Search Results

This computed tomography system is intended to generate and process cross-sectional images by computer reconstruction of x-ray transmission data within a 25 cm field-of-view, primarily for the head and neck.

The images delivered by SOMATOM On.scene can be used by a trained physician as an aid in diagnosis.

The Siemens SOMATOM On scanners are comprised of a Computed Tomography (CT) Scanner System (SOMATOM On.scene) which can be mounted on an optional motorized base (SOMATOM On.site). The CT scanner features one continuously rotating tube-detector system that functions according to the fan beam principle. The system software is a command-based program used for patient management, data management, X-ray scan control, image reconstruction, and image archive/evaluation.

The SOMATOM On scanners produce CT images in DICOM format, which can be used by trained staff for post-processing applications commercially distributed by Siemens and other vendors as an aid in diagnosis and treatment preparation. The computer system included in the CT Scanner is able to run optional post processing applications.

The software version for the SOMATOM On scanner system is Somaris/10 syngo CT VA35A, is a command-based program used for patient management, X-ray scan control, image reconstruction, and image archive/evaluation. The software platform SOMARIS/10 syngo CT VA35A is designed to provide a plugin interface to integrate potential advanced post processing tasks, tools, or extendable functionalities.

As with the primary predicate device, the SOMATOM On. Scanners will be available in a 32 row, 32 slice configuration.

This FDA 510(k) submission for the SOMATOM On.site and On.scene CT systems does not contain specific acceptance criteria or a study directly aimed at proving algorithm performance against a pre-defined set of metrics. Instead, the submission focuses on demonstrating substantial equivalence to predicate devices primarily through non-clinical testing, phantom studies, and adherence to various industry standards and guidance documents.

Here's an analysis based on the provided text, addressing your points where information is available:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of acceptance criteria for specific quantitative performance metrics of an AI algorithm, nor does it report the device performance against such criteria. The "Performance Data" section discusses:

• Non-Clinical Testing (Integration and Functional): Included phantom tests and volunteer human scans.
• Electrical Safety and EMC Testing: Conformance to IEC 60601-1, 60601-2-44, and 60601-1-2 standards.
• Radiation Safety Testing: "Under published limits," with the protection curtain reducing radiation significantly.
• Usability Testing: Formative and Summative evaluations, identifying "No new use errors, hazards or hazardous situations."
• Imaging Studies: Phantom scans (adult heads, pediatric bodies, pediatric heads) and volunteer scans (brain with/without contrast, ankle, hand). All imaging results were "as expected and were determined by a board-certified radiologist to be of high diagnostic quality." This last point is the closest to a performance statement, but it's qualitative and without specific acceptance thresholds.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: Not explicitly stated. The document mentions "phantom tests and volunteer human scans" for non-clinical testing and "phantom scans... Additionally, volunteer scans" for imaging studies. There is no quantification of the number of volunteers or specific phantom cases used for performance evaluation that could be considered a "test set" for an AI algorithm.
• Data Provenance: Not explicitly stated. The document doesn't specify the country of origin for the volunteer scans or if the data was retrospective or prospective. Given that this is a Siemens product with manufacturing in Germany and sales in the US, the volunteer data could be from either region.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

• Number of Experts: For the imaging studies, it states: "All imaging results were as expected and were determined by a board-certified radiologist to be of high diagnostic quality." This implies at least one board-certified radiologist provided an opinion.
• Qualifications: "board-certified radiologist." No specific experience level (e.g., "10 years of experience") is mentioned.

4. Adjudication Method for the Test Set

The document does not describe any formal adjudication method (e.g., 2+1, 3+1) for establishing ground truth from multiple experts. The statement "determined by a board-certified radiologist" suggests a single expert assessment for the imaging quality.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, a multi-reader multi-case (MRMC) comparative effectiveness study focusing on how human readers improve with AI vs. without AI assistance was not mentioned or performed. The submission is for the CT system itself, not an AI-powered diagnostic aid that assists human readers.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

The document describes the CT system as providing images that "can be used by a trained physician as an aid in diagnosis." It also states "The computer system included in the CT Scanner is able to run optional post processing applications." While the software platform (SOMARIS/10 syngo CT VA35A) is designed to "provide a plugin interface to integrate potential advanced post processing tasks, tools, or extendable functionalities," the submission does not describe the performance of any specific standalone AI algorithm (i.e., operating without human-in-the-loop for diagnosis). The focus is on the CT system's ability to produce diagnostically acceptable images.

7. The Type of Ground Truth Used

For the imaging studies described, the ground truth was based on expert consensus/opinion (specifically, a "board-certified radiologist" determining "high diagnostic quality"). There is no mention of pathology or outcomes data being used as ground truth for the device's image quality evaluation.

8. The Sample Size for the Training Set

The document does not provide any information regarding a training set sample size. This is consistent with the submission not detailing the performance of a specific AI algorithm for diagnostic aid, but rather the CT system's foundational image generation capabilities. The software updates mentioned are primarily for mobile workflow adaptation and hardware/reconstruction support, not a defined AI model.

9. How the Ground Truth for the Training Set Was Established

Since no training set is discussed, there is no information on how ground truth was established for it.

Ask a specific question about this device