HLTH.Rad Day 1 will trace the shift from traditional automation to truly multimodal, generative intelligence and separate hype from real, deployable value. Leaders will unpack why AI adoption stalls, how to build clinician trust, and what effective human-AI collaboration actually looks like in practice.

From single-site pilots to system-wide transformation, we’ll explore the operating models that scale, the evidence frameworks that matter, and the continuous-learning pipelines needed to keep AI safe, validated, and impactful in the real world. We’ll close by examining the foundations of an AI-ready health system required to power radiology for the next decade.

Radiology has moved beyond diagnostics to become the connective matrix of modern health systems, linking care pathways, enabling proactive population health, and powering smarter screening at scale. 

Day 2 will unpack the economics behind this shift: value-based imaging, evolving reimbursement, and the real ROI of intelligent imaging beyond productivity, before moving into the next frontier of operations: end-to-end workflow intelligence, predictive capacity planning, and the rise of semi-autonomous departments built for system-level efficiency. Finally, we tackle the strategic asset at the centre: data. We will map how institutions can compete, collaborate and innovate responsibly.

The Blue Marble, the first image of Earth taken by a human astronaut, showed our planet as a fragile, pale blue dot suspended in darkness. Commercial space travel may one day let more people witness that view. Suppose we apply the same futuristic lens to radiology. What do we see? 

Radiology’s “big picture” doesn’t come from zooming out, but from zooming in, capturing the physiological details of disease. The future lies in functional imaging that shows not just anatomy but how it works; integrated diagnostics that unite imaging, pathology and genomics; and AI accelerating every step.

In this session, early adopters explore where imaging innovation is headed. Which emerging technologies are today’s radiology leaders betting on for tomorrow?

Beneath the forest floor, threads of fungi speak a colloquial language, trading tiny molecules of energy between each other. Nature rarely works in siloes; everything is interdependent. Human-made systems struggle to mirror that balance. AI is not an exception.

AI is falling into a well-known healthcare trap. Radiology AI tools can spot lung nodules, fractures, and hemorrhages; some can even triage cases and streamline diagnosis. But they are not interconnected, there’s no all-in-one solution, and the true impact of AI is currently ill-defined. The future for radiology visionaries is one in which AI tools deliver on efficiency, integrate into existing workflows, talk to each other seamlessly and deliver on impact.

Integration is the future. Imaging revolutionaries have already made strides towards this. How can AI be accessed, adopted, implemented, and scaled, with that goal in mind?

In 2016, Geoffrey Hinton, AI’s “godfather” and a future Nobel laureate, predicted that AI would replace radiologists, saying we should “stop training radiologists now.”

Nearly a decade later, we face the opposite reality: the largest radiologist shortage in history, with a projected 30,000-specialist gap in the EU by 2034. For new radiologists, it’s a buyer’s market.

We don’t need to stop training radiologists; we need to train more, and fast. And ironically, the technology once predicted to make them obsolete may now be crucial to developing the next generation. AI-driven training platforms are already emerging, offering adaptive, personalised learning. High-fidelity simulation environments, much like aviation training, could allow radiologists to dedicate regular time to practising technique and reducing variability.

What other strategic approaches can health systems adopt to fix the radiology workforce gap and ensure long-term retention?

Advanced, analytical and interventional imaging is increasingly informing clinical decision-making. With this in mind, clinical care pathways need to be rethought. 

Overused, delayed, or poorly integrated imaging doesn’t improve patient care or deliver value for hospitals. For imaging to deliver all that it supposedly promises - earlier diagnosis, reduced unnecessary testing, and improved efficiency - actionable results are needed.

What information do physicians need to make these decisions? How do we maintain imaging as an enabler, not an obstacle? A gateway to diagnosis, not a gatekeeper?