RSNA 2019 market report—the top trends you should know about

Enterprise imaging—image access from anywhere

Like last year, enterprise imaging was a hot topic among both exhibiting vendors and attendees at RSNA 2019.

There is currently a shift away from the de-constructed PACS approach towards buying one EI solution for all imaging, provided by a single vendor, where the main goal is enterprise-wide image access. The viewer, workflow, VNA and advanced tools do not necessarily need to come from one vendor, but customers want one provider and contractor.

Few health providers have made any significant progress in the adoption of EI due to difficulties in terms of organization and project orchestration, demonstrating a return on investment (ROI), and the lack of standardization. Issues related to implementation were a common theme. However, there are several successful cases and adoption is increasing.

Another interesting finding is that the ability to handle digital pathology in the EI solution has become almost a prerequisite when selecting a vendor. Speed of viewing, supporting the pathology workflow and, in the US, an FDA-cleared solution are key.

Workflow orchestration—improving operational efficiency

Improving operational efficiency was one of this year’s main topics and was promoted by almost all EI vendors. This year, it was more obvious that workflow solutions are provided by the big vendors rather than smaller third-party vendors (as an additional layer). Workflow functionality seems to be sold under the AI umbrella by most vendors.

AI is still a hot topic—but focus has changed

The discussion as to whether radiologists will lose their jobs because of AI seems to have shifted to a sense of impatience for real implementation to take place. Many expressed that they “have moved from fear to impatience.”

Another issue when it comes to AI is differentiation. Most vendors presented very similar AI solutions, probably because these areas have displayed a need and can deliver a positive ROI, or have available data.

The general perception is that visitors have now shifted focus from smaller start-up companies that provide one/a few applications to larger EI/modality vendors that can offer a full portfolio of AI apps under a single offering and platform (so-called marketplaces).

This year’s RSNA AI Showcase was somewhat of a disappointment. Its size was in inverse proportion to the amount of visitor traffic, the maturity of the products and the number of implementations. The number of companies in the AI Showcase increased yet again—from 43 in 2017 and 75 in 2018, to 129 in 2019— although those that participated are clearly facing tougher financial demands. We have probably reached the peak in terms of the number of companies (and perhaps the peak of the hype?) and can expect a consolidation in the coming years. We can also conclude that AI still has a long way to go before large-scale deployment, especially for diagnostic applications.

However, AI-assisted workflow functionality and structured information exchange have made further inroads into clinical practice than diagnostic tools. Workflow AI is now offered by most EI/modality vendors in the technical exhibition.

Another finding is that some teleradiology companies are building their own AI solutions and PACS applications, and they are among those showing the greatest interest in AI.

Cost models, payments and ROI were the unspoken questions this year—with no clear answers.

Breast imaging is advancing fast

Digital breast tomosynthesis (DBT) is developing very quickly and was promoted by several vendors (together with ultrasound and ABUS). When it comes to DBT screening, significant challenges and questions remain in terms of efficiency, efficacy, safety, and quality assurance as well as the long-term effect on mortality (according to at least two RSNA conference sessions).

MRI for breast imaging was a hot topic in sessions and among vendors, partly fueled by the recent study by Wouter Veldhuis, radiologist at the University Medical Center Utrecht in The Netherlands, showing a significant reduction in interval cancers for women with dense breasts.

Digital pathology—starting to take off in the US

Many conference sessions mentioned that recent FDA clearances have allowed the implementation of digital pathology. One session mentioned that the lag in comparison with the EU “is allowing the US to learn from early experiences and become leading edge instead of bleeding edge.”

Radiologists highlighted the need for pathology report access as this provides a valuable datapoint as diagnostics become more advanced (advanced cancer treatments).

Photographs assist radiologists

Photos taken by the GP, primary care provider or patients are more commonly being requested by radiologists as they can assist in many diagnoses. Numerous vendors showcased different solutions for this. Again, many care providers are facing issues with EI systems not supporting such workflows, and privacy/anonymization is a problem.

Structured reporting—a main topic as expected

A hot topic as usual, where radiologists highlighted the need for fully integrated structured reporting templates and integrations with AI.

Integrated diagnostics

There were several good conference presentations on the topic, including a presentation by Mitchell D. Schnall, Penn Medicine’s Chair of Radiology, who shared his thoughts.

Although the clinical benefits of integrated diagnostics are large, there was little focus from vendors other than Sectra on the topic, probably due to the lack of concrete customers.

The revival of the cloud

‘The cloud’ received much greater focus than at RSNA 2018 and was promoted by several vendors. Cutting costs and better accessibility to patient data seem to be the main drivers for going cloud. The question is whether the market is ready for utilizing public clouds yet. A major difference this time seems to be that patient integrity is being taken seriously.

Personalized re-planning—increased collaboration between vendors

We are seeing increased collaboration between medical imaging and radiation therapy. This is mainly driven by the need to adjust the treatment plan based on the initial CT scan and tailor it to changes in tumors and surrounding organs.

For example, Canon announced a partnership with RaySearch, and Fuji acquired the Japanese radiation therapy company Yokogawa. This was also a hot topic during various conference sessions.

Evidence-based purchasing

Evidence has become an important factor in radiology purchasing, which is reflected in vendors’ marketing efforts to back up their claims with proof of how their solutions can boost productivity and/or quality. This was especially noticeable looking at the messages in some of the leading vendors’ booths and pre-show articles.