Rapid

The Rapid software by iSchemaView is a comprehensive medical image processing system utilizing CT and MRI data to enhance diagnostic capabilities, particularly for acute stroke assessment and patient selection for thrombectomy. It integrates advanced analytical modules, including AI/ML for NCCT motion artifact detection and specific tools for perfusion, diffusion, and angiographic analysis, aiming to provide clinicians with timely and quantitative insights for critical treatment decisions.

Rapid, by iSchemaView Inc., is an advanced image processing software package designed for trained medical professionals to view, process, and analyze medical images from CT and MRI scanners. It operates on standard off-the-shelf computers or virtual platforms, acquiring DICOM-compliant image data to assist physicians in diagnosis. The software suite includes modules for CT Perfusion (CTP), CT Angiography (CTA), and MRI, featuring a Diffusion Weighted MRI (DWI) Module for visualizing local water diffusion properties and a Dynamic Analysis Module for assessing changes in contrast over time, thereby calculating parameters related to tissue flow and blood volume. A primary application is its utility in aiding physicians in the selection of acute stroke patients with known intracranial occlusions for endovascular thrombectomy, based on specific imaging criteria in conjunction with clinical assessment; contraindications include inadequate contrast bolus, excessive patient motion, or presence of hemorrhage. The Rapid system's CT analysis capabilities include generating Non-Contrast CT (NCCT) maps that delineate areas of hypodense and hyperdense tissue. A notable technological feature is its NCCT Motion Artifact AI/ML Module, which utilizes artificial intelligence and machine learning—trained on a dataset of 23,066 axial image slices and validated on an independent set of 619 cases from multiple scanner vendors—to detect motion artifacts with reported optimal performance (AUC=0.95, Sensitivity=0.95, Specificity=0.96 in training; independent validation AUC=0.96, Sensitivity=0.91, Specificity=0.86). The software is designed for streamlined workflow integration with minimal user interaction post-installation and features a server-based architecture (Rapid Server as a processing engine, connecting to a hospital LAN or via a virtualized appliance) for image processing and secure data management, including encrypted hard disks and secure network connections (VPN, SSH). Performance validation testing has confirmed its accuracy in representing key processing parameters and its compliance with DICOM standards.

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 NCCT Motion Artifact AI/ML Module vary across different CT scanner manufacturers and imaging protocols not included in the initial training and validation datasets?; What is the typical processing time for a complete stroke protocol dataset (e.g., NCCT, CTA, CTP) using Rapid, and how does this impact time-critical emergency stroke workflows?; Are there specific thresholds or quantitative outputs from the Dynamic Analysis Module (e.g., tissue flow, blood volume) that have been correlated with patient outcomes post-thrombectomy?; How does the software handle co-registration of images from different modalities (e.g., CT and MRI) or images acquired at different time points, and what is the typical accuracy of this co-registration?; What are the specific imaging criteria that Rapid CT-Perfusion and Rapid MR-Perfusion utilize to aid in the selection of acute stroke patients for thrombectomy, and how sensitive are these criteria to variations in contrast agent administration or bolus quality?