NeuViz 128 Multi-Slice CT Scanner System

The NeuViz 128 Multi-Slice CT Scanner System is designed to provide whole-body computed tomography imaging by acquiring X-ray transmission data with a continuously rotating X-ray tube and detector assembly, subsequently reconstructing cross-sectional images. Its key scientific advancement involves the integration of the ClearInfinity deep learning-based reconstruction algorithm, complementing existing Filtered Back-Projection and ClearView iterative methods, to significantly enhance image quality metrics such as low-contrast detectability and spatial resolution, while simultaneously reducing image noise and enabling substantial radiation dose reduction.

The NeuViz 128 Multi-Slice CT Scanner System is a whole-body computed tomography X-ray system designed for acquiring cross-sectional images. It operates using a continuously rotating X-ray tube and detector array, employing the fan beam principle to collect X-ray transmission data, which is then computer-reconstructed. This modified version introduces several technological enhancements, most notably a new optional image reconstruction algorithm called ClearInfinity, which utilizes a trained deep learning neural network. This is offered alongside existing methods like Filtered Back-Projection (FBP) and the ClearView iterative reconstruction algorithm. The system also features an extended generator mA range (10mA-667mA), new couch options including an intelligent positioning device, an optional extended 700mm FOV, a wider spiral scan pitch range, and additional optional software functionalities such as real-time MPR, 4D scanning, dual-energy imaging (Prism Imaging), enhanced metal artifact reduction (MAR+), arrhythmia handling, and Auto FOV+. Non-clinical performance testing demonstrated significant improvements with the new ClearInfinity reconstruction compared to FBP. Phantom studies indicated potential for substantial radiation dose reduction (50-60% for head, 75-85% for body) while maintaining image quality, and notable enhancements in low-contrast detectability (35-65% for head, 60-115% for body) and noise reduction (up to 90%) at the same dose; spatial resolution was also reported to improve up to 88% at equivalent image noise. The ClearView iterative reconstruction method also showed benefits over FBP, including dose reduction (40-50% head, 45-60% body) and improved low-contrast detectability (30-35% head, 35-45% body). Retrospective clinical image evaluations by radiologists found ClearInfinity to be equivalent or better than FBP in diagnostic quality, including at simulated low doses. However, both ClearView and ClearInfinity, as well as the Auto FOV+ feature for children under three, have not been thoroughly tested for all advanced post-processing applications or across all pediatric populations, necessitating consultation with radiologists and physicists for dose optimization and appropriate 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 performance of the ClearInfinity deep learning algorithm compare to other commercially available deep learning reconstruction algorithms in terms of specific image quality metrics (e.g., CNR, SNR, spatial resolution) across different anatomical regions and patient sizes?; What are the computational requirements (e.g., reconstruction time, hardware dependencies) for ClearInfinity, and how do these compare to FBP and ClearView, particularly in a high-throughput clinical setting?; Given the noted limitations in pediatric testing for ClearInfinity and Auto FOV+, what specific phantom studies or pre-clinical evaluations are planned or recommended to establish safe and effective imaging protocols for different pediatric age groups and clinical indications?; The document mentions "US patient-based testing data" for ClearInfinity; what was the demographic and clinical diversity of this patient cohort, and were there any specific challenges or image quality variations observed across these cases?; For the new optional features like "Dual Energy Imaging (Prism Imaging)" and "4D scanning," what specific phantom and pre-clinical validation was performed to demonstrate their accuracy and clinical utility beyond basic functionality?