ClosedLoop.ai Uses AI and Machine Learning to Help Users Identify and Predict High-Risk Populations

2020

Industry Challenge

Healthcare organizations are navigating a digital “perfect storm.” They are faced with an industry-wide data tsunami, a need for digital transformation, and an unprecedented shift into business models that reward value over volume and outcomes over activity. These changes are motivating payers and providers to find ways to proactively manage the health of their populations, which begins with finding ways to identify and predict high-risk individuals.

ClosedLoop Utilizes AI and Machine Learning to Help Users Identify and Predict High-Risk Populations
kr_quotemark

The idea of risk stratification has been around for a long time. We’ve just brought it into the 21st century. What used to be the inputting of insurance claims data into prebuilt formulas is now the ability to use any person-linkable data and the latest AI and machine learning methodologies."

Carol McCall
Chief Health Analytics Officer, ClosedLoop.ai

ClosedLoop.ai’s Solution

ClosedLoop.ai brings explainable artificial intelligence (AI) to healthcare. Its data science platform creates and deploys models able to predict patients at risk across a variety of negative health outcomes, ranging from unplanned admissions and ER visits to serious fall-related injuries and near-term mortality. Models built using ClosedLoop’s data science platform are highly accurate and easily explainable. They optimize prediction and risk stratification and include detailed information that helps explain an individual’s risk. This helps payers and providers concentrate on and tailor their intervention resources to patients that have the greatest need.

ClosedLoop.ai’s data science platform, which is purpose built and dedicated to healthcare, is a work bench for data scientists to create better models faster and to manage and monitor them efficiently once deployed. The platform inputs an array of patient-linkable data to create models that produce detailed risk profiles, including a risk percentile, change in risk, contributing factors, and possible next actions.

“The idea of risk stratification has been around for a long time,” says Carol McCall, chief health analytics officer for ClosedLoop.ai. “We’ve just brought it into the 21st century. What used to be the inputting of insurance claims data into prebuilt formulas is now the ability to use any person-linkable data and the latest AI and machine learning methodologies. And while the scientific and engineering breakthroughs needed to achieve this are impressive, the opportunities it opens up to promote health and reduce healthcare costs are even more impressive.”

In one of the platform’s first applications, Medical Home Network, the largest Medicaid-accountable care organization in the United States, used ClosedLoop.ai’s platform to risk stratify new Medicaid patients. ClosedLoop.ai was able to improve the accuracy of its risk stratification by 63 percent, reduce false positives by more than 80 percent, and hone the list of high-risk individuals for interventions. This made it possible for Medical Home Network’s care-management teams to more effectively focus their efforts, a change estimated as being worth $1.5 million the first year.

In one of its more recent efforts, ClosedLoop.ai developed and released the COVID-19 Vulnerability Index (C19 Index), a free open-source tool designed to help healthcare organizations identify and protect individuals who are most vulnerable to COVID-19. The C19 Index does not predict who will become infected with COVID-19 or identify where the virus might spread. Instead, it was created to help identify people with a heightened risk of severe complications should they become infected. When it was built, and because no data on COVID-19 cases was available, the C19 Index was developed using a surrogate endpoint for similar proxy events (e.g., pneumonia or influenza). It calculates vulnerability as a person’s near-term risk of severe complications from these respiratory infections. The response to the tool has been overwhelmingly positive. Since its launch, it has been downloaded and used by healthcare organizations serving more than 10 million people in the United States.

Benefits of Using AWS

The ClosedLoop.ai platform is a healthcare-specific, cloud-based, automated machine learning platform that enables rapid experimentation in a collaborative environment. Every step in the machine learning pipeline—from data cleanup, normalization, feature creation, model training, hyperparameter tuning, and explainable calculations to deployment and ongoing monitoring—requires substantial cloud resources. AWS brings significant expertise in cloud infrastructure and architecture and a robust, HIPAA-compliant environment that is critical to ClosedLoop.ai’s AI solutions, which allows ClosedLoop.ai to focus on data science and the needs of its customers.


About ClosedLoop.ai

ClosedLoop.ai is healthcare’s data science platform. It makes it easy and affordable for healthcare organizations to use data science to improve health, reduce costs, and change the experience of care. It is committed to bringing healthcare organizations the power to predict health outcomes in ways that are highly accurate and easily explained and to help them learn what interventions work and for whom.


Get Started

Companies of all sizes across all industries are transforming their businesses every day using AWS. Contact our experts and start your own AWS Cloud journey today.