CardIQ Suite

CardIQ Suite is a non-invasive software application intended to analyze CT cardiovascular anatomy and pathology, aiding in diagnosis and treatment path determination by providing visualization, measurement, and advanced calcium scoring capabilities. It utilizes deep learning algorithms for automated calcium segmentation and scoring from DICOM CT data, supplemented by manual tools, and has shown performance comparable to existing methods in clinical evaluations.

CardIQ Suite is a non-invasive software application designed for the analysis of cardiovascular anatomy and pathology using 2D or 3D CT cardiac DICOM data, from both non-contrast and angiography acquisitions. Its primary purpose is to aid in the diagnosis and determination of treatment paths for various cardiovascular diseases, including coronary artery disease, functional heart parameters, heart structure abnormalities, and for follow-up on stent placements, bypasses, and plaque imaging. The software provides capabilities for visualizing and measuring vessels, assessing chamber mobility, and notably features a calcium scoring function to evaluate calcified plaques in coronary arteries, heart valves, and great vessels like the aorta, which can also be used to monitor calcium progression or regression over time. The technology integrates an automated calcium scoring component that utilizes a new deep learning algorithm for segmenting and labeling calcifications within coronary arteries, alongside traditional Agatston/Janowitz 130 (AJ130) and Volume scoring methods. While automation aims to improve workflow efficiency, CardIQ Suite also offers manual calcium scoring capabilities, allowing users to edit auto-scored lesions or manually score calcific lesions in various cardiac structures. The software includes a Coronary 2D Review toolset for interactive review of cardiac exams, including visualization of coronary artery paths and chamber motion. Non-clinical testing confirmed compliance with DICOM standards and successful design controls, while clinical testing involving three radiologists demonstrated very high correlations between manual scores (using a predicate device) and the automated scores from CardIQ Suite, indicating equivalent performance.

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 deep learning algorithm for calcium segmentation and labeling compare in accuracy and processing time to other commercially available AI-driven tools across diverse patient populations and varying image qualities?; What is the inter-observer and intra-observer variability associated with the manual editing features for calcium scoring, and how does this compare to the variability of the fully automated deep learning-based results?; Beyond the Agatston and Volume scoring methods, are there plans to incorporate advanced plaque characterization features, such as differentiating between calcified and non-calcified plaque or assessing plaque vulnerability, within the CardIQ Suite?; How does the 'Adaptive Volume' scoring method for calcium quantification differ from the standard Volume scoring, and what are the specific clinical scenarios or patient types where it might offer superior diagnostic or prognostic value?; What were the specific demographic and clinical characteristics of the patient cohort used in the retrospective CT exam database for validating the calcium scoring algorithm, and how does the algorithm perform on underrepresented patient groups or rare cardiac pathologies?