Aquilion Exceed LB (TSX-202A/3) V10.6 with AiCE-I

The Aquilion Exceed LB (TSX-202A/3) V10.6 is a computed tomography system utilizing a Deep Convolutional Neural Network, AiCE-i, for advanced image reconstruction to enable whole-body imaging with significantly reduced image noise and radiation dose. Phantom-based testing demonstrates its science through improved image quality metrics, including enhanced low-contrast detectability and spatial resolution, particularly in specified anatomical applications.

The Aquilion Exceed LB (TSX-202A/3) V10.6 with AiCE-i is a whole-body multi-slice helical computed tomography (CT) scanner designed to acquire and display cross-sectional volumetric data of the entire body, including the head, for specialized diagnostic studies. Its core technology utilizes the PURE ViSION detector system and introduces the Advanced Intelligent Clear-IQ Engine (AiCE-i), an advanced reconstruction technology that employs Deep Convolutional Neural Network (DCNN) methods; this DCNN algorithm is engineered to differentiate anatomical structures from image noise, thereby facilitating fast, high-quality CT reconstructions across applications such as abdomen, pelvis, lung, extremities, head, and inner ear imaging. Key features of the Aquilion Exceed LB include a wide 900mm gantry bore. Performance evaluations using phantoms highlighted AiCE-i's capabilities, demonstrating up to 82% radiation dose reduction for abdomen scans compared to traditional Filtered Backprojection (FBP), and up to 87% improvement in low-contrast detectability with a 67.2% noise reduction for body imaging at equivalent dose levels; AiCE-i also showed improved high-contrast spatial resolution with simultaneous 50% noise reduction, while the PUREViSION Optics technology separately contributed to enhanced low-contrast detectability and dose reduction for body and brain CT applications.

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 performance of AiCE-i in the Aquilion Exceed LB compare to other DCNN-based reconstruction algorithms from different vendors in terms of specific image quality metrics (e.g., Noise Power Spectrum, Modulation Transfer Function) and computational time across various anatomical regions?; The document states AiCE-i is indicated for abdomen, pelvis, lung, extremities, head, and inner ear; what were the specific quantitative improvements in dose reduction and low-contrast detectability observed for each of these anatomical regions individually during phantom testing?; Given the larger 900mm gantry opening and increased couch lateral movement (±85mm) of the Aquilion Exceed LB, what specific clinical scenarios or patient populations (e.g., bariatric patients, radiotherapy planning) are anticipated to benefit most from these hardware modifications compared to systems with smaller bores?; While phantom studies demonstrate significant dose reduction with AiCE-i (up to 82%), what is the expected range of dose reduction and its impact on diagnostic confidence when applied to diverse clinical patient populations and varying protocols, particularly for traditionally challenging low-dose examinations?; Since the AiCE DCNN algorithm was not modified or retrained from the version used in the predicate device, are there plans for future retraining or optimization of the model using data acquired from the Aquilion Exceed LB to potentially further enhance its performance or expand its applications, for instance, to cardiac imaging which was listed for the predicate's AiCE?