Arterys- - Version Cardio AI - Assisted Cardiac MRI Software
Visualize and quantify flow precisely anywhere in the entire heart. Enabling a better patient experience, faster CMR and precise analysis Arterys Cardio AI is built on the most advanced web rendering technology for medical imaging. You can load studies of any size, from 2D PC to multiple 4D Flow datasets. Enjoy breathtaking graphics and automation for a fast, comprehensive view. You’ve got to see it — and experience it — to believe it.
Save up to 25 minutes per study! Arterys Cardio AI is a cardiac MRI software solution that powers your workflow with deep learning and cloud supercomputing – while keeping you in full control. Generate and edit your reports faster and with precision. That’s what it means to be AI-assisted.
Arterys MICA includes an optional Cardio AI module which is used to analyze the heart and its major vessels using multi-slice, multi-phase, and velocity-encoded cardiovascular magnetic resonance (MR) images. It provides clinically relevant and reproducible, quantitative data, and has been tested and validated on MR images acquired from both 1.5T and 3.0 T MR Scanners. Arterys MICA software is intended to be used as a support tool by trained healthcare professionals to aid in diagnosis. It is intended to provide image and related information that is interpreted by a trained professional to render findings and/or diagnosis, but it does not directly generate any diagnosis or potential findings.
Information on training data
The 4D Flow Deep Learning based landmarks detection algorithm was trained on a database of 500 4D Flow MRI and expert landmarks annotations. For each study, the 6 landmarks (LVA, RVA, AV, PV, MV, TV) have been annotated on at least one time point by a cardiac radiologist. For all of LVA, RVA, AV, PV, TV, MV landmarks, we compute the median of the distance to the ground truth, and the 95% percentile of the median P1 using the bootstrap with 100,000 bootstrap samples. We require P1 to be less than or equal to 33mm for all detected landmarks. We exclude landmarks that are discarded by our quality detection algorithm from this measurement. The Deep Learning based SSFP segmentation algorithm was trained on 1143 cine cardiac MRI short axis series. The algorithm segmentation masks were compared to radiologist experts manual segmentation and DICE score was computed for comparison.
Model performance metrics
The value of P1 for the different landmarks are: LVA: 8.4mm RVA: 11.6mm MV: 10.1mm PV: 11.3mm TV: 12mm AV: 8.3mm For each landmark, the detection rate is: LVA: 90% RVA: 89% MV: 91% PV: 88% TV: 99% AV: 88%