At Intelligent Health 2022, the world’s leading healthcare AI summit, I was delighted to take part in a panel discussion on unlocking the value of medical data in the cloud. The panel was expertly moderated by Elena Bonfiglioli, GM Health & Life Sciences and Global Strategy Lead for Pharma Life Sciences at Microsoft, and alongside me were Matjaz Ladava, Microsoft’s Group Program Manager for Health Cloud & Data, and Julie Agez, IT Project Manager at the Centre Antoine Lacassagne, a cancer centre in France.
It was a fascinating discussion, so for those of you who couldn’t join us at the summit, I’ve captured some of the highlights here.
Barriers to overcome
While the panellists agreed that the cloud is vital to harness the full power of healthcare data, we acknowledged that the industry’s journey to the cloud is long and challenging.
A stringent regulatory environment can put many organisations off using data in the cloud to support healthcare AI use cases. And when healthcare organisations do implement data-driven solutions, there’s often strong resistance to change from clinical and support staff, making it hard to deliver full value from cloud and AI investments.
There are also broader industry challenges around interoperability and data standards. As Matjaz said: “The biggest problem we have to solve is interoperability. We need to get the data flowing before we can go into AI. The amount of data in healthcare is staggering, and if we could just unlock part of it and make it available for research, that would be incredible—that’s the core problem we’re trying to solve.” Matjaz explained that the industry is already making progress on interoperability, with the FHIR standard now mandated in the US and growing in the UK.
“Driving a culture and a practice of interoperability in healthcare is a journey we are on as we recognize the value and innovation that standards can bring to healthcare,” Elena said. “If we streamline interoperability— at a national level and across nations— as a key element of procuring any solution, we will build a stronger foundation to unlock the impact of existing and emerging health technologies,” she added.
The panel then discussed why structured data is essential for AI to support clinical decision-making and accelerate research. We also talked about how the workflow assistant Dragon Medical One turns unstructured narrative data into structured data to automatically create high-quality medical documentation and make it easier for clinicians to retrieve relevant patient information.
Cloud and AI in action
Centre Antoine Lacassagne is a Dragon Medical One customer, and Julie told us how the cloud-based documentation companion is helping clinicians and patients. She explained how it’s supporting the HOP’EN programme, a five-year plan to transform digital healthcare across France. As part of the programme, patients and clinicians can access and share medical reports on a cloud-based patient portal called DMP, aligned with “Ma santé 2022” strategy —and AI is making that much easier to achieve.
Julie explained that, because Dragon Medical One allows clinicians to automatically create medical reports directly in the EHR using their voice, it saves them time to focus on patients, and reports are quickly available on the patient portal.
“Dragon Medical One in the cloud is a game-changer,” Julie said. “Patients can access their medical reports quickly and easily, and all physicians can access the information they need about each patient.”
Focus on the users, not the technology
After many years of working with fellow clinicians to increase the adoption of digital solutions, I’ve learnt that deploying technology is the easy part of the project. Change management is the complex part of any digital transformation, so it’s vital to show users how the solution will have a positive impact on their working lives.
Julie explained that the technical implementation of Dragon Medical One was fast and simple, and that most of her preparation for the switch was focused on user adoption. “We didn’t spend a lot of time preparing servers and internal systems,” she said. “We just prepared at each step to manage the change with physicians and medical assistants.”
Julie’s team identified a few people in each department who could act as champions for the new solution, helping to overcome resistance to change. With careful change management, Centre Antoine Lacassagne now has 80% adoption of Dragon Medical One, which is a fantastic achievement.
The centre’s physicians can now spend more time with patients, and the solution has also elevated the role of medical assistants. “People love the solution because they can do more for their patients,” Julie said. “And we’ve noticed that medical assistants are reimagining their missions; they’re not just typing medical reports, they’re assisting patients along their complex cancer care journeys.”
The future of healthcare AI in the cloud
Cloud-based AI solutions like Dragon Medical One are already transforming healthcare by focusing on outcomes rather than technology environments. “Clinicians can see more patients if they can save time,” Elena said. “It’s not so much about whether the technology is in the cloud—it’s about whether it helps us give people the care they need.”
For Matjaz, regulation related to the use of data in the cloud needs to foster innovation, not hamper it. “Innovation means we need to think outside of the box,” he said. “Regulators need to think about are the outcomes we want to drive instead of the means we want to proscribe.”
When it’s focused on outcomes, AI can be a powerful ally for clinicians and patients. It’s clear from success stories like Julie’s, that voice is the new frontier of healthcare, and I’m excited to see how conversational AI in the cloud will transform the industry in the future. As Julie said: “We have a long road in front of us—but it’s really exciting.”