Vulnerabilities

Orthanc DICOM Vulnerabilities Lead to Crashes, RCE

Attackers could exploit these vulnerabilities in denial-of-service, information disclosure, and arbitrary code execution attacks.

Orthanc DICOM medical imaging vulnerability

Nine vulnerabilities in the open source Digital Imaging and Communications in Medicine (DICOM) server Orthanc allow attackers to crash servers, leak data, and execute arbitrary code remotely.

A lightweight standalone DICOM server for healthcare and medical research, Orthanc supports the automated analysis of medical images and does not require complex database administration or third-party dependencies.

The nine security defects in Orthanc, tracked CVE-2026-5437 to CVE-2026-5445, are rooted in insufficient validation of metadata, missing checks, and unsafe arithmetic operations, CERT Coordination Center (CERT/CC) notes in an advisory.

The first bug is an out-of-bounds read issue affecting the meta-header parser, caused by insufficient input validation in the parsing logic.

Next is a GZIP decompression bomb flaw in the processing of specific HTTP requests. Because no limit is enforced on decompressed size, and memory is allocated based on attacker-controlled metadata, a malicious payload could be used to exhaust system memory.

Another memory exhaustion defect was discovered in ZIP archive processing, where the server trusts metadata describing the uncompressed size of the archived files, allowing an attacker to forge size values and cause the server to allocate extremely large buffers during extraction.

Advertisement. Scroll to continue reading.

The HTTP server was also found to allocate memory directly based on user-supplied header values, allowing attackers to craft an HTTP request containing an extremely large length value, triggering server termination.

Orthanc’s decompression routine for the proprietary Philips Compression format is affected by an out-of-bounds read vulnerability, where escape markers at the end of the compressed data stream are improperly validated.

“A crafted sequence at the end of the buffer can cause the decoder to read beyond the allocated memory region and leak heap data into the rendered image output,” the CERT/CC advisory reads.

Another out-of-bounds read weakness was identified in the lookup-table decoding logic for Palette Color images, which fails to validate pixel indices. The flaw can be exploited via crafted images with indices larger than the palette size.

The last three security defects are heap buffer overflow issues impacting the image decoder, Palette Color image decoding logic, and PAM image parsing logic. Successful exploitation of these vulnerabilities could lead to out-of-bounds memory access.

“The most severe issues are heap-based buffer overflows in image parsing and decoding logic, which can crash the Orthanc process and may, under certain conditions, provide a pathway to remote code execution (RCE),” the CERT/CC advisory reads.

Orthanc versions 1.12.10 and earlier are affected by these bugs. Users are advised to update to version 1.12.11, which addresses all of them.

The vulnerabilities were discovered by researchers at Machine Spirits, who published their own advisories

Related: Critical Marimo Flaw Exploited Hours After Public Disclosure

Related: Microsoft Finds Vulnerability Exposing Millions of Android Crypto Wallet Users

Related: Data Leakage Vulnerability Patched in OpenSSL

Related: RCE Bug Lurked in Apache ActiveMQ Classic for 13 Years

Related Content

Mobile & Wireless

The chain involved the exploitation of several vulnerabilities in the Samsung Members and Samsung Account applications.

Vulnerabilities

The flaws can be exploited for remote code execution, authentication bypass, and EncryptInterceptor bypass.

Network Security

Forescout researchers have found 15 new vulnerabilities in the TP-Link Omada networking ecosystem.

Artificial Intelligence

A crafted prompt to a low-privilege Google ADK agent could be used to pass a malicious hand-off comment to a privileged agent.

Vulnerabilities

Over 24,000 internet-accessible server-management interfaces disclose authentication hashes before login.

Vulnerabilities

The biggest single reward paid out by Microsoft between July 1, 2025, and June 30, 2026, was $200,000.

Vulnerabilities

The N‑central vulnerability CVE-2026-18577 has been exploited in the wild after threat actors found a patch bypass.

Ransomware

The INC Ransomware gang has been targeting vulnerable SMA1000 appliances for root access and lateral movement.

Copyright © 2026 SecurityWeek ®, a Wired Business Media Publication. All Rights Reserved.

Exit mobile version