Where radiology, AI and interoperability meet Leave a comment
Radiology is among the busiest departments in healthcare, with radiologists spending their days poring through a non-stop barrage of images to render their expert opinions.
Artificial intelligence has been making inroads into radiology in recent years. The technology is being used to assist – not replace – radiologists as they examine imagery.
AI programs, for example, can make a first pass at a batch of images, pointing out for radiologists what they identify as potential problems.
Another area of technology important to radiologists is interoperability – the lack of which is still often making it difficult for many radiology departments to make use of leading-edge technology, many radiologists say.
We spoke with two leaders from Philips’ enterprise diagnostic informatics business, who say interoperability and AI can transform radiology to help lighten the load on radiology staff, enhance patient-centric care and lower costs.
Madhuri Sebastian is general manager and head of data management and interoperability solutions at Philips, and Tanuj Gupta is its general manager and head of artificial intelligence informatics. They offered their perspective on the future of interoperable, AI-enabled radiology.
Q. How can digital transformation, interoperability and AI help lighten the load on radiology staff and deliver better patient-centric care?
Sebastian: Current trends in healthcare, such as the digitization of data and personalized medicine, are setting the stage for better care. But they’re accompanied by – and, in some cases, create – complexities that can jeopardize efficiency and outcomes.
For example, massive amounts of data have been and continue to be generated, but the data is fragmented among multiple disparate specialties and silos and is often inaccessible. To improve care, health systems urgently need improved interoperability and collaboration. They must find a way to transition to a streamlined, fully interoperable IT solution that benefits both administrators and clinicians and helps drive personalized medicine.
Gupta: In the “2021 Future Health Index,” a global survey of healthcare leader perspectives commissioned by Philips, more than half the radiology leaders surveyed cite a lack of interoperability and data standards as barriers to adoption of digital health technologies.
At the same time, they express a strong belief in the potential for artificial intelligence to help them do their job. A clear majority (82%) plan to invest in AI over the next three years, and they cite two very practical roles this technology could play.
“We take a collaborative ecosystem approach to AI innovation and implementation, supported by new business models such as cloud-based software marketplaces that can help accelerate adoption of AI.”
Tanuj Gupta, Philips
The first is to leverage AI to develop software tools that can assist with diagnosis, such as detecting pulmonary nodules. The second is to use AI to shift the burden of routine or mundane tasks like data entry from humans to technology to reduce clinician burnout.
Q. Every year, dozens of AI applications are being released in the market. What is Philips doing to leverage some of these innovations?
Gupta: Healthcare providers are seeking integrated, end-to-end solutions. At Philips, we take a collaborative ecosystem approach to AI innovation and implementation, supported by new business models such as cloud-based software marketplaces that can help accelerate adoption of AI.
Philips has built the infrastructure to allow our customers access to automated insights from third parties, either through partnership with companies such as IBEX and DiA Imaging Analysis, but also through integration into our Vue PACS primary reading environment, with the added ability to review, approve or reject automated findings.
With many potential partners already lined up, we envision a time when customers will have seamless access to multiple enhancements to their clinical and operational workflows – not just those developed by Philips, but also from third parties supported by our HealthSuite cloud-enabled platform and delivered where they’re needed most.
Q. How can interoperability enhance patient-centric care and lower costs?
Sebastian: Healthcare expenditure as a percentage of gross domestic product is too high in many countries around the world. In the U.S., it is close to 18% of GDP. When you consider the statistic that three out of every 10 tests are reordered because the original results can’t be found, you realize that the potential cost savings in radiology alone are very large.
Philips aims to address this issue by delivering interoperability solutions based on open standards that facilitate data exchange and collaboration across clinical domains, such as radiology, oncology, cardiology and image-guided therapy.
As an example, Philips is working with the Kentucky Health Information Exchange (KHIE) to help address challenges around data exchange and interoperability. KHIE is a state-run health information exchange program whose participants include more than 1,600 healthcare organizations representing 8,300 physical sites.
KHIE enables electronic exchange of patient health information – including lab results, radiology reports, continuity of care documents, and other clinical reports, immunization records and medication histories – to support improved health outcomes and reduce healthcare costs. KHIE also collects communicable disease and toxicology data for the state.
Philips’ vendor-neutral HealthSuite Interoperability solution is a key component of KHIE’s system, bringing together data from multiple vendors and sources in a scalable, future-proof platform to create a collaborative, patient-centered ecosystem.
The HealthSuite Interoperability solution uses open, global standards, such as XDS, XDW, HL7, FHIR and DICOM, allowing for providers across Kentucky to exchange clinical information, regardless of the EHR, PACS, or other IT systems and infrastructures they use.
Gupta: Another example includes the deployment of the latest Imaging Object Change Management technology in our new clinical informatics solutions, enabling automatic synchronization of data with a vendor-neutral archive.
This reduces the administrative burden of managing multiple archives – while ensuring data quality interoperability in this case – and facilitates long-term storage and recovery, further reducing the total cost of ownership as well as benefiting patients and clinicians by providing easy access to longitudinal data across a health network.
Twitter: @SiwickiHealthIT
Email the writer: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.
