(125 days)
The Imbio RV/LV Software device is designed to measure the maximal diameters of the right and left ventricles of the heart from a volumetric CTPA acquisition and report the ratio of those measurements. RV/LV analyzes cases using an artificial intelligence algorithm to identify the location and measurements of the ventricles. The RV/LV software provides the user with annotated images showing ventricular measurements. Its results are not intended to be used on a stand-alone basis for clinical decision-making or otherwise preclude clinical assessment of CTPA cases.
The Imbio CT RV/LV Software is a set of medical image post-processing computer algorithms that together perform automated image segmentation and diameter measurements on computed tomography pulmonary angiography (CTPA) images. The device then reports the ratio of those diameter measurements. The Imbio CT RV/LV Software is a single command-line executable program that may be run directly from the command-line or through scripting and thus the user interface is minimal.
Imbio RV/LV Software is a Software and Medical Device (SaMD) intended to provide annotated images and a PDF report that will be read most typically at a PACS workstation. Imbio RV/LV Software is an aid only used to support a physician in the analysis of CTPA images.
The Imbio RV/LV Software program reads in DICOM CPTA image datasets, processes the data, then writes output DICOM files and summary reports to a specified directory. Imbio RV/LV Software outputs DICOMs of the original input DICOM CPTA images overlaid with color-codings representing the results of RV/LV computer caliper measurement. Additionally, a summary PDF report is output.
Imbio RV/LV Software does not interface directly with any CT scanner or data collection equipment; instead the software imports data files previously generated by such equipment and is integrated as part of the radiological work-flow, reducing the risk of use errors.
Here's an analysis of the Imbio RV/LV Software based on the provided FDA 510(k) summary, structured to answer your questions:
1. Table of Acceptance Criteria and Reported Device Performance
The FDA 510(k) summary provided does not explicitly state numerical acceptance criteria in terms of performance metrics (e.g., accuracy thresholds, sensitivity, specificity targets). Instead, it describes clinical performance testing designed to demonstrate improvement in agreement among general radiologists and accuracy compared to radiologists' measurements.
| Performance Metric | Acceptance Criteria (Not Explicitly Stated in Document) | Reported Device Performance (Summary of Findings) |