Aquilion Serve (TSX-307A/1) V1.2 with AiCE-i

The Aquilion Serve (TSX-307A/1) V1.2 is an advanced whole-body CT scanner employing AiCE, a Deep Convolutional Neural Network-based image reconstruction technology, to deliver superior image quality by reducing noise, enhancing low-contrast detectability, and improving spatial resolution. Its methodology combines helical scanning with sophisticated data processing and new hardware features, validated through rigorous testing that demonstrated significant improvements in image diagnostics and potential for optimized radiation dose.

The Aquilion Serve (TSX-307A/1) V1.2 is a whole-body multi-slice helical CT scanner designed to acquire and display cross-sectional volume data sets for a wide range of anatomical regions, including the head, facilitating specialized diagnostic studies. Its core technology includes the AiCE (Advanced intelligent Clear-IQ Engine), a noise reduction algorithm that utilizes Deep Convolutional Neural Network methods to enhance image quality and decrease image noise in applications such as abdomen, pelvis, lung, cardiac, extremities, head, and inner ear imaging. The system consists of a gantry, couch, and a console for data processing and display, and incorporates several technological advancements including a 3D Landmark Scan for positioning, Anatomical Landmark Detection, scan planning assistance, Flex e-Tilt for gantry movement, a larger 800mm gantry opening, an additional SilverBeam wedge filter, internal gantry cameras, and a NewUX console with a 27-inch high-resolution display. Extensive bench and phantom testing validated the Aquilion Serve's performance, particularly the AiCE technology, which demonstrated equivalent or improved image quality metrics compared to its predicate device and traditional Filtered Back Projection (FBP) or Adaptive Iterative Dose Reduction 3D (AIDR 3D). Key findings from testing AiCE include improved low-contrast detectability, significant noise reduction, enhanced quantitative spatial resolution for body and cardiac imaging, and potential for radiation dose reduction, especially in abdomen scans compared to FBP. Functionality of new features like automatic scan planning and Flex e-Tilt was confirmed, and representative clinical images reviewed by a board-certified radiologist were found to be of diagnostic quality, supporting the system's effectiveness for its intended use.

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. How does the 'Anatomical Landmark Detection' and 'Scan planning assist function' specifically improve workflow efficiency or scan accuracy compared to systems without these features?; What are the quantitative differences in radiation dose reduction achieved with AiCE for various anatomical regions (e.g., head, cardiac, abdomen) when compared to AIDR 3D and FBP at comparable diagnostic image quality?; Can the 'Flex e-Tilt' feature be utilized during dynamic scanning sequences, and how does it impact image acquisition speed or potential for motion artifacts?; What is the impact of the 'NewUX Console' and larger, higher-resolution display on user experience, particularly in terms of image review, manipulation capabilities, and the effect of simplified ROI/windowing options?; The AiCE technology claims 'noise texture as natural as filtered backprojection.' What specific metrics or qualitative assessments were used to substantiate this claim beyond NPS and kurtosis values, and how does it compare to other MBIR techniques?