We take pride in personally connecting with all interested partners, collaborators and potential clients. Please email us with a brief description of how you would like to be connected with Galois and we will do our best to respond within one business day.
Galois’s UPGRADE project provides a multi-layered solution that helps hospital IT teams make sense of their complex IoT systems, automates security where it’s possible, and assists and accelerates human IT teams where it is not.
Hospitals are facing an escalating cybersecurity crisis. The stakes are high: successful cyberattacks can lead to patient data theft, massive compliance fines, and even the failure of critical systems, directly endangering patient lives and causing entire hospitals to grind to a halt.
Today’s hospitals depend on vast, heterogeneous networks of interconnected devices – from infusion pumps and imaging systems to patient monitors and printers. A single hospital may operate thousands, or even tens of thousands, of devices across multiple vendors and operating systems. As these networks continue to grow, each new device added expands the attack surface and adds complexity, making them increasingly difficult to understand and secure.
Hospital IT departments are chronically understaffed and underfunded relative to the complexity they manage. Manually monitoring, updating, and securing all of the thousands of devices in a hospital is not realistic.
Networked medical devices don’t operate in isolation. A vulnerability in one system can serve as an entry point for an attack on a connected, far more critical system, while even well-intentioned changes can trigger unexpected, cascading failures. Yet hospital systems are so big and complex that no one knows where to begin to fix the problem. Manually mapping dependencies, vulnerabilities, and change impact is simply not feasible.
Medical devices routinely remain in service for decades, long after vendors stop providing updates. When manufacturers go out of business or discontinue products, vulnerabilities can remain indefinitely.
Even when patches exist, applying them sometimes requires complicated or risky procedures – such as disabling firewalls or taking critical systems offline. In a healthcare environment where lives are at stake, these steps can introduce more immediate risk than simply leaving vulnerabilities unpatched.
Hospitals prioritize uptime and patient safety, placing IT teams under constant pressure to avoid disruption. Without full visibility into the system-wide impacts, even necessary patches may be delayed to avoid unintended outages. This caution is reinforced by regulatory frameworks where noncompliance penalties are immediate, predictable, and expensive – often perceived as a higher-certainty risk than a ransomware event, thus discouraging changes that could meaningfully improve security but threaten audit outcomes.
Galois’s UPGRADE project provides a multi-layered solution that helps hospital IT teams make sense of their complex IoT systems, automates security where it’s possible, and assists and accelerates human IT teams where it is not.
At the core of our solution is 5STARS. This state-of-the-art verification engine uses automata learning to passively observe the live hospital network – every device, every packet, every path – and automatically generate a precise model of the system, tracing dependencies and clearly defining "good" network behavior.
This model serves as a baseline for formal reasoning and verification. Any proposed change or security mitigation – from device-specific patches to complex network configuration changes – can be checked against this 5STARS model. This allows the system to simulate the outcome in advance and formally verify that the change preserves all critical functionality.
For device-level security, UPGRADE uses a structured, validated AI to automatically crawl the web, along with device documents provided to IT departments, to find scattered documentation and vulnerability reports for each of a hospital network’s many devices. The AI even tracks patches for devices from companies that have been acquired or have discontinued their product, maintaining a traceable, digital thread to validated databases, including those hosted by CISA, MITRE, and the FDA. From this, we synthesize clear, concise, actionable instructions for safely fixing identified vulnerabilities, making securing a hospital’s massive network of devices actually manageable.
This same task would be an incredibly time-consuming, likely impossible effort for a human IT team – the number and diversity of devices, and the amount of constantly-updated documentation, is simply too large. But analyzing and summarizing vast amounts of technical data is what AIs do best, empowering the humans to simply work through the “fix it” list, using their judgment along the way.
Connected to hospital IT networks are a range of devices, including those that are in-service by the manufacturer, as well as legacy systems. For devices that are currently in-service, deploying hospital-specific patches may be infeasible as it may cause incompatibilities with future software updates from the manufacturer. Similarly, for legacy systems, patching the device itself may be infeasible due to sparse documentation, lack of ongoing vendor support, and risk of disrupting critical systems elsewhere in the broader hospital network.
Towards this end, we have partnered with Kry10 to develop a hardware dongle called Mitigation Guard. This device can be plugged in upstream of vulnerable devices, providing fine-grained traffic filtering and microsegmentation – detecting and blocking malicious network packets, and blocking access to non-essential or buggy features at the network (rather than device) level. This approach shuts down the highest risk attack paths and filters out potential threats without requiring changes to the medical devices themselves.
While current FDA regulations require a time consuming and expensive approval process before updating any medical device, the FDA is moving toward a framework that allows software updates for devices through Predetermined Change Control Plans (PCCP). This signals a shift toward the kind of informed, verifiably safe update process that UPGRADE provides. With UPGRADE, making hospitals safe from cyberthreats is not only possible, it’s the only feasible path forward for the future of reliable patient care.
5STARS provides full system understanding, helping identify the patches, changes, or strategies that will have the highest impact.
Gives IT staff the confidence that a patch or change will not break a critical system, as the mitigation is formally verified against the system model.
Dramatically reduces the time, effort, and expertise required for vulnerability mitigation, effectively boosting the capacity of a hospital’s IT staff to manage an otherwise overwhelming number of devices and updates.
By securing hospital networks against attack, UPGRADE significantly reduces the risk of massive compliance fines (such as those associated with HIPAA violations that could result from data breaches), while simultaneously reducing the risk of ransomware attacks that could cause systems to fail and put patient lives at risk.
Portland, OR
421 SW 6th Avenue, Suite 300
Portland, Oregon 97204
Arlington, VA
901 N Stuart Street, Suite 501
Arlington, Virginia 22203
© Copyright Galois, Inc. 2024. All Rights Reserved | Terms of Use | Privacy Policy
