AI Guidance Meets Hospital Hardware: HeartFocus Link Extends Cardiac Ultrasound Training to Cart-Based Systems

DESKi just released a product that could shift how hospitals approach cardiac ultrasound training. HeartFocus Link lets any existing cart-based machine deliver real-time AI instructions through a simple tablet and HDMI connection. The French company behind it aims to solve a stubborn mismatch between sophisticated equipment sitting in hospitals and the limited number of professionals skilled enough to use it for heart imaging.

Cardiac ultrasound demands precise probe movements and pattern recognition that take months or years to master. Yet demand keeps climbing. Between 2011 and 2021 ultrasound exam volume in the U.S. jumped 55 percent, from 38.6 million to 59.8 million. The sonographer workforce expanded only 44 percent in that span. Training program capacity grew a mere 23 percent. The result? Echocardiography reaches just 8 percent of eligible hospitalized patients even though studies show it lowers mortality and improves outcomes. Sonographer vacancy rates hit 16.7 percent in 2023 and stood at 12.4 percent in 2025. The Next Web reported these figures in its coverage of the launch.

HeartFocus Link attacks this gap from a different angle. Instead of asking hospitals to buy new handheld devices, it overlays guidance on the systems they already own. A tablet runs the software. An HDMI cable feeds the live ultrasound image to it. The AI then projects positioning cues directly onto that image. Users keep their eyes on one screen. No flipping between the machine display and a separate reference app. The system supports all 10 standard transthoracic echocardiographic views.

Two training-focused features stand out. Auto Record saves clips the moment image quality crosses a preset threshold. Real-time scoring delivers an objective percentage rather than subjective instructor judgment. These tools reduce the burden on expert sonographers who otherwise supervise every practice scan. Bertrand Moal, PhD, MD, and CEO of DESKi, put it plainly. “Cart-based systems make up the majority of clinical ultrasound infrastructure, yet utilization is often limited to a small pool of specialists and other professionals. The equipment is already there. What’s been missing is a scalable way to help more healthcare professionals build familiarity, competency, and confidence in cardiac ultrasound acquisition.” The statement appeared in the company’s Business Wire announcement.

DESKi founded in 2016 by brothers Bertrand and Olivier Moal. Bertrand holds a medical degree and PhD in biomechanical engineering. Olivier trained at Berkeley and EPFL. The Bordeaux-based team raised a $6 million seed round in 2025 led by impact investors. Their first product, the HeartFocus app, earned FDA clearance in April 2025 accompanied by a Predetermined Change Control Plan that eases future algorithm updates. The underlying models trained on more than 10 million data points.

Clinical evidence backs the approach. A peer-reviewed study in JACC: Advances showed that novice users, including nurses with no prior ultrasound experience, could acquire diagnostic-quality images of all 10 standard views. In one cohort 100 percent of nurses succeeded across 100 percent of patients. The AI demonstrated specificity above 99 percent. Participants reported confidence after roughly nine practice acquisitions. Research presented at the American College of Cardiology meeting found novice scans achieved more than 85 percent agreement with expert readings on key parameters.

The company already partnered with Inteleos to create the first AI-specific cardiac point-of-care ultrasound certification. That credential lets clinicians and institutions prove competence in AI-assisted imaging. It arrived in April 2026, just weeks before the Link launch. Early availability targets medical schools, residency programs, and dedicated ultrasound training centers. Clinical use on cart-based systems will expand later once regulatory clearances align.

But the HDMI overlay carries limits. Because the software sits outside the ultrasound machine’s native software stack it cannot process raw data directly. Guidance remains visual rather than deeply analytical. Competitors such as Israel’s UltraSight have secured FDA clearance for similar AI guidance across multiple systems. Larger imaging vendors hold established distribution networks and deeper pockets. DESKi’s modest funding means it must prove rapid adoption or secure strategic partnerships to scale.

Still the strategy feels pragmatic. Hospitals spent heavily on cart-based fleets over decades. Many of those machines sit idle for cardiac work because staff lack training. HeartFocus Link turns that installed base into a teaching tool without new capital outlay. It meets providers where they work. And it does so at a moment when health systems hunt for ways to stretch personnel amid chronic shortages.

Recent coverage echoes the excitement. A BriefGlance article published one day ago noted that 80 percent of clinicians list lack of training as the top barrier to point-of-care ultrasound adoption. Traditional paths require 20 to 50 supervised scans for basic proficiency. The AI route compresses that timeline dramatically by supplying constant, objective feedback. Supervisors gain breathing room. Trainees build muscle memory faster.

Industry observers on X described the move as practical AI adoption. One post called it “a smart implementation that bridges the gap between complex software and human health.” Another highlighted how the tablet-plus-HDMI approach could spread quicker than hardware replacement cycles. These reactions surfaced within hours of the May 2026 launch.

The broader HeartFocus platform already works on Butterfly Network handheld devices and through certain OEM integrations. Link completes the picture. It brings the same patented 3D multi-planar guidance, semi-transparent on-image cues, quality scoring, and auto-capture to the dominant hospital infrastructure. The company brands the full offering as an ecosystem that spans learning, certification, and earlier cardiac assessment in primary care, emergency departments, and rural clinics.

Questions remain about long-term integration. Will major ultrasound manufacturers embed similar AI directly? Can overlay guidance evolve into tighter software partnerships that unlock deeper analytics? DESKi clearly bets that starting with education and training builds user familiarity and data that later support diagnostic clearances. The Predetermined Change Control Plan signals regulators accept iterative improvement.

For now the product lands at an opportune time. Hospitals face pressure to do more with existing assets. Clinicians want tools that shorten the path to competence. Patients in settings far from echo labs stand to gain if more providers can obtain usable heart images quickly. HeartFocus Link does not replace sonographers. It multiplies their reach by helping others contribute reliable scans.

That distinction matters. The company stresses the current version supports education, training, and practice only. It carries no indication for diagnostic use or clinical decision-making. Future clearances could change that. Until then the focus stays on building skills at scale. The data suggest the need is urgent. So does the response from training institutions already lining up demonstrations.

DESKi’s bet rests on a simple idea. Give people the right prompt at the right moment and they learn faster. Overlay that prompt on hardware they already trust and resistance drops. Early trial results and the swift move to certification indicate the approach works. Whether it scales across thousands of hospitals will depend on execution, partnerships, and continued clinical validation. But the direction feels clear. AI guidance is moving from specialized devices into the everyday equipment that fills hospital imaging suites.