Q&A: An ultrasound solution clinicians adopt in a heartbeat

Ian Bolland caught up with Hila Goldman-Aslan, CEO and co-founder of DiA, which provides an AI-powered ultrasound analysis solution – in particular focusing on its cardiac device to assist clinicians to measure ejection fraction more accurately. 

Tell us about how DIA started?

The story behind DiA begins at Israel’s Ben Gurion University, where Michal Yaacobi, one of DiA’s founder, was a M.Sc. student in bioengineering seeking a compelling project for her graduate thesis. That’s when she met Dr. Noah Liel-Cohen, head of Echocardiography from the nearby Soroka University Medical Center, and together they decided to address a major challenge they saw facing ultrasound users: easily and accurately analysing ultrasound images. The goal was to see if an automated, objective solution could effectively address the subjectivity typically associated with analysing ultrasound images visually so clinicians – with all levels of experience or ultrasound training – could make more confident clinical decisions. Together with Hila Goldman-Aslan and Arnon Cohen, DiA was established in 2010 to bring the benefits of AI-powered ultrasound imaging analysis to all ultrasound clinical stakeholders in more clinical settings.

Give us some insight in your product range and what it offers?

Ultrasound market is going through a major disruption. Millions of ultrasound scans are performed annually. Yet the acquisition and analysis of ultrasound images is mostly done by visually ‘eyeballing’ the scan. This makes the process of capturing images and conducting an objective, reproducible analysis both prone to errors and highly dependent on each individual user’s experience.

DiA’s goal is to address these challenges by empowering ultrasound users with a clinical support tool for automatically and objectively acquiring and analysing ultrasound images. Our proprietary AI-based solutions use a combination of advanced pattern recognition, machine learning algorithms and deep learning technology to provide a powerful solution for ultrasound analysis that works within clinicians’ existing workflow. By simplifying and automating the process, DiA enables clinicians to objectively identify clinical abnormalities in order to guide patient management. This extends to clinicians with various levels of ultrasound experience and in a range of hospital or point of care (PoC) settings, including clinics, emergency departments, ambulances, or the ICU.

DiA has already made a significant impact on cardiovascular ultrasound, with its LVivo Toolbox. DiA’s solutions can be used for ejection fraction (EF) measurements, heart disease detection and identifying abnormalities during coronary events. In recent years, DiA has rapidly expanded its presence in the ultrasound market through partnerships with several medical imaging manufacturers as well as technology giants, Google Cloud, and IBM Watson Health. 

How are DIA’s products different from other ultrasound devices?

With four FDA approvals and CE marks to date, DiA is the leading provider of automated AI-powered ultrasound analysis, addressing a market of 8,000 hospitals and 18 million clinicians in the U.S. alone. Its AI solutions are used by thousands of end-users and sold to clinicians and hospitals through collaborations with industry leading distribution partners, including GE Healthcare, Konica Minolta and IBM Watson Health.

Because DiA’s AI-based solutions are cross-platform and vendor-neutral, they can be easily added to any ultrasound device or healthcare IT system, just like an app, to easily support clinicians’ workflow. With just one click, ultrasound users can run automated AI-based analysis on cardiac ultrasound images to receive results within seconds for various cardiac measurements, including ejection fraction, strain, segmental wall motion and from the Short Axis view. DiA solutions will soon extend beyond cardiology, to other anatomical areas for ultrasound including lung, bladder and more.

DiA’s AI solutions are designed to run on any ultrasound device or healthcare IT system - from handheld devices requiring low CPU or memory, to cloud-based imaging systems. DiA is already offered as an add-on solution to hospitals, clinics and other healthcare providers through channel partners, on devices like GE Healthcare’s portable ultrasound, Vscan Extend and more recently, as part of IBM Watson’s AI Marketplace.

What was the aim of the product in the first place?

DiA’s aim remains true to its co-founder’s original goal: to help the growing number and different types of ultrasound users leverage ultrasound imaging to support their patient management goals. By eliminating the variability associated with visual ultrasound analysis, DiA empowers clinicians to make faster and more accurate clinical decisions about appropriate treatment. Our ultimate vision is to create AI-powered solutions for clinicians’ toolbox that extend across all ultrasound segments.

How have clinicians responded to it?

Clinical response to DiA has been overwhelmingly positive. We work in close partnership with several leading KOLs to understand DiA’s impact in the field and gain insights on evolving clinical needs. Our advisory board includes prominent clinicians from world-leading medical centres, including Dr. Martin Goldman and Dr. Bret Nelson from Mount Sinai Hospital, Dr. Steven Feinstein from Rush University Medical Center and Dr. Christopher Moore from Yale University School of Medicine. Additional evidence of DiA’s impact with clinicians exists in demand for our AI-based cardiac ultrasound solutions from thousands of end-users, who have added LVivo to their handheld device or healthcare IT/ PACS through DiA’s channel partnerships. To date, DiA’s solutions are being used by 1000+ end-users in over 20 countries, including UK, Italy, Germany, Australia, South Korea and Japan.