A critical security vulnerability has been disclosed in the widely deployed pynetdicom medical imaging library, exposing healthcare systems globally to unauthenticated file write attacks. With a CVSS v3 score of 9.1, this flaw ranks among the most severe vulnerabilities disclosed in medical-sector open-source software this year — and the attack surface is significant.
What Is pynetdicom and Why Does It Matter?
pynetdicom is a Python implementation of the DICOM networking protocol, a foundational standard used across hospitals, radiology departments, imaging centers, and clinical research facilities worldwide. The library underpins how medical imaging data — X-rays, MRIs, CT scans — is transmitted between systems. Its parent project, pydicom, serves as a cornerstone of open-source medical imaging workflows, meaning the library is embedded in countless healthcare IT environments, often without direct awareness by security teams.
That ubiquity is precisely what makes this vulnerability so consequential. An attacker does not need credentials to exploit it. They simply need network access to a system running a vulnerable instance.
The Vulnerability: A Path Traversal Hidden in DICOM Handling
The flaw — tracked as CVE-2026-56445 — resides in the qrscp application's C-STORE request handler. In DICOM terminology, a C-STORE operation is how imaging data is pushed from one system to another. The handler in question accepts DICOM datasets from external sources and then passes a specific instance value directly into Python's os.path.join() function.
The critical failure here is the absence of input sanitization before the attacker-supplied value is incorporated into a filesystem path — a textbook path traversal condition that allows directory boundary escapes.
An attacker can craft a malicious DICOM dataset containing a manipulated instance value — using sequences like ../../ — to navigate outside the intended storage directory and write files to arbitrary locations on the host filesystem. Depending on the permissions of the running process, this could enable an attacker to overwrite system files, plant web shells, drop malicious executables, or corrupt critical application data.
Affected Versions and Scope
All versions of pynetdicom from v1.0.0 through v3.0.3 are confirmed vulnerable. The fix was introduced in v3.0.4. Organizations operating any deployment of pynetdicom below this threshold — whether in production clinical environments, research pipelines, or development infrastructure — should treat this as an urgent remediation priority.
Why This Is Especially Dangerous in Healthcare Environments
Healthcare networks present unique challenges for vulnerability response. Medical imaging systems are frequently embedded in long-running clinical workflows where downtime carries real patient safety implications. This operational pressure often delays patching. Moreover, DICOM services are sometimes exposed on internal networks with limited segmentation, meaning a single compromised entry point can propagate laterally with relative ease.
The unauthenticated nature of this exploit compounds the risk considerably. Unlike vulnerabilities requiring prior credential compromise, this flaw can be triggered by any actor with network visibility to the DICOM service — which, in many hospital environments, includes a broad swath of internal network users and, in misconfigured deployments, potentially external actors as well.
The prospect of arbitrary file writes in a medical context is not merely an IT concern. Tampered configuration files, planted scripts, or overwritten application binaries could directly affect the availability or integrity of clinical systems — with downstream implications for patient care delivery.
What You Should Do
- Patch immediately: Upgrade pynetdicom to version 3.0.4 or later. This is the vendor-confirmed fix and should be deployed as a priority in all environments.
- Audit your dependencies: Conduct a software composition analysis across your infrastructure to identify all deployments of pynetdicom, including transitive dependencies embedded in third-party tools or imaging platforms.
- Restrict DICOM service exposure: Ensure DICOM services are not reachable from untrusted network segments. Apply strict firewall rules and network segmentation to limit who can initiate C-STORE requests.
- Review filesystem permissions: Ensure pynetdicom processes run with the minimum necessary OS privileges. Least-privilege configurations limit the blast radius of a successful path traversal exploit.
- Monitor for anomalous file writes: Deploy file integrity monitoring on systems running DICOM services. Unexpected writes to directories outside configured storage paths should trigger immediate investigation.
- Engage your vendor ecosystem: If your imaging platform or PACS solution incorporates pynetdicom, contact the vendor to confirm their patch status and expected update timeline.
The Broader Lesson
This vulnerability is a reminder that open-source libraries powering critical healthcare infrastructure deserve the same security scrutiny applied to commercial software — and in many cases, they receive far less. Path traversal is one of the oldest and most well-understood vulnerability classes in software security. Its presence in a library handling sensitive medical data underscores the need for security-aware code review practices and robust dependency management within healthcare IT procurement and operations.
For security teams in healthcare organizations: this is not an advisory to file away. It demands immediate action.
