Key trends in enterprise imaging
Radiology is going though a period of disruption with a growing radiologist shortage, decreasing reimbursements and increasing numbers of exams, making workflow efficiency a critical concern. Health systems are also facing increased financial pressures and hospitals are looking for ways to reduce their operational costs. Improving operation efficiencies is also a goal to help reduce burnout and address the current radiologist and technologist shortage.
Technologies, including enterprise imaging platforms to replace traditional PACS, cloud data storage, artificial intelligence (AI), workflow orchestration software, remote work capabilities and ways to mitigate cybersecurity are all playing a bigger role for radiology departments are looking to replace IT systems.
Health Imaging spoke with numerous experts in radiology IT system technologies to get their thoughts on what health systems are looking for in the post-COVID world impacted by new market pressures. While many of the technologies in customer discussions have been around for a while, there is new emphasis on issues bringing them front and center for adoption.
"You want to maximize what the IT tools are providing to you, not just to create better clinical outcomes, but also make your reading more efficient. You want to make sure that you are leveraging all of what that application is providing so you're not just getting the clinical information, but you're utilizing that clinical information for higher efficiency," explained Kyle Souligne, director of enterprise imaging radiology at Agfa Healthcare.
He said enterprise imaging enables all image-producing departments in a healthcare system to deposit their data into one location rather than having numerous, disparate data silos and then attempting to make these systems work together. With most hospitals now using electronic medical records (EMR) to manage all patient data, imaging and reports need to be readily accessible and it is much easier to do that when everything is located in one place.
Long-held IT management strategies are also being challenged as health systems are looking for ways to get out of the data management business and refocus their IT department efforts on clinical systems and improving staff efficiency and workflow. This includes a rapidly rising interest in cloud storage rather than expanding and maintaining ever growing on-premise server farms.
"Healthcare systems are struggling with managing everything on their radar whilst dealing with financial pressures, inflation and the workforce crisis. So that's where outsourcing is becoming more prominent within the market," explained Amy Thompson, research manager for healthcare IT at the healthcare market analysis firm Signify Research.
The need to eliminate data silos across healthcare
Thompson said there is a noticeable shift towards cloud adoption and the development of more cloud-native architectures in radiology informatics systems. Cloud technology, coupled with AI, has also become crucial for telecardiology practices meeting and maintaining service level agreements (SLAs) with healthcare providers, she explained.
"RSNA is a really great show to start seeing more progress towards cloud adoption and more cloud native architecture. A lot of these bigger, longer-term trends are more iterative stages for vendors. So there are not necessarily big, bold out-of-the-blue announcements that take me aback and go, okay, I wasn't necessarily expecting that. They're more so these phase approaches towards achieving these long-term goals." Thompson said.
The global healthcare workforce crisis has prompted a strategic approach in teleradiology, not only as a response to the crisis, but also as a retention strategy. Thompson emphasized that offering a more innovative and streamlined technological solution can make the field more attractive to radiologists, contributing to both recruitment and retention efforts.
“There is huge interest in cloud transformation, just moving enterprise imaging platforms to the cloud. And it's really driven by two things. One is the operational efficiency, everything from the place of care moving outside the hospital—that's true for radiologists and where they do their work, to where patients are getting their care. And the second is the financial pressures that so many health systems are under, and looking to find ways to invest in technologies that help them lower their operating costs," explained Tracy Byers, the CEO of enterprise imaging at Change Healthcare.
She said institutions seek avenues to either generate more revenue or manage costs more effectively. The adoption of cloud-based systems alleviates the challenge of managing multiple software versions. This not only ensures heightened cybersecurity measures, but also streamlines maintenance and support, significantly prolonging the life and relevance of IT systems.
Most hospitals have moved away from separate silos for echocardiograms, CT, MRI, nuclear imaging and X-ray. But new silos have emerged as health systems attempted to merge all their imaging data into one location.
"They are realizing dermatology photos are in one location, wound care photos are in another. The radiology DICOM images that they've been managing for years could be in 12 different picture archiving and communication systems (PACS) systems they have. They are also trying to capture digital pathology and its large gigabyte size files, and they're not able to bring it all together in one place. So these things that we've been talking about for 15 or 20 years—how do you eliminate silos—it now seems to be taking on new life with our customers because they want to drive AI initiatives. They want to drive research, they want to be interoperable, make the data available to their clinicians. And they're struggling with the old ways of doing things where it is just not possible to make all those new initiatives a reality," explained Lyle McMillin, principal healthcare product manager at Hyland.
