Multi-Vendor Image Quality Comparison of Precise Image

This white paper presents a detailed comparison of the image quality of the Philips Precise Image deep-learning reconstruction (DLR) technique against traditional computed tomography (CT) methods and other manufacturers' DLR techniques. The document outlines how Precise Image addresses the limitations of conventional filtered back projection (FBP) and iterative reconstruction (IR) methods, providing low-noise images with a familiar appearance. It quantifies improvements in image quality using advanced metrics such as the detectability index, which assesses the ability to detect lesions based on noise, resolution, and specific characteristics. The methodology employed includes the use of an ACR 464 phantom to evaluate noise and resolution properties, with results indicating significant enhancements in image quality, particularly in detecting low-contrast chest nodules. The findings suggest that Precise Image not only surpasses traditional methods but also outperforms competitors in the AI reconstruction domain, thereby enhancing diagnostic confidence and patient outcomes.