(77 days)
AutoMIStar is a medical image post-processing software package to be used by trained professionals, including but not limited to physicians and medical technicians. The software runs on a standard "off-the-shel" computer or a virtual platform with Windows operating system, and can be used to perform image viewing, processing and analysis of brain images acquired by DICOM compliant imaging devices.
AutoMIStar provides general viewing of DICOM images, with analysis capabilities for functional imaging data including a CT Perfusion Module and a DWI (diffusion-weighted MRI) Module.
The CT Perfusion Module is used for visualization and analysis of contrast enhanced CT Perfusion (CTP) dataset, showing properties of changes in contrast over time. This function includes calculation of various parameters related to tissue flow (perfusion) and tissue blood volume.
The DWI Module is used to visualize local water diffusion properties from the analysis of diffusion-weighted MRI data.
AutoMIStar is a software package that provides visualization and processing of medical imagesacquired by CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) scanners, as an aid to physician diagnosis. It can be installed and run on a PC, laptop or virtual machine running a Microsoft Windows operating system. It provides viewing, quantification, analysis and reporting capabilities.
AutoMIStar also provides processing and analysis of functional and dynamic imaging datasets including CT Perfusion (CTP) and diffusion-weighted MRI (DWI). It provides tools to perform the following type of advanced analysis:
- time intensity plots for dynamic time courses;
- volumetry of threshold maps;
- calculation of mismatch between volumes of different threshold maps;
AutoMIStar can be configured to connect to various DICOM devices (modality scanners, workstations, PACS) via the hospital medical imaging network to receive CT or MRI images as they become available. It supports transfer of DICOM images using the DICOM standard network protocol, and allows import of DICOM images from storage media. It provides automated workflow to processes received data and sends post-processed results to designated DICOM devices. It can also be configured to send post-processed results to designated recipients via email through the hospital email system.
AutoMIStar is a DICOM-compliant PACS softwarethat provides functionality to transfer, process, and display of CT and MR imaging data including dynamically acquired CT perfusion imaging data and diffusion-weighted MRI (DWI).
AutoMIStar runs on a standard "off-the-shelf" computer with Windows operating system, and can seamlessly integrate into an existing radiological data network. AutoMIStar is entirely independent from CT, MRI, or PACS platforms.
Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't explicitly state quantitative acceptance criteria in a structured table. Instead, it describes general performance goals and compares performance to a predicate device and other software packages.
| Acceptance Criterion (Inferred from Text) | Reported Device Performance |
|---|---|
| Accurate calculation of perfusion parameters (CBF, CBV, MTT, DT). | Achieved pre-established performance goals for all perfusion parameters (CBF, CBV, MTT, DT). Regression analysis between AutoMIStar output and ground truth showed strong correlation. Performance was comparable to published results of seven other commercially available and academic CTP software packages. |
| Accurate calculation of diffusion parameters (ADC). | Achieved pre-established performance goals for all diffusion parameters (ADC). Regression analysis between AutoMIStar output and ground truth showed strong correlation. Performance was comparable to published results of seven other commercially available and academic CTP software packages. |
| Accurate representation of perfusion threshold maps and volumes. | Performance validation was conducted on perfusion threshold maps and volumes, demonstrating substantial equivalence. |
| Accurate representation of diffusion threshold maps and volumes. | Performance validation was conducted on diffusion threshold maps and volumes, demonstrating substantial equivalence. |
| DICOM compliance. | AutoMIStar complies with DICOM (Digital Imaging and Communications in Medicine), developed by the American College of Radiology and the National Electrical Manufacturers Association. NEMA PS 3.1-3.20. |
| Satisfy all design requirements and device specifications. | Software verification and validation testing demonstrated that AutoMIStar met all design requirements and specifications. Performance validation testing demonstrated that AutoMIStar provides accurate representation of key perfusion and diffusion parameters associated with the intended use of the software. |
| Substantial equivalence to predicate device (IschemaView RAPID) for CTP/DWI. | AutoMIStar has the same intended use and similar indications, technological characteristics, and principles of operation as its predicate device (IschemaView RAPID). While it lacks dynamic contrast-enhanced MRI (DCE-MRI) functionality present in the predicate, this minor difference does not alter the intended diagnostic use or affect safety/effectiveness for CTP/DWI. The Delay Time (DT) parameter in AutoMIStar is considered similar to the Tmax parameter in RAPID, both relating to arterial delay effects. |
2. Sample Size Used for the Test Set and Data Provenance:
- Sample Size: The document does not explicitly state the specific number of cases or images used for the test set. It mentions "published perfusion phantom and diffusion phantom data."
- Data Provenance: The data used for testing was derived from "published perfusion phantom and diffusion phantom data." This suggests the data is likely from research studies or publicly available datasets specifically designed for validating perfusion and diffusion analysis software. The country of origin is not specified, but the nature of phantom data implies it's synthetic or laboratory-generated rather than patient-specific. The data would be considered retrospective in the sense that it was pre-existing for testing purposes, not prospectively collected for this device's validation.
3. Number of Experts Used to Establish Ground Truth and Qualifications:
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified. The ground truth was established by "published perfusion phantom and diffusion phantom data," implying that the ground truth was inherent to the design of these phantoms, which are validated by established methodologies in the field.
4. Adjudication Method for the Test Set:
- Adjudication Method: Not applicable/not specified. Since the ground truth was established by phantom data, there was no need for human expert adjudication of individual cases in the traditional sense. The phantom data itself provided the ground truth values.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- Was an MRMC study done? No. The document does not mention an MRMC study or any comparison of human readers with vs. without AI assistance. The performance testing focused on the standalone algorithm's accuracy against phantom data and comparison to other software.
- Effect size of human readers improve with AI vs without AI assistance: Not applicable, as no MRMC study was conducted.
6. Standalone (Algorithm Only) Performance:
- Was standalone performance done? Yes. The entire performance data section describes the evaluation of the AutoMIStar device itself against ground truth from phantom data. This is a standalone algorithm performance assessment. The "regression analysis between the output of the AutoMIStar device and the ground truth values" is a direct measure of its standalone accuracy.
7. Type of Ground Truth Used:
- Type of Ground Truth: The ground truth was established using phantom data (specifically, "published perfusion phantom and diffusion phantom data"). This type of ground truth provides precise, known values for the parameters being measured (e.g., CBF, CBV, MTT, ADC), which is ideal for quantitative validation.
8. Sample Size for the Training Set:
- Sample Size: The document does not provide any information regarding the sample size of a training set. This suggests that the device, or at least the part being validated, does not rely on a machine learning model that requires a distinct training phase with labeled data in the way a deep learning algorithm would. It appears to be based on established algorithms for perfusion and diffusion analysis.
9. How the Ground Truth for the Training Set Was Established:
- How Ground Truth for Training Set Established: Not applicable. As no training set information is provided, the method for establishing its ground truth is also not mentioned.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).