Beyond just being able to access patient data more easily, cloud data storage can help health systems that want to leverage AI models and support research initiatives by making data more accessible.
"What we're seeing is customers are wanting to make that next leap forward. They are wanting to put that data to use, whether it be to drive AI models that they're interested in procuring and bringing on site. They're interested in it for academic medical centers. They're interested in providing that content to the researchers so they can drive new research initiatives within the organization, and then just overall helping the clinicians make better decisions with the data they manage. But just like any other large activity around data, most of the time is spent wrangling it into one place," McMillin said.
Google Cloud is one of the big players radiology IT vendors are partnering with to increase bandwidth and cybersecurity of data. Image courtesy of HIMSS
Movement to managed cloud services
Cloud also allows hospitals to outsource management of some IT functions to free up their own IT staff to work on more pressing projects. This has led to an increase in managed services offering to help simplify IT needs, lower costs and improve efficiencies.
These managed services are all hosted on the cloud, which provides not only off-site, managed data storage, but also software as a service offering. This includes enterprise imaging, PACS and access to advanced visualization, subspecialty reporting modules, and the ability to access AI algorithms integrated into the radiology workflow. The advantages of this are immediate updates for the software on a regular basis, access anywhere via web connections, elimination or reduction of maintaining on-site servers, increased cybersecurity, and the ability to off-load some tasks from hospital IT teams.
“We saw this trend coming. It was inevitable and it has happened in other industries. The shift from big capital purchases into ongoing consumption-based subscription models where organizations can scale as they grow and pay for the services that they actually use is here to stay,” explained Isaac Zaworski, president of Sectra U.S.A. Inc. "There has been a big push to look at how to modernize infrastructure to realize efficiencies at the broader enterprise organizations level. So we see a lot of trends of organizations that are looking to make the pivot into managed services and to transition out of hosting on-premise major IT infrastructure projects.”
"The executives we speak to are keen on moving away from managing on-premise data centers due to staffing shortages, rising security demands, and the overall burden of maintaining infrastructure. The cloud is emerging as a viable solution to free up resources for more valuable tasks," McMillin added.
Vendors partnering with major cloud providers to boost cybersecurity
Some vendors offer their own cloud services, but all of the major vendors appear to be partnering with big names in cloud storage to ensure higher levels of cybersecurity. In doing so, they are able to take advantage of dedicated IT staffing to maintain system up time and make large data transfers and scaling capacity easier.
Several vendors have announced partnerships over the past couple years with Google, Amazon Web Services or Microsoft Azure.
Many healthcare organizations are realizing they lack the expertise to maintain robust cybersecurity measures 24/7, so there is a growing willingness to entrust this responsibility to cloud providers with dedicated resources. The shift is seen as a pragmatic response to the increasing complexity and sophistication of cybersecurity threats, explains Monique Rasband, vice president of strategy and research for imaging, cardiology and oncology at KLAS Research.
"Providers are saying they now don't have the staff to keep up with it and there are so many things they need to work on. They want to work on a strategy to broaden their enterprise imaging offering, deciding what they are going to be putting in the VNA or possibly replacing a legacy PACS. They have a lot of problems to solve, so all of a sudden giving up cybersecurity to somebody else sounds really good. And so, we have definitely seen a shift post-pandemic of them being more willing to hand that piece over," Rasband said.
In today's world where ransomware is a major concern for healthcare institutions, which have become the No. 1 target of these cyberattacks, cloud makes a lot of sense.
Zaworski stressed the importance of partnering with specialized security providers like Microsoft to fortify institutions against evolving cyberthreats, allowing hospital IT teams to focus on other projects at their institutions.
Leveraging AI for workflow orchestration
Workflow orchestration tools are increasingly being in integrated into enterprise imaging systems. The technology not only orchestrates the radiology reading list, but also manages various tasks, including wet reads, post-reads, and the utilization of artificial intelligence to prioritize studies.
