(365 days)
SAFIRE is a CT reconstruction software. The end user can choose to apply either SAFIRE or the weighted filter back-projection (WFBP) to the acquired raw data.
Depending on the clinical task, patient size, anatomical location, and clinical practice, the use of SAFIRE can help to reduce radiation dose while maintaining Pixel noise, low contrast detectability and high contrast resolution. Phantom measurements showed that high contrast resolution and pixel noise are equivalent between full dose WFBP images and reduced dose SAFIRE images. Additionally, SAFIRE can reduce spiral artifacts by using iterations going back and forth between image space and raw data space.
A Model Observer evaluation showed that equivalent low contrast detectability can be achieved with 54% to 60% less dose using SAFIRE at highest noise reduction strength for thin (0.6 mm) reconstruction slices in simulated body and head phantoms for low contrast objects with different contrasts.
Images reconstructed with SAFIRE are not intended to be evaluated with syngo Osteo CT or syngo Calcium Scoring.
Siemens SAFIRE is a software option to for CT operating system SOMARIS/7. It provides an improved image quality. Reciprocally the use SAFIRE allows the physician to acquire scans with reduced radiation dose without reduction of image quality compared to today's standard.
Here's a breakdown of the acceptance criteria and study information for the SAFIRE device, based on the provided text:
SAFIRE Device Performance and Acceptance Criteria
The SAFIRE software is designed to improve image quality and enable radiation dose reduction in CT scans. The acceptance criteria focus on maintaining or improving image quality metrics (pixel noise, low contrast detectability, high contrast resolution) while allowing for significant dose reduction.
Table of Acceptance Criteria and Reported Device Performance:
| Performance Metric | Acceptance Criteria (Stated Goal) | Reported Device Performance |
|---|---|---|
| Radiation Dose Reduction | Reduce radiation dose without reduction of image quality compared to today's standard. | Achieved 54% to 60% less dose for equivalent low contrast detectability using SAFIRE at highest noise reduction strength (for thin (0.6 mm) reconstruction slices in simulated body and head phantoms for low contrast objects). Clinical environment evaluation confirmed use of up to 60% less radiation dose with SAFIRE provided images with same results in terms of pixel noise and spatial resolution compared to today's standard scan techniques. |
| Pixel Noise | Maintain Pixel noise with reduced radiation dose. | Phantom measurements showed pixel noise in images is equivalent for full dose WFBP images and SAFIRE images acquired with reduced dose. Clinical evaluation confirmed same results in terms of pixel noise with reduced dose and SAFIRE. |
| Low Contrast Detectability | Maintain low contrast detectability with reduced radiation dose. | Model Observer evaluation showed equivalent low contrast detectability can be achieved with 54% to 60% less dose using SAFIRE at highest noise reduction strength. |
| High Contrast Resolution | Maintain high contrast resolution with reduced radiation dose. | Phantom measurements showed high contrast resolution in images is equivalent for full dose WFBP images and SAFIRE images acquired with reduced dose. |
| Spiral Artifact Reduction | Reduce spiral artifacts. | SAFIRE can reduce spiral artifacts by using iterations. Validated in clinical images and simulations with established image quality phantoms. |
Study Details:
Based on the provided text, the verification and validation efforts included non-clinical tests, a Model Observer evaluation, phantom measurements, and clinical environment evaluations.
-
Sample Size Used for the Test Set and Data Provenance:
- Test Set Sample Size: The text does not explicitly state a numerical sample size for the "test set" in terms of number of patient cases or images. It mentions "simulated body and head phantoms" for the Model Observer evaluation and "clinical images" for artifact reduction.
- Data Provenance: The data provenance is mixed:
- Phantoms: Simulated body and head phantoms.
- Clinical Data: "Clinical environment" and "clinical images," suggesting prospective or retrospective data from a healthcare setting, likely within Germany (manufacturing site SIEMENS AG Healthcare Sector Siemensstrasse 1 D-91301 Forchheim) or where Siemens conducts clinical studies. The specific country of origin is not detailed.
-
Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:
- The document does not provide information regarding the number of experts or their qualifications used to establish ground truth for any clinical test set. The Model Observer evaluation uses a quantitative, objective metric (low contrast detectability) and phantoms, which typically do not involve human experts for ground truth establishment in the same way clinical image interpretation would.
-
Adjudication Method for the Test Set:
- The document does not specify any adjudication method (e.g., 2+1, 3+1, none) for the test set.
-
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No, an MRMC comparative effectiveness study is not explicitly mentioned. The document focuses on demonstrating that SAFIRE allows for dose reduction while maintaining image quality metrics (pixel noise, low contrast detectability, high contrast resolution) through phantom studies and some clinical evaluation of pixel noise/spatial resolution. It doesn't describe a study comparing human reader performance with and without AI assistance to quantify an "effect size" of improvement.
-
Standalone Performance (Algorithm only without human-in-the-loop performance):
- Yes, a standalone performance evaluation was done. The Model Observer evaluation and the phantom measurements for pixel noise, low contrast detectability, and high contrast resolution are examples of standalone algorithm performance. These evaluations assess the algorithm's direct impact on image characteristics without involving human interpretation until later clinical validation steps.
-
Type of Ground Truth Used:
- The ground truth used appears to be primarily established physical properties and objective measurements from phantoms (e.g., known contrasts in phantoms for low contrast detectability) and computational models (Model Observer).
- For "spiral artifacts," the ground truth was also established through "simulations with established image quality phantoms" and "clinical images," indicating a qualitative assessment potentially against a reference or expert understanding of artifact presence.
-
Sample Size for the Training Set:
- The document does not provide any information about the sample size used for the training set. SAFIRE is "iterative reconstruction" software, and while it learns and refines, the text doesn't detail a traditional machine learning "training set" in the sense of a distinct dataset used for supervised learning.
-
How the Ground Truth for the Training Set Was Established:
- As no information on a distinct "training set" is provided, there is no information on how its ground truth was established. Iterative reconstruction algorithms like SAFIRE often use mathematical models of physics and image formation, rather than relying on labeled training data with external ground truth in the same way many modern AI systems do.
§ 892.1750 Computed tomography x-ray system.
(a)
Identification. A computed tomography x-ray system is a diagnostic x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data from the same axial plane taken at different angles. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.