StrategyCore
Back to Resources
13/cybersecurity 7 min read

Securing Medical Devices in Japanese Hospitals: An IoMT Field Guide

Why connected medical devices are the hardest OT security problem, and how agentless intelligence covers them

A modern Japanese hospital runs thousands of connected medical devices, infusion pumps, patient monitors, MRI and CT scanners, and lab analysers, alongside the same building automation and network gear as any large facility. Almost none of them can run a security agent, most cannot be taken offline for a scan, and a growing share are past vendor support and will never be patched. The Internet of Medical Things, or IoMT, is where every hard problem in OT security meets patient safety and the PMD Act. Aimed at the vendor or hospital security lead who needs coverage without ever touching the device.

Medical device tiers

SC-LY-02 · REV A · 2026.07

Clinical network

hl7 · fhir

T1

Device segment

vlan · dicom · astm

T2

Monitoring + inventory

cmdb · passive

T3

No agent on the pump · no probe to a device in clinical use

Layer diagram of a hospital estate: the clinical network at the top, a segmented device VLAN speaking medical protocols beneath it, and a monitoring and inventory tier that identifies devices passively.
01

Why medical devices are the hardest case

A hospital device estate combines everything that makes OT security hard, then adds patient safety on top. An infusion pump or a bedside monitor runs closed, certified firmware that cannot host an agent and cannot be modified without breaching its medical certification. An MRI or CT scanner cannot be taken offline for an active scan without cancelling patient appointments. Many devices run embedded operating systems years past end of support. And where a factory PLC failure halts production, a compromised or failed medical device puts patients at direct risk, which raises the stakes well above an operational outage.

02

The protocols a general scanner misses

Medical devices speak their own protocols, and a generic network scanner does not read them. Imaging systems use DICOM, clinical messaging runs on HL7, lab instruments use ASTM, and newer systems expose FHIR over HTTPS. Without those protocols a scanner sees only an unidentified host on port 104 or 2575, with no model or firmware attached. Device intelligence that decodes DICOM, HL7, ASTM, and FHIR can name the device down to model and firmware, which is the identity a vulnerability match depends on.

03

Agentless coverage for devices you cannot touch

Because no agent can be installed and no scan can be risked, the workable path is the catalog-based approach: identify each device from its vendor, model, and firmware, taken from the hospital's existing inventory or read passively off the network, then look up its exact vulnerabilities, lifecycle status, and remediation. Nothing runs on the pump or the scanner, and no probe is sent to a device in clinical use. For medical equipment the vendor-direct data matters: in one comparison DeviceTotal produced firmware-exact vulnerability mapping for a GE LOGIQ E10 ultrasound that the manufacturer's own public data did not provide.

04

When the device cannot be patched

Much of a hospital estate cannot be patched at all: an estimated 46 percent of devices run firmware past vendor support. Disconnecting them is rarely an option and replacing them is a multi-year capital programme. The realistic control is to know precisely which devices are affected, which vulnerabilities are actively exploited, and what compensating controls exist, from network segmentation to access restriction and monitoring, for the cases where a patch never will. Lifecycle tracking flags end-of-life devices before support lapses, so replacement can be planned ahead rather than discovered during an incident.

05

Compliance and the Japanese context

In Japan medical device security sits under the PMD Act, and hospital procurement increasingly expects a defensible device risk position, alongside international frameworks such as IEC 62443 and FDA 21 CFR Part 11 for exporters. Audit-ready evidence of what devices exist, what they are exposed to, and how that risk is being managed is becoming a procurement requirement in its own right. For a foreign medical-device or security vendor, Japanese-language deployment and support, and on-premise options for patient-data residency, are the practical entry conditions.

// Key Takeaways

What to remember

  • Connected medical devices combine every OT security constraint with direct patient-safety and PMD Act stakes
  • Medical devices speak DICOM, HL7, ASTM, and FHIR, which general network scanners do not decode
  • Agentless catalog-based intelligence identifies medical devices without an agent or a risky active scan
  • An estimated 46 percent of devices cannot be patched, so compensating controls and lifecycle tracking are the realistic defence
  • Japanese hospital procurement increasingly expects audit-ready device risk evidence under the PMD Act

// FAQ

Frequently asked questions

Q1

What is IoMT security?

IoMT security protects the Internet of Medical Things, the connected medical devices in a hospital such as infusion pumps, patient monitors, imaging scanners, and lab analysers. Most cannot run a security agent and cannot be taken offline for scanning, so they need agentless approaches that identify and assess each device without touching it.

Q2

Why can't hospitals just scan or patch medical devices?

Active scanning can disrupt a device in clinical use, and an MRI or CT scanner cannot be taken offline without cancelling appointments. Patching is often impossible because the firmware is certified and closed, or the device is past vendor support. An estimated 46 percent of devices cannot be patched at all.

Q3

How do you identify a medical device without an agent?

By decoding the protocols medical devices actually use, DICOM for imaging, HL7 for clinical messaging, ASTM for lab instruments, and FHIR over HTTPS, a device can be named down to vendor, model, and firmware. That identity is looked up against a vendor-direct vulnerability database, with nothing installed on the device.

Q4

What compliance frameworks apply to medical device security in Japan?

Japan's PMD Act governs medical devices, and hospital procurement increasingly expects a defensible device risk position. Exporters also encounter IEC 62443 and FDA 21 CFR Part 11. Audit-ready evidence of what devices exist and how their risk is managed is becoming a procurement requirement.

Q5

What happens when a medical device cannot be patched?

You manage the risk instead of removing it. Know exactly which devices are affected and which vulnerabilities are actively exploited, then apply compensating controls such as network segmentation, access restriction, and monitoring. Lifecycle tracking flags end-of-life devices so replacement can be planned before support lapses.

Last updated:

Scoping Japan entry in this category?

If your company is weighing Japan entry in the work above, StrategyCore is the operating layer that carries it from first assessment to live deployments, run locally and in Japanese.