Aquilion ONE (TSX-308A/3) V1.4 with PIQE Reconstruction System

The Aquilion ONE (TSX-308A/3) V1.4 system aims to advance computed tomography diagnostics through its PIQE deep learning reconstruction, which enhances spatial resolution and reduces noise in cardiac, abdominal, and pelvic imaging, complementing iterative (FIRST) and AI-based (AiCE) reconstruction methods. Its methodology combines whole-body volumetric scanning with spectral imaging for material differentiation and features designed for radiation dose optimization, with performance characteristics established via comprehensive bench testing and clinical image assessments.

The Aquilion ONE (TSX-308A/3) V1.4 is a whole-body multi-slice helical CT scanner designed to acquire and display cross-sectional volumetric data. It can image whole organs such as the brain, heart, and pancreas in a single rotation, providing volume sets for specialized studies. This iteration introduces the PIQE Reconstruction System, a deep learning method enhancing spatial resolution and reducing noise for cardiac, abdominal, and pelvic applications; crucially, original image data remains accessible. The system also integrates FIRST, an iterative reconstruction algorithm for dose reduction and improved high-contrast spatial resolution, and AiCE, utilizing Deep Convolutional Neural Networks for noise reduction across various applications including the inner ear. A spectral imaging system allows dual-energy acquisition via rapid kV switching for material composition analysis. Performance assessments, including phantom studies and clinical image reviews, support the device's capabilities. Bench tests indicated image quality metrics (e.g., contrast-to-noise ratio, CT number accuracy, Modulation Transfer Function) were equivalent or improved over a predicate device. Specific low contrast detectability was reported, such as 2 mm at 0.3% with 14.7 mGy CTDIvol using AiCE, and the SilverBeam Filter was noted for dose reduction. Radiologists confirmed the diagnostic quality of images from various body regions. The system also incorporates hardware updates from its predicate, including a larger 800mm gantry aperture, faster couch speed, anatomical landmark detection, and higher resolution display.

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 PIQE deep learning reconstruction algorithm's performance in improving spatial resolution and reducing noise for cardiac, abdomen, and pelvis applications compare quantitatively against other leading deep learning methods, particularly concerning subtle lesion conspicuity?; What is the specific impact of the new 'Anatomical Landmark Detection' feature on workflow efficiency, scan time, and inter-operator variability in complex CT examinations?; The PIQE system offers L1, L2, and L3 intensity levels. What are the precise trade-offs between these levels in terms of quantitative noise reduction, spatial resolution gains (e.g., MTF), and the potential introduction of image artifacts or loss of low-contrast information across different anatomical regions?; Considering the spectral imaging system's capability for material differentiation, what is the demonstrable improvement in diagnostic accuracy or diagnostic confidence for specific clinical applications, such as differentiating incidentalomas or characterizing plaque, compared to conventional CT acquisitions or the predicate's spectral capabilities?; How do the synergistic effects of PIQE, AiCE, FIRST, and the SilverBeam filter quantitatively impact the overall radiation dose reduction achievable for standardized whole-body protocols while maintaining or improving specific diagnostic image quality metrics like low-contrast detectability, especially in dose-sensitive patient populations?