Excellence in Clinical Science Cardiovascular Research, which is the top award in the junior faculty clinical research category and includes a prize of $10,000.

His abstract, “Race/Ethnicity and Sex Differences in Lifetime Healthcare Expenses Across Cardiovascular Risk Factors,” examines the role of socioeconomic factors on healthcare expenses across the lifetime of individuals and demonstrated that there are patterns suggestive of deferred care manifesting as excess spending among minorities in later life.

Established in 2005, the Jeremiah Stamler Distinguished Young Investigator Research Award recognizes progressive work in cardiovascular research and preventive medicine. The $10,000 award supports current or future cardiovascular research.

“Participating in the NCYIF provides an opportunity to interact with the most talented peers in basic, clinical, population, and translational science and obtain valuable feedback from pillars in the field,” said Khera. “My research interests are focused on defining precision strategies to improve cardiovascular care and build a better understanding of the challenges faced by our patients. This award will help further my investigation.”

Machine Learning-based Tool Chooses Between Imaging Tests

In April 2021, the authors Khera and Oikonomou published a paper in the European Heart Journal on another machine learning-based tool called ASSIST,© which helps physicians choose between two imaging tests for patients with chest pain.

The tool was created for symptomatic outpatients without diagnosed coronary artery disease (CAD), whose physicians believe that nonurgent, noninvasive cardiovascular testing is necessary for the evaluation of suspected CAD.

Originally published Feb. 4, 2022; updated May 11, 2022.

Submitted by Elisabeth Reitman on February 04, 2022