AI-Rad Companion Engine

The AI-Rad Companion Engine is an advanced software platform designed to enhance diagnostic capabilities by providing visualization and enabling modular post-processing extensions for medical images using DICOM standards. Its methodology focuses on integrating various clinical applications and improving user workflow, supported by rigorous non-clinical testing to ensure performance and safety.

The AI-Rad Companion Engine is a software platform developed by Siemens Medical Solutions, designed to assist technicians and trained physicians in the diagnostic use of medical images. Its primary purpose is to provide basic image visualization and enable external post-processing extensions for multi-modality and multi-vendor DICOM data. The platform facilitates qualitative and quantitative measurement and analysis of clinical data, supports data storage, and allows transfer to other systems like PACS. Key functionalities include an interface for DICOM data input/output, data validity checks using DICOM tags, an interface for post-processing extensions, and user interfaces for image visualization and system configuration. The system receives imaging data from modalities or PACS via auto-routing, with data transfer managed by the "teamplay" infrastructure. The AI-Rad Companion Engine, specifically version VA20, builds upon a previously established platform. Its technological foundation relies on DICOM communication standards for image handling and interoperability. Version VA20 introduces enhancements such as support for additional clinical extensions and an improved workflow for better usability. The system's reliability and performance were evaluated through non-clinical bench testing, including software validation (Unit, System, Integration testing) and human factors usability validation, demonstrating conformance with multiple industry standards like IEC 62366-1, ISO 14971, IEC 62304, and DICOM. While no new clinical tests were conducted for this version, extensive verification and validation activities were performed to ensure the enhancements maintain an equivalent safety and performance profile to its predicate. The platform also incorporates cybersecurity measures according to relevant guidelines.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific types of "additional clinical extensions" are supported by version VA20, and what new diagnostic capabilities do they offer compared to previous versions?; How does the "teamplay" infrastructure ensure secure and efficient data transfer, particularly for large imaging datasets, and what are its interoperability features with non-Siemens systems?; Could you elaborate on the "modified workflow" in VA20 and provide specific examples of how it enhances usability for technicians and physicians in a clinical setting?; What were the key findings from the human factors usability validation, and how did they influence the final design of the user interface and interaction with clinical extensions?; Beyond DICOM conformance, what specific measures are implemented within the AI-Rad Companion Engine to ensure data integrity and validity from multi-vendor imaging sources before post-processing?