IEC 81001-5-1 & MDCG 2019-16: Medical Device Cybersecurity

  • IEC 81001-5-1 plugs security activities directly into IEC 62304 software life cycle processes.
  • Manufacturers must produce a machine-readable Software Bill of Materials, typically in CycloneDX or SPDX format.
  • MDCG 2019-16 and IEC 81001-5-1 both require a formal Coordinated Vulnerability Disclosure process post-market.
  • Structured threat modeling, such as STRIDE, must trace identified threats to design requirements and tests.
  • Securing hospital-facing DICOM connections with mutual TLS reduces the attack surface under these frameworks.

International standard IEC 81001-5-1 and European guidance MDCG 2019-16 require medical software teams to build security controls into every stage of the software life cycle. Instead of relying on generic advice like code sanitization, these frameworks mandate formal threat modeling, an accurate Software Bill of Materials (SBOM), and post-market vulnerability handling. These security controls link directly to clinical patient safety under ISO 14971 risk management.

This article provides engineering orientation, not legal advice, and a notified body or regulatory consultant decides for a specific product. It shows software engineers how to meet medical device cybersecurity expectations using practical development workflows.

What do IEC 81001-5-1 and MDCG 2019-16 require for medical device cybersecurity?

Under European Union Medical Device Regulation (EU MDR 2017/745), cybersecurity is a legal market requirement. Annex I General Safety and Performance Requirements (GSPR) 17.2 and 17.4 require manufacturers to develop software using the state of the art. Software must incorporate information security, risk management, and unauthorized access protection across its life cycle.

The European Medical Device Coordination Group issued MDCG 2019-16 Rev.1 to interpret these rules. The guidance covers pre-market design, manufacturing, post-market surveillance, and vigilance reporting. It establishes that security is not a single feature. Regulators evaluate whether your software maintains confidentiality, integrity, and availability in its operating environment.

To satisfy the state-of-the-art requirement, engineering teams turn to IEC 81001-5-1:2021. This international standard sets security activities across the health software life cycle. Both the United States Food and Drug Administration (FDA) and European notified bodies recognize it as an authoritative standard.

IEC 81001-5-1 does not replace IEC 62304 software safety classification. Instead, it plugs directly into IEC 62304 development processes. Where IEC 62304 requires software requirements, IEC 81001-5-1 adds security requirements derived from threat analysis. Where IEC 62304 requires problem resolution, IEC 81001-5-1 adds vulnerability disclosure and security patch management.

Why is generic secure coding advice not enough for medical devices?

Generic web security guides often suggest basic hygiene: escape SQL queries, use HTTPS, and update packages regularly. While these practices are necessary, regulators and notified bodies reject submissions that rely only on generic rules.

Medical device regulations require systematic evidence rather than good intentions. You must prove that you analyzed threats systematically, implemented specific controls, and verified their effectiveness before release.

Security risks in medical devices differ from standard enterprise IT systems. In standard IT, a security breach usually means financial loss or data theft. In healthcare software, a security breach can cause physical harm to patients.

Consider an imaging pipeline that receives computed tomography (CT) scans. If an attacker tampers with image slice order or corrupts pixel data in transit, a diagnostic algorithm or radiologist could misdiagnose a patient. Under ISO 14971 risk management for imaging pipelines, that defect creates a critical clinical hazard.

Clause 5.3 of IEC 81001-5-1 requires structured threat modeling during architectural design. Frameworks like STRIDE (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege) identify how attackers might breach trust boundaries. You must document identified threats, trace them to design requirements, and prove that tests verify each defensive control.

How do you generate and maintain an SBOM for health software?

Modern medical software relies heavily on third-party libraries, container images, and open-source packages. Both MDCG 2019-16 and FDA premarket cybersecurity guidance require manufacturers to know every component in their software stack.

They enforce this through a Software Bill of Materials (SBOM). An SBOM is a machine-readable inventory of all third-party software, libraries, tools, and dependencies included in your product.

Under Section 524B of the United States Federal Food, Drug, and Cosmetic Act, an SBOM is mandatory for commercial clearance. MDCG 2019-16 expects equivalent transparency to demonstrate supply chain control.

A compliant SBOM must contain specific metadata for each component:

  • Component name and exact version string.
  • Primary supplier or author.
  • Cryptographic hash of the source code or binary.
  • Unique package identifiers, such as Package URLs (PURL).
  • Upstream dependency relationships.

Regulators require standard machine-readable formats. Two formats dominate the industry: CycloneDX and Software Package Data Exchange (SPDX). Both formats allow automated tools to parse your software inventory.

Do not treat an SBOM as a static document created once before submission. A compliant team automates SBOM generation inside their continuous integration (CI) pipeline. Every build generates an updated inventory alongside application artifacts.

