AWS Public Sector Blog

Cleerly uses AI-driven heart imaging technology to help save lives with AWS

More than 50 percent of people who will suffer a heart attack—the most common cause of death in both men and women—will do so without experiencing any symptoms. Cleerly is a digital health company innovating a new standard for coronary artery disease (CAD). Cleerly goes beyond the traditional methods of diagnosing heart disease by using artificial intelligence (AI) and the Amazon Web Services (AWS) Cloud to enable non-invasive, comprehensive quantification and characterization of atherosclerosis (plaque) building in arteries.

Clinicians use coronary computed tomography angiograms (CCTA) as the first-line test for the evaluation of symptomatic patients with suspected CAD. Current clinical interpretation of CCTA is largely qualitative and significant variability has been observed between readers. There is an urgent unmet need to support physicians in the interpretation of the nearly 80 million pixels of data embedded within a single CCTA examination. An individual cannot ingest and accurately interpret all the data in a CCTA imaging dataset and also manage the current patient study volume in today’s healthcare environment.

Cleerly developed Cleerly Coronary and Cleerly Labs to address the qualitative analysis of CCTAs and provide comprehensive and quantitative assessment of CAD in a matter of minutes in ways previously not possible. With Cleerly, clinicians can quantitatively determine the burden and type of CAD to support physicians and patients in identifying the key features of atherosclerosis that have been shown in large-scale clinical trials to increase risk of heart attack.

Cleerly applies its AI algorithms to non-invasive CCTAs, an approach that has been shown to reduce heart attacks by 41 percent in stable symptomatic patients with suspected CAD, according to a recent study in The New England Journal of Medicine. Further, the latest research in large-scale multicenter clinical trials demonstrated that atherosclerotic (plaque) features are the strongest discriminants of future heart attack risk, findings that are uniquely and non-invasively visualized on a CCTA.

In the Cleerly workflow, Cleerly receives auto-uploaded DICOM CCTA data from a healthcare facility and then applies a series of FDA-approved software as a medical device machine learning (ML) algorithms to quantify atherosclerotic plaque characteristics and vascular morphology features for all identified arteries in the CCTA data. The Cleerly platform provides a rich diagnostic user experience for clinicians to review and interpret these values with clear actionable clinical insights they can then review with patients.

Figure 1: The Cleerly dashboard, featuring an example report and actionable insights for the reviewing clinician from an example patient's coronary computed tomography angiogram (CCTA).

Figure 1: The Cleerly dashboard, featuring an example report and actionable insights for the reviewing clinician from an example patient’s coronary computed tomography angiogram (CCTA).

Cleerly uses AWS to deliver these clinical insights at scale and with rapid turnaround times, which requires significant infrastructure and processing power. Researchers, developers, startups, and larger enterprises are using ML enabled by AWS to transform medical imaging by taking advantage of hyper-scale storage with cost-optimized life cycle, scalable microservices, and integrated AI and ML pipelines. AWS Batch helps to orchestrate the GPU-heavy Amazon Elastic Compute Cloud (Amazon EC2) P3 instances needed for Cleerly’s machine learning pipeline, which is an effective way to keep costs down and keep performance high. Amazon Simple Queue Service (Amazon SQS) is used to trigger the machine learning pipeline and return processing results as Cleerly receives incoming DICOM studies.

Deploying code through AWS CodeBuild and AWS CodeDeploy into the Amazon Elastic Container Service (Amazon ECS) cluster gives a simple, containerized runtime system that’s horizontally scalable. Serving Cleerly’s React-based web application through Amazon CloudFront provides flexibility in release cadence for the user experience with a highly available, fast user interface (UI) for use by clinicians. From a security perspective, leveraging HIPAA-compliant offerings in AWS made securing the environment and patient data privacy with end-to-end encryption straightforward to implement and audit.

