(262 days)
This device is indicated to acquire and display cross-sectional volumes of the whole body, to include the head. The Aquilion Precision has the capability to provide volume sets. These volume sets can be used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician.
FIRST 3.0 is an iterative reconstruction algorithm intended to reduce exposure dose and improve high contrast spatial resolution for abdomen, pelvis, chest, cardiac, extremities and head applications.
AiCE is a noise reduction algorithm that improves image quality and reduces image noise by employing Deep Convolutional Neural Network methods for abdomen and pelvis applications.
Aquilion Precision (TSX-304A/1 and /2) V8.8 with AiCE is an ultra-high resolution whole body multi-slice helical CT scanner, consisting of a gantry, couch and a console used for data processing and display. Aquilion Precision incorporates a 160-row, 0.25 mm detector, a 5.7-MHU large-capacity tube, and 0.35 s scanning, enabling wide-range scanning with short scan times to capture cross sectional volume data sets used to perform specialized studies, using indicated software/hardware, by a trained and qualified physician. In addition, the subject device incorporates the latest reconstruction technology, AiCE (Advanced intelligent Clear-IQ Engine), intended to reduce image noise and improve image quality by utilizing Deep Convolutional Neural Network methods to 1024x1024 HR/SHR images. These methods can more fully explore the statistical properties of the signal and noise. By learning to differentiate structure from noise, the algorithm produces fast, high quality CT reconstruction.
Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criterion (Implicit) | Reported Device Performance and Confirmation |
|---|---|
| Image Quality (General) | - CT image quality metrics were performed: Contrast-to-Noise Ratios (CNR), CT Number Accuracy, Uniformity, Slice Sensitivity Profile (SSP), Modulation Transfer Function (MTF)-Wire, Modulation Transfer Function (MTF)-Edge, Standard Deviation of Noise (SD), Noise Power Spectra (NPS), Low Contrast Detectability (LCD), and Pediatric water phantom. |
- "AiCE is substantially equivalent to the predicate device as demonstrated by the results of the above testing."
- "The AiCE reconstructed images were of diagnostic quality," as reviewed by an American Board Certified Radiologist. |
| Low Contrast Detectability (LCD) Improvement | - "AiCE demonstrated 13% improved low contrast detectability" for super-high resolution body at the same dose compared to AIDR 3D. - "superior LCD performance for super-high resolution body at the same dose for AiCE vs AIDR 3D." |
| Noise Reduction | - "AiCE demonstrated... 42% noise reduction for super-high resolution body at the same dose compared to AIDR 3D." |
| Dose Neutrality | - "it was demonstrated that there is dose neutrality between super-high resolution mode with AiCE and normal resolution mode with AIDR." |
| Spatial Resolution Improvement (High Contrast) | - "A spatial resolution comparison study was conducted to support a high contrast spatial resolution improvement claim of 8.8 lp/cm at 10% of the MTF for AiCE relative to AIDR 3D Standard for abdomen/body." |
| Diagnostic Quality of AiCE Images (Clinical) | - "Representative abdomen/pelvis diagnostic images, reviewed by an American Board Certified Radiologist, were obtained using the subject device and it was confirmed that the AiCE reconstructed images were of diagnostic quality." |
| Safety and Regulatory Compliance | - "The device is designed and manufactured under the Quality System Regulations as outlined in 21 CFR § 820 and ISO 13485 Standards." - Conforms to applicable parts of IEC60601-1, IEC60601-1-2, IEC60601-1-6, IEC60601-2-28, IEC60601-2-44, IEC60825-1, IEC62304, IEC62366, NEMA PS 3.1-3.18, NEMA XR-25, NEMA XR-26 and NEMA XR-29.
- Complies with all applicable requirements of the radiation safety performance standards, as outlined in 21 CFR §1010 and §1020.
- Software Documentation for a Moderate Level of Concern included per FDA guidance.
- Cybersecurity documentation included per FDA guidance.
- Testing conducted in accordance with applicable IEC standards. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Test Set: The document mentions "Representative abdomen/pelvis diagnostic images" but does not specify the numerical sample size for this clinical review. For phantom studies, the concept of a "sample size" of images is less directly applicable; rather, it's about the number of measurements taken from the phantom.
- Data Provenance: The document does not explicitly state the country of origin for the data. Given the manufacturer is Canon Medical Systems Corporation based in Japan, it's possible the data originated there or from a multi-national study. The document also does not specify if the data was retrospective or prospective. The use of "representative diagnostic images" suggests real-world data, but the study design (retrospective vs. prospective) is not detailed.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
- Number of Experts: "an American Board Certified Radiologist" (singular).
- Qualifications of Experts: American Board Certified Radiologist. No specific number of years of experience is mentioned.
4. Adjudication Method for the Test Set
- The document describes a review by a single American Board Certified Radiologist. Therefore, there was no adjudication method described for resolving disagreements, as only one expert was involved in this specific clinical image evaluation aspect. The primary testing relies on objective phantom measurements.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size
- No, an MRMC comparative effectiveness study was not explicitly described. The document mentions a single radiologist's review for diagnostic quality confirmation.
- The study primarily focuses on standalone performance improvements demonstrated through objective phantom measurements (noise reduction, LCD, spatial resolution) and the radiologist's qualitative assessment of diagnostic quality. It does not compare human readers with AI assistance versus human readers without AI assistance to quantify an "effect size" of improvement in human performance.
6. If a Standalone (i.e., Algorithm Only Without Human-In-The-Loop Performance) Was Done
- Yes, a standalone performance evaluation was done. The phantom studies (CNR, CT Number Accuracy, Uniformity, SSP, MTF, SD, NPS, LCD, Pediatric water phantom) directly measure the performance characteristics of the AiCE algorithm as applied to CT data, without human interaction in the measurement process.
- The reported improvements in low contrast detectability (13%), noise reduction (42%), and spatial resolution (8.8 lp/cm) are all measures of the algorithm's standalone performance.
7. The Type of Ground Truth Used
- For the phantom studies, the "ground truth" is established by the known physical properties and measurements derived from the phantom itself, and standardized metrology for CT image quality. This is an objective, quantitative ground truth.
- For the clinical image evaluation, the ground truth for "diagnostic quality" was established by expert consensus (of one expert), specifically an American Board Certified Radiologist.
8. The Sample Size for the Training Set
- The document does not provide any information about the sample size used for the training set of the Deep Convolutional Neural Network (DCNN) for AiCE. This information is typically proprietary and not included in 510(k) summaries unless specifically requested or deemed critical for substantial equivalence in a novel device.
9. How the Ground Truth for the Training Set Was Established
- The document states that AiCE "improves image quality and reduces image noise by employing Deep Convolutional Neural Network methods...By learning to differentiate structure from noise, the algorithm produces fast, high quality CT reconstruction." However, the document does not describe how the ground truth was established for the training data used to train this Deep Convolutional Neural Network. This is a critical piece of information for AI/ML-based devices but is not detailed in this 510(k) summary.
§ 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.