Arterys MICA

Arterys MICA is a medical diagnostic software platform that integrates DICOM and non-DICOM image management with advanced analysis capabilities, functioning on a client-server architecture with optional AI modules for specialized cardiovascular and oncological applications. The system provides quantitative data and visualization tools from MR and CT imaging to support, but not replace, clinical diagnosis by trained healthcare professionals.

Arterys MICA is a medical diagnostic software application designed to assist healthcare professionals by displaying, processing, storing, and transferring DICOM and non-DICOM medical data, with the exception of mammography. It functions as a comprehensive data management system for images and patient information, offering tools for filtering, digital manipulation, and quantitative measurements; however, it acts as a support tool and does not generate diagnoses directly. The system utilizes a client-server architecture, with the client software running on standard computers via a web browser and server software on Linux. Its basic image Viewer allows for 2D and 3D image interaction, study selection from a unified worklist, and management of uploaded studies, integrating functionalities that were previously part of separate predicate devices into a single platform. The Arterys MICA platform is extended by optional AI-powered modules for specialized clinical workflows: the Cardio AI module and the Oncology AI module. The Cardio AI module analyzes cardiovascular magnetic resonance (MR) images (multi-slice, multi-phase, velocity-encoded) from 1.5T and 3.0T scanners to provide reproducible quantitative data on the heart and its major vessels, featuring capabilities like flow-4D analysis, automatic landmark identification (user-editable), normalized volume measurements, and eddy current correction. The Oncology AI module provides analytical tools to help assess and document changes in morphological activity of lesions from MR and CT scans during diagnostic and therapy follow-up examinations. It supports the oncological workflow by aiding in the confirmation of lesion presence or absence, and providing tools for evaluation, quantification, longitudinal tracking of lesions (e.g., in lung and liver), and standardized reporting (LI-RADS, Lung-RADS). Both modules employ semi-automated segmentation techniques to enhance clinical assessment.

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 Cardio AI module's automated landmark identification and quantitative measurements compare to manual expert analysis in terms of accuracy and inter-observer variability across diverse patient populations and imaging protocols?; What specific machine learning models are employed in the Oncology AI module for lesion detection, segmentation, and longitudinal tracking, and what were the key metrics from the validation studies for these AI components?; Can the Arterys MICA software be seamlessly integrated with a wide range of existing PACS and Electronic Medical Record (EMR) systems, and what are the typical technical requirements or challenges encountered during such integrations?; What is the average processing time for computationally intensive tasks, such as 4D flow quantification in the Cardio AI module or multi-timepoint lesion tracking in the Oncology AI module, when running on recommended hardware configurations?; Beyond the stated validation on 1.5T and 3.0T MR scanners for the Cardio AI module, has its performance been evaluated or benchmarked on images acquired from different MR scanner vendors or those using non-standard acquisition parameters?