Arterys Stands Out for Having Multiple FDA Clearances for AI Enabled Medical Devices

In 2016 Arterys became the first AI company to receive?FDA clearance to use cloud-based deep learning in a?clinical setting. The company now has 6 FDA clearances 2 of which are for AI in clinical applications.

In 2011 four Stanford University graduates wanted to help doctors treat newborn babies born with congenital heart defects.?At that time, newborn infants with heart issues were subjectively diagnosed using ultrasound without quantification and partial views. They founded a company called Arterys and developed technology that made it possible for physicians to have a full view of a newborn's heart, see blood flow, and accurately diagnose newborn babies in 6 minutes.?

Prior to Arterys, nobody had developed a reliable, non-invasive, comprehensive diagnostic tool that enabled physicians to visualize and quantify blood flow for cardiovascular disease. The non-invasive standard of care was the echocardiogram, which was the first line of diagnosis but lacked blood flow accuracy. The alternatives to obtain more precise quantitative measures were expensive, time consuming, and invasive, and this data could not be read by legacy infrastructure. The cardiac application that Arterys developed reduced the time to process a case from 36 minutes to 6 minutes. Arterys’ cardiac application has already helped doctors treat more than 50,000 newborn babies with heart defects and is considered by doctors to be one of the top of AI solutions available.

Albert Hsiao, MD, PhD, Fabien Beckers, PhD, John Axerio-Cilies, PhD, and Shreyas Vasanawala, the founders of Arterys.

The World’s First Online Medical Imaging Platform

Arterys was founded in 2013 by Fabien Beckers, PhD, John Axerio-Cilies, PhD, Albert Hsiao, MD, PhD, and Shreyas Vasanawala. The founders ran the MR group at Stanford and thought that medicine should be powered by data and that doctors should be empowered with every piece of information they need to make the best diagnosis and treatment decisions. They also believed that every patient should have access to the best technology regardless of where they live or how they live. Healthcare technology infrastructure was not capable of getting medicine where it needed to go - so they decided to change the infrastructure.

One of the founders, Dr. Albert Hsiao, MD, PhD, is a radiologist with an undergraduate degree in computer science and a PhD in bioengineering. Dr. Hsiao initially designed software to analyze raw data from Stanford’s MR scanners but when he ran the software it captured so much raw data that hospital computers couldn’t efficiently process it into images. Arterys developed a cloud-based system to manage and process the gigabytes of MR data in each cardiac image. Arterys changed both the way image analysis is done and the infrastructure used.

Hospital IT Infrastructure is Antiquated

Early on the Arterys team concluded that the main limiting factor was the IT infrastructure. Radiology hasn't been upgraded in over 40 years and some hospital IT systems are in a pre-Internet state with workstations and local servers that lack the computational power that Al demands. There are thousands of data scientists and hundreds of startups building AI models that can’t be deployed on current legacy systems. Radiology is a fundamental pillar of patient care and the cloud is a critical component because it provides the compute power to effectively analyze and derive insights from Al, ML, and DL.

Healthcare needed a new deployment environment and platform infrastructure for AI - and this is what Arterys built. Arterys developed a true cloud native platform for diagnostic imaging, with unlimited computation, viewer processing images of any size at real speed, workflow integration, data privacy management, clinical workflow and AI combined, that is regulated. Arterys' platform brings together previously siloed data streams and enables greater collaboration and information sharing between physicians and patients.

6 FDA Clearances

After 6 years of development, Arterys has 6 FDA clearances (class II 510k) and radiologists around the world can receive automatic and very precise measurements of medical images of brain, heart, lung, liver, breast, and everything in between. Arterys provides clinicians with medical imaging viewing, Al-based analysis for interpretation, case sharing, and clinical collaboration all through a web browser. Physicians can interact with medical images of any size at real time speed, and Arterys provides the computation power necessary to process Al models on demand. The platform is in compliance with data privacy and security standards in over 95 countries.

15 seconds vs. 30+ minutes

To receive FDA clearance, Arterys had to pass a series of tests showing that its Al could produce results as accurately as human radiologists. Arterys' Al proved to be as accurate as humans but where the Al really excelled was in efficiency. In tests it took Arterys' Al 15 seconds to analyze a medical image whereas it took human radiologists between 30 minutes to an hour to analyze the same image (15 seconds vs 30+ minutes). And, unlike humans, Al can analyze medical images 24 hours a day, 365 days a year, and the more images it analyzes, the better it gets.

There's a severe worldwide radiologist shortage . To address this radiologists need a well optimized workflow where everything is streamlined. Every click counts. From the time it takes to open a case, receive it, process it - it all has to fit into their clinical constraint. It was a big challenge to make this available in the cloud and fully integrated with radiologists' daily practice, but after years of iteration, the Arterys team managed to solve this and now it's available at scale.?Arterys provides performance that reproduces a local experience - even if the medical images are processed thousands of miles away.

In 2018 Arterys received a unique FDA clearance of 2D and 3D (volumetric) quantification of lesions using deep learning across the whole body. This allows for much higher measurement of tumor progression, tumor response to treatment, and reduction of cohort size in clinical trials. The company is leading the industry with solutions designed to make reading and analyzing medical images of the brain, heart, lung, liver, breast, and other organs more efficient and accurate for radiologists. Doctors can use Arterys to measure and track blood flow in the heart, spot tumors and potential cancers, and easily apply radiological standards.

Healthcare is one of the last industries to leverage web-based technology. Before Arterys, the cloud wasn't embraced in healthcare in the way that it was in other areas such as FinTech. Due to fear of data piracy, hospitals were reluctant to send patients' health information to the cloud. Arterys solved this problem by creating a system called PHI Service which enables personal identifying information to be stripped from the imaging data when it is initially collected. Neither the doctors nor Arterys receive information that can be used to identify individual people.

This year Arterys is opening its API and an?appstore on its platform so that developers can build AI algorithms for medical imaging and within minutes make the algorithms available?to radiologists. AI models are analogous to mobile apps. It’s as though there’s been thousands of mobile?apps but no smart phone to host all of these apps. In essence Arterys built the first “smart phone” for AI algorithms that analyze medical images.

The World Needs Data Driven Medicine

Although an AI model can be built in 2 - 3 months it can take 6+ years to solve data privacy issues in different regions. Arterys' viewer can process medical images of any size from anywhere in real time. Arterys shifts the paradigm of medicine from observational to data-powered, and leverages millions of data-points from patients to provide predictive analytics and personalizing medicine, all via a simple web browser.

The world needs data driven medicine that leverages millions of patient cases to improve prediction, diagnosis, and treatment with an infrastructure designed to facilitate collaboration and knowledge sharing. It needs to be fully cloud native with an unlimited amount of computation at the fingertips of physicians and researchers. AI applications need to be created, distributed, and used openly, and data needs to be aggregated in a central location. It must be open and available to everyone. Arterys has built such a platform. Today Arterys has 6 FDA clearances, 70 employees, and the platform is used in over 100 hospitals around the world. The last mile of this journey is to bring predictive analytics to healthcare.?

Subscribe & Comment

I'm interested in your feedback - please leave your comments. Please click the subscribe button at the top of this article to subscribe.

Copyright ? 2019 Margaretta Colangelo. All Rights Reserved.

This article was written by?Margaretta Colangelo. ?Margaretta is a leading AI analyst who tracks significant milestones in AI in healthcare. She's consulting at AI healthcare companies and she writes about some of the companies she's consulting with. Margaretta serves on the advisory board of the AI Precision Health Institute at the University of Hawai?i?Cancer Center?@realmargaretta