"What we're doing is helping prioritize to get the right radiologist the right read. But it's not just about that; radiology is complex. So most radiologists aren't just sitting in one location reading for one facility. They may be reading for multiple facilities and it doesn't matter where those studies live. We're just going to make sure that all of the work that needs to be done is orchestrated efficiently to keep the doctors in the flow so that they're getting the most done and have time for their life after their shift," Souligne said.
He and other vendors we spoke with said workflow orchestration can help address challenges with readers cherry-picking studies and workload balancing. The orchestration tool's flexibility allows facilities to set rules that fit their organizational needs. It ensures that radiologists receive a balanced mix of study complexities, aligning with their expertise.
"Cherry-picking is definitely an issue, and one of the things with workflow orchestration is how strict you want to be with it. So we can push just five studies, we can push 10 studies, we can push just one study. So, it's read next, read next, read next. The orchestration tool has to make sure that it is feeding the radiologist not just the right study, but the right series of studies so that they're being the most efficient with their time," he said.
AI can serve as a radiology workflow assistant
AI has seen an explosion in radiology, with medical imaging making up more that 75% of 700 U.S. FDA cleared clinical algorithms. AI is reaching a point of maturity and data where many health systems want to adopt the technology to make radiologists more efficient and to act as a second set of eyes.
"Now we're starting to see the point where algorithms are stable enough that organizations are deploying this at scale in a production environment,” Zaworski explained.
All the major radiology IT vendors now offer both AI they developed themselves or via partnerships with numerous start-up AI companies that have emerged as leaders in specific imaging areas. However, Zaworski and other vendors we spoke with said seamless integration of AI into the radiologists' workflow is critical, otherwise no one will utilize it.
"They are continuing to evolve multiple AI solutions in the market, but they don't fit readily into the radiologist workflow quite yet. And while there's a desire to leverage these technologies where applicable, nobody is really solving that last mile problem of embedding it appropriately in the radiologist's workflow," explained Ashish Sant, general manager for imaging at Merge by Merative. He said radiologists need to ask users of a particular PACS or enterprise system how well the AI integrates with the workflow.
One of the recent additions in AI that is gaining a lot of attention is natural language processing to automate and structure report data. Zaworski highlighted the potential of generative AI models in enabling radiologists to articulate their interpretations naturally, while AI algorithms autonomously structure the reports, significantly expediting the diagnostic process. There have been many complaints that forcing radiologists to used structured reporting drop down menu choices just does not work because there are so many variables.
“At RSNA we are demonstrating some really early prototypes where we can automatically structure report data without forcing the radiologist to conform to a structured template, clicking through a lot of boxes and ticking through things," Zaworski said. "They can just speak naturally, interpret in a way that focuses on the images and is comfortable to them, and then use the generative AI technology on the backend to pick out the relevant details and automatically structure them into reports that then add a whole lot of value downstream in the overall diagnostic chain."
Use of AI is also being increasingly viewed as being on the cutting edge of new technology, especially when an algorithm has study data backing its claims of increasing accuracy, cutting read times and catching findings that might otherwise have been missed.
"A big trend within the world of teleradiology is AI adoption increasingly seen in the mainstream hospital segment. We are seeing an increasing amount of deals and AI vendors announcing partnerships with teleradiology providers. When you are looking at the efficiency gains, that is one of the more critical markets that need it because the teleradiology providers see it as an opportunity to stand up competitively against the others," Thompson explained.
Video interviews on enterprise imaging and PACS from the past year:
3 key radiology IT systems trends - Rik Primo, consulting firm Primo Medical Imaging Informatics
Cloud image storage for radiology is a growing trend in healthcare - Monique Rasband, KLAS Research
Cloud and AI are key to teleradiology success in post-COVID resurgence - Amy Thompson, healthcare market analysis firm Signify Research
Trends in radiology enterprise imaging systems - Isaac Zaworski, president of Sectra U.S.A. Inc.
Optimizing reading efficiency to address radiologist shortages - Kyle Souligne, Agfa Healthcare
Radiology staffing shortages increasing demand for AI and cloud - Ashish Sant, Merge by Merative
Radiology embracing the flexibility of imaging data stored off-site in the cloud — Lyle McMillin, Hyland
Trends and tips in healthcare cybersecurity - Jon Kimerle, Pure Storage
Improving interoperability between the PACS and EMR - Rik Primo, consulting firm Primo Medical Imaging Informatics
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: dfornell@innovatehealthcare.com