The SIIM 2026 trends shaping radiology’s future
The top SIIM 2026 trends made it clear that radiology has shifted from a discipline of “big iron” to one dominated by software. At the annual meeting of the Society for Imaging Informatics in Medicine, held in Pittsburgh, seven movements stood out and help explain where imaging informatics is heading through 2026: from structured reporting to the evolving artificial intelligence platform segment.
SIIM may not be the largest event on the radiology calendar, yet many consider it the most valuable one outside of RSNA precisely because it gathers the people who think about the information-technology side of the specialty. Below is what these seven themes mean in practice for service managers, radiologists and healthcare IT teams.
Reporting and AI adoption stay red hot
Radiology reporting remains on fire. It was already the dominant theme at SIIM 2025, but it grew even hotter after Microsoft announced it would sunset PowerScribe 360, one of the most widely used reporting platforms in the world. That decision opened the door to a wave of new competitors, and in 2026 the pattern was clear: enterprise imaging companies began embedding reporting modules directly into their solutions. This echoes what we covered when interactive multimedia reporting gained ground in radiology, blending text, key images and measurements into a single navigable document.
The second trend is AI adoption, which is advancing slowly but surely. At SIIM 2026, roughly 68% of radiology-oriented papers had some AI component. The new generation of foundation models and vision-language models drew particular attention, allowing targeted algorithms to be developed faster than ever. Integrating these tools into the workflow recalls initiatives such as NewVue integrating the report into the radiologist’s cockpit, cutting clicks and windows.
AI governance and the rise of other ‘ologies
The third trend may be the most mature: AI governance became a concrete agenda item. With dozens of algorithms in production, hospitals and clinics must answer hard questions. Who is accountable when an algorithm fails? How do you monitor a model’s performance after deployment (the so-called drift)? What are the legal and ethical duties of those who deploy AI in routine care? At SIIM 2026, providers debated the need for formal frameworks to manage the adoption, deployment and ongoing use of these tools, a topic tightly linked to clinical validation and risk.
The fourth trend shows radiology is not alone. Although it likes to treat SIIM as “its own” conference, the event also embraces other specialties moving into digital image management, such as pathology and ophthalmology. Several imaging IT vendors demonstrated integration with data from these disciplines, giving institutions a single source for healthcare data management. This convergence is an important step toward genuine interoperability between systems that historically lived in silos.
All-in-one vendors and the rise of agentic AI
The fifth trend is the consolidation of so-called all-in-one vendors. A growing number of imaging IT firms now offer solutions that combine viewer, worklist and reporting in a single platform. For the buyer, this simplifies purchasing, deployment and maintenance, avoiding the nightmare of stitching together components from different manufacturers. Many of these firms appeared to gain real traction with prospective customers, suggesting the all-in-one concept may finally have arrived.
The sixth trend is agentic AI. Instead of algorithms that merely flag a finding, developers are starting to build agents capable of taking on repetitive, administrative tasks, freeing radiologists to focus on what they do best: interpreting images. A fair question lingers, though: will agentic AI truly work in the real world, or simply pile more technology onto already overburdened clinicians? The balance between useful automation and technological noise will be decisive.
The uncertain future of AI platforms
The seventh and final trend concerns the AI platform model. Bayer’s exit from this market, when it pulled support for Blackford in 2025, raised questions that persisted at SIIM 2026. Platforms seem to be evolving beyond a simple algorithm “marketplace,” adding services such as AI monitoring and governance, precisely the pain points that bother customers most today.
Taken together, the seven trends paint a coherent picture rather than a list of isolated novelties. Reporting, AI adoption and governance reinforce one another: as more algorithms reach production, structured reports become the natural place to surface their outputs, and governance becomes the framework that keeps that output trustworthy. All-in-one platforms and agentic AI, meanwhile, are two answers to the same productivity pressure that has long defined the specialty. The common thread from Pittsburgh is that radiology’s value is increasingly created in software, and that the winners will be the vendors and institutions able to orchestrate it responsibly.
For practitioners worldwide, these trends carry immediate practical implications. Services planning to renew their PACS or reporting systems should evaluate all-in-one vendors with a critical eye, demand evidence of AI governance and monitoring, and plan for digital pathology integration now. Mid-2026 radiology confirms a direction: less iron, more software, and a growing demand for solid governance over every algorithm entering the clinical routine.
To follow the sector, it is also worth watching how major congresses set these agendas, as seen in our coverage of SNMMI 2026, with theranostics and cardiac PET leading the discussions. SIIM 2026 closes this picture by showing the pace of digital transformation in medical imaging.
Source: The Imaging Wire
