Ultrasound tool, fueled by artificial intelligence software, works like a navigation system to help providers capture the best cardiac images.
Capturing high-quality four-chamber cardiac ultrasound images can be a difficult task even for highly trained sonographers who have years of experience. Imagine if it were possible, though, to get a diagnostic-quality image after only one day’s instruction.
That’s the goal of UltraSight, a cardiac imaging company based in Israel. It’s an achievement that company leaders hope will facilitate easier care delivery, particularly to patients who currently do not have sufficient cardiac care access.
“Annual mortality from cardiovascular disease is a big, growing problem,” said Davidi Vortman, UltraSight chief executive officer. “We want to prevent a lot of these deaths. Many are avoidable if you can detect the disease at the right time, and ultrasound is the perfect tool to do that.”
Currently, cardiovascular disease (CVD) is the No. 1 cause of death globally and in the United States. Around the world, CVD claims 17.9 million lives annually, with three-quarters of these deaths occurring in low-to-middle income countries that have little access to cardiac care or early detection programs. The United States’ CVD problem is significant, as well, claiming the lives of 655,000 Americans yearly. And, caring for these patients carries a high price tag – it’s a $219-billion expenditure that is expected to top $1 trillion by 2035.
Ultrasound imaging is an excellent method for detecting and diagnosing cardiovascular problems, Vortman said, but the service isn’t offered everywhere largely because providers must have extensive training to achieve proficiency. With UltraSight’s new artificial intelligence (AI) platform, he said, every doctor and every nurse can learn to capture high-quality cardiac images in a day.
How It Works
To train sonographers, Ultrasound provides a basic overview of cardiac terminology and anatomy, as well as several simulations that give learners the opportunity to practice and test their new skills. The tool itself looks like most point-of-care systems. The probe is attached to a standard tablet, but it is augmented with AI software that works as an ultrasound guidance system, Vortman said.
The software, which was tested and validated with image data provided by New York University Langone Medical Center, assesses the input from the probe and creates a target with a crosshair guide that alerts the provider when the optimal image is achieved. As the sonographer moves the probe, a blue cross hovers around the grey crosshairs. When both crosses align, they turn green, indicating a high-quality image that should be captured.
“It’s so simple. It’s like a game,” he said. “You just need to put the cross in the right position.”
For example, he said, a four-chamber view of the heart is particularly difficult to visualize unless the provider is highly experienced. By using this tool, less experienced sonographers can feel confident that they’re securing diagnostic-quality images.
Vortman likened UltraSight’s ultrasound tool to a driving navigation system.
“Before Google Maps or Waze, you needed to study maps. You needed to be really good at navigating. But, now, you don’t need those skills,” he said. “In much the same way, we’re helping providers become much better doctors, much better nurses, by introducing a tool that can produce a much faster assessment of the patient that can lead to good clinical decisions.”
Alongside better image capture, he said, the clips from the patient's scan can, then, be sent to a cardiologist located elsewhere for timely diagnosis. By doing so, UltraSight can connect patients who live in low-resource, low-access areas to expert cardiac care.
Impact for Radiologists
In addition to the expanded care access patients will enjoy, radiologists will also experience benefits, said Laurence Grossman, M.D., a radiologist with the Cleveland Clinic who has observed UltraSight in action.
For example, it can help providers evaluate the potential cardio-toxic effects of chemotherapy with patients being treated for cancer.
“These patients need ejection fractions, and they need to know whether they can have another round of chemotherapy,” he explained. “In many instances, they’re a long distance from a major cancer center, but with this tool, you can see if the heart is functioning properly.”
In addition, he said, there is significant potential for it to be used for triage. It could allow providers to pivot a patient’s body, looking for ascites in the abdomen, distended bladder, bladder outflow obstruction, or hydronephrosis, and send these images to a diagnostic center for quicker access-to-care.
Using the tool does present challenges, though, said Grossman, who recently became part of UltraSight's Medical Advisory Board. A patient’s unique anatomy characteristics can present some AI challenges, but further work with the software will improve its visualization capabilities.
Trust in the technology could also be a factor, Vortman said. Overcoming the need to intently examine an image can be difficult for some providers.
“For some physicians, it’s hard for them to simply focus on the blue cross. They always like to look at the picture and don’t completely trust the cross,” he said. “We remind them that they don’t really need to look at the picture. The cross will guide them to the best image.”
Ultimately, Vortman said, UltraSight hopes to positively impact patient care.
“It’s about risk management. With value-based care become more and more prominent in the healthcare system, you really want to manage the risk of the patient,” he said. “If you catch things earlier, it’s a different ball game – costs will go down, and patient quality-of-life will go up.”
For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.