Cleerly built their platform in AWS from the beginning, which allowed the team to focus on building the product rather than focusing on maintaining infrastructure and constantly expanding it. Instead of dealing with the maintenance and management of GPU servers in a co-location data center for the ML pipeline, Cleerly uses spot instances when necessary and moves to reserved instances when it makes sense to scale with ease. CodeBuild and CodeDeploy saves on the costs of owning complex build infrastructure within a virtual private cloud (VPC) or on premise and keeps the build definitions within source control for simple DevOps maintenance. From a redundancy standpoint and a disaster recovery perspective, using AWS helps Cleerly keep patient data secure and recoverable in disaster scenarios. Cleerly can shift the operation environment to any number of AWS Regions, quickly leveraging the data replication capabilities and infrastructure as code.

Cleerly’s data science organization can leverage AWS to quickly experiment with new algorithms and machine learning models, driving rapid innovation and decision making. Having the ability to roll out new features and experimentation to environments for testing, validation and review has been invaluable in Cleerly’s product development process. Learn more about Cleerly here.

For more resources regarding AWS in public sector healthcare, visit the public sector healthcare hub. AWS is the trusted technology and innovation partner to the global healthcare and life sciences industry, providing unmatched reliability, security, and data privacy. Learn how to get started with AWS for health here. Curious how AWS can help your healthcare organization innovate with new technology solutions? Contact us to learn more.

Subscribe to the AWS Public Sector Blog newsletter to get the latest in AWS tools, solutions, and innovations from the public sector delivered to your inbox, or contact us.

More AWS in healthcare stories:

James K. Min, MD

James K. Min, MD

James Min is the founder and CEO of Cleerly, a mission-driven healthcare company whose digital care pathway supports physicians and patients to reduce the risk of heart attacks. Previously, Jim served as a Professor of Radiology and Medicine (Cardiology) at the Weill Cornell Medical College and as the Director of the Dalio Institute of Cardiovascular Imaging at the NewYork-Presbyterian Hospital. There, he was the founder of the HeartHealth cardiovascular disease prevention program and led an array of seminal multicenter clinical trials that defined the relationship of coronary heart disease phenotyping using advanced non-invasive imaging with patient-centered outcomes. Jim has published more than 500 peer-reviewed scientific publications, and served as the editor for several textbooks, including most recently Artificial Intelligence in Medicine - Technical Basis and Clinical Applications.

Nick Nieslanik

Nick Nieslanik

Nick Nieslanik is the chief technology officer (CTO) of Cleerly, a mission-driven healthcare company whose digital care pathway supports physicians and patients to reduce the risk of heart attacks. Previously, Nick was the Head of Engineering at myStrength, a behavioral health startup providing self-service digital tools and content in areas such as depression and anxiety. There, Nick built the myStrength platform and technology organization from the ground up, leading to the acquisition of myStrength by Livongo in 2019. At Livongo, Nick was a Senior Director of Engineering, driving the integration of behavioral health into chronic condition products, such as diabetes and hypertension, to deliver a platform focusing on whole-person care. Most recently, Nick was VP of Engineering at Teladoc Health, leading the organization formerly known as Livongo and focusing on integrating chronic condition and behavioral health capabilities into the Teladoc telemedicine platform. Prior to his time in digital health, Nick was an engineer and manager at Microsoft spanning multiple lines of business including Bing, SQL Server, and financial products.

Razvan Ionasec

Razvan Ionasec

Razvan Ionasec, PhD, MBA, is the technical leader for healthcare at Amazon Web Services in Europe, Middle East, and Africa. His work focuses on helping healthcare customers solve business problems by leveraging technology. Previously, Razvan was the global head of artificial intelligence (AI) products at Siemens Healthineers in charge of AI-Rad Companion, the family of AI-powered and cloud-based digital health solutions for imaging. He holds 30+ patents in AI/ML for medical imaging and has published 70+ international peer-reviewed technical and clinical publications on computer vision, computational modelling, and medical image analysis. Razvan received his PhD in Computer Science from the Technical University Munich and MBA from University of Cambridge, Judge Business School.