Once generated, your pipeline checks the SBOM against public vulnerability catalogs. Tools scan your dependencies against Common Vulnerabilities and Exposures (CVE) feeds automatically. If a library contains a known vulnerability with an active exploit, your build flags the defect before deployment.

How does coordinated vulnerability disclosure work in practice?

Cybersecurity does not end when your software clears regulatory review. In fact, software vulnerabilities appear continuously throughout a product operating lifetime. Section 5 of MDCG 2019-16 and Clause 7 of IEC 81001-5-1 demand an active post-market surveillance system.

A core component of this system is Coordinated Vulnerability Disclosure (CVD). CVD provides a safe, structured way for security researchers, hospitals, and users to report security flaws to your engineering team.

A compliant CVD process requires 4 operational pieces:

  1. Public reporting intake: You must publish a clear security contact. Teams often place a security.txt file on their website or publish an email address such as security@example.com.
  2. Defined intake timelines: You must acknowledge receipt of vulnerability reports quickly, typically within 24 to 72 hours.
  3. Structured risk evaluation: When a researcher reports a vulnerability, evaluate its Common Vulnerability Scoring System (CVSS) score. Then assess whether the flaw affects patient safety under your ISO 14971 risk management file.
  4. Remediation and communication: You must deploy patches within defined timeframes. If a vulnerability exposes patients to immediate harm, you must notify hospital operators and competent authorities through formal vigilance reporting.

Many small teams hesitate to publish a security contact because they fear receiving invalid bug bounty reports. However, auditors look for an operational intake process during design reviews. A documented disclosure policy demonstrates regulatory maturity.

Managing network attack surfaces for hospital data pipelines

Connecting cloud medical software to hospital imaging networks introduces significant cybersecurity exposure. Medical modalities such as MRI and CT scanners send studies over the DICOM protocol. These transfers often run across the public internet to reach cloud archives or AI analysis tools.

Exposing custom network listeners directly to internet traffic expands your software attack surface. Legacy DICOM connections rely on raw TCP sockets, complex state machines, and long-lived sessions. Every exposed port, TLS configuration, and socket handler falls under your IEC 81001-5-1 threat model.

Securing these endpoints requires strict cryptographic controls. MDCG 2019-16 Section 3.6 calls for state-of-the-art transport encryption and strong endpoint authentication. For medical imaging connections, mutual TLS provides the standard defense. You can read our guide to mutual TLS for DICOM connections to understand how client certificates authenticate senders.

Maintaining custom DICOM listener infrastructure requires ongoing engineering overhead. Your team must track vulnerabilities across networking libraries, manage certificate renewals for each hospital scanner, and patch protocol parsers promptly.

A managed gateway such as Dicomly handles this network edge by terminating mutual TLS and translating DICOM traffic into standard HTTPS webhooks. This architecture keeps public internet DICOM listener infrastructure outside your core application codebase. In exchange, you pay a monthly service fee and lose raw TCP socket control. Your team must also authenticate and validate incoming payloads at your webhook receiver, as shown in how to receive deliveries in your backend.

If you partition your platform cleanly, you can separate transport utilities from your clinical software. As explained in our guide to MDCG 2019-11 software qualification and module boundaries, modular architecture lets you focus your regulatory cybersecurity documentation on the clinical device itself.

Where should your engineering team start today?

Meeting cybersecurity expectations under IEC 81001-5-1 and MDCG 2019-16 can seem overwhelming at first. However, the standards rely on systematic software engineering practices that fit modern development tools.

You can take 3 concrete steps today:

  1. Automate your SBOM in CI: Add an open-source tool like Syft or CycloneDX CLI to your build pipeline. Export a software bill of materials on every pull request.
  2. Publish your vulnerability disclosure policy: Add a security.txt file to your web domain. Define a monitored contact email and document your response timeline for security researchers.
  3. Draft a data flow diagram and threat model: Map where external network connections enter your system. Identify trust boundaries between hospital networks, public ingress endpoints, and clinical data stores.

Starting these practices early keeps compliance manageable. Documenting your supply chain and security lifecycle from day 1 allows your team to pass regulatory audits without slowing feature development.

Dicomly builds infrastructure for healthcare software teams that receive DICOM studies from hospitals. Why we built Dicomly →

Sources

  1. IEC 81001-5-1:2021 Health software and health IT systems safety, effectiveness and security (webstore.iec.ch)
  2. MDCG 2019-16 Rev.1 Guidance on Cybersecurity for medical devices (health.ec.europa.eu)
  3. FDA Recognized Consensus Standards: IEC 81001-5-1 Edition 1.0 (www.accessdata.fda.gov)
  4. FDA Guidance: Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions (www.fda.gov)

Start receiving DICOM today.

Early access. No credit card. First endpoint free.