Healthcare organizations are facing massive increases in network traffic and complexity from the adoption of AI, cloud services, electronic health records, virtual healthcare, digital imaging, and connected devices – and these pressures aren’t going away anytime soon. These organizations need to act quickly before the network becomes a bottleneck, holding back patient care and standing in the way of innovation and cost savings.
In our last blog, we looked at the pressures that new technology developments are putting on networks in this sector. What approaches, then, should healthcare organizations take to build a network that can cope with these evolving demands?
First, they need to go beyond the idea of bigger and bigger pipes, and explore a range of connectivity options that are more adaptable to the varying requirements of different applications and use cases.
They need to abandon outdated security models and adopt zero trust and cloud-based approaches to protecting complex, disparate networks. And finally, they need to shift from reactive management to a proactive and intelligent network operations model that takes the pressure off overstretched IT teams.
The Range of Connectivity Types Is Expanding – and That’s a Good Thing
Connectivity options are developing all the time, so healthcare organizations can expect to see their networking choices continuing to widen. The ongoing rollout of low Earth orbit (LEO) satellites, for example, offers healthcare organizations an increasingly affordable form of wireless connectivity, particularly in areas where 5G isn’t yet available.
Although satellite connectivity still can’t reliably provide low enough latency for many critical healthcare applications, techniques like adaptive latency compensation may be able to offset that issue. A surgeon in Lhasa, for example, recently carried out successful liver surgery on two patients in Beijing using satellite connectivity and latency compensation mechanisms.
A range of connectivity options is also important for creating true network diversity. So if the primary connection goes down (say, a fiber link is damaged or a cellular network provider has an outage), the network can fail over to a backup option that uses a different technology, making sure critical healthcare services aren’t interrupted.
Patient Care Needs More Versatile Connections, Not Just Bigger Ones
As we explored in our last blog, newer developments like AI, IoT, and wearable devices can have wildly different performance, latency, and security demands.
Throwing more capacity at the problem will only take healthcare organizations so far in meeting these requirements, and this approach could end up being very costly. Instead, they need to look at networking types that can be more closely tailored to the needs of the applications and services running over them.
For example, 5G standalone (5G SA) has great potential for the healthcare sector because it offers new capabilities that earlier versions of 5G can’t, like higher speeds, lower latency, and network slicing.
This fully fledged version of 5G can provide the extremely low latency that applications like remote surgery need to be safe and effective, avoiding the lag that plagued older cellular technologies. It could also help to improve patient outcomes by allowing experts to collaborate on complex cases even when they’re in different locations.
5G SA can also support an enormous density of devices – potentially up to one million per square kilometer. This allows it to comfortably support the rising numbers of devices within hospitals that need network connectivity (which include equipment such wheelchairs and portable ultrasound machines, observation devices like blood pressure and blood glucose monitoring devices, and smart devices like ventilators and infusion pumps).
Why 5G SA Is So Adaptable and How It Helps Save Lives
Most importantly, though, the network slicing capabilities of 5G SA allow IT teams to create dedicated virtual network “slices” with the performance and security that each healthcare application needs.
Remote surgery, say, could be given the highest priority on the network so it’s not affected even if there’s congestion, while patient information could be transmitted over a slice that’s ultra-secure and compliant with data protection regulations.
5G SA also helps to join up different elements of critical care to improve outcomes. Bangkok’s Siriraj Hospital, for example, has improved the survival rates of critically ill patients transported by ambulance by deploying a 5G SA network. This high-speed and ultra-low latency connectivity saves the emergency department critical preparation time by transmitting essential data like vital signs. It also supports high-definition video streaming so that specialists within the hospital can consult with paramedics during the journey.
Security Needs to Support a Distributed Organization
Because healthcare networks no longer have a perimeter that can be secured, many organizations are moving toward a “zero trust” approach.
Zero Trust Network Access (ZTNA) continuously verifies the identity and privileges of users, applications and devices before allowing them access to network and IT resources.
This is a much more effective approach to securing the enormous numbers of endpoints, devices, systems, resources, and users that are involved in healthcare operations. It also prevents attackers moving laterally from one system to another (from a compromised IoT device to a patient records database, for example) if there’s a breach.
Secure Access Service Edge (SASE), which brings together SD-WAN and cloud-based security capabilities like ZTNA, is another approach that’s gaining traction.
SASE provides consistent security policies and network performance from all locations. So a medical professional might access patient records or conduct a video consultation from within the hospital or from a mobile or satellite clinic, but the network performance and level of security stay the same.
SASE can be extended to medical devices that are part of “hospital at home” services, protecting data that’s transmitted from patient homes too.
Complex Networks Need Intelligence for Proactive Management
In the future, hospitals could have tens of thousands of devices connected to networks that are already extremely complex. Managing this infrastructure is becoming more and more difficult, network visibility is often poor, and IT and security teams are forced to operate reactively to problems as they arise.
Building agentic AI into network and security operations allows healthcare providers to intelligently tie together all the different parts of the network – the connections, clouds, vendors, tools, platforms, and devices – into a single system. This lets network teams:
• unify visibility, orchestration, management and analytics across the whole network, including wireless
• automate standard workflows like triaging connectivity problems
• combine network and security to reduce blind spots and vulnerabilities and speed up threat detection and breach resolution
• reduce “alert fatigue” and proactively identify problems before they start impacting users or critical systems
In the future, we’re likely to see networks becoming increasingly autonomous, allowing them to head off problems before they arise. So, for example, an autonomous network could predict a bandwidth bottleneck, reroute traffic to avoid app performance being affected, and analyze whether more capacity needs to be ordered.
The Future of Healthcare Will Be Built on Intelligent, Adaptive Networks
To address growing data flows and network complexity, then, healthcare organizations need to explore connectivity that responds more dynamically to the traffic and apps running over it. New approaches to security will protect the growing infrastructure sprawl, and network intelligence will help simplify management and put tech teams on the front foot when it comes to security and troubleshooting.
To see network modernization in action, find out how a scalable, cloud-ready network implementation helped radiologists at a leading US healthcare provider improve productivity by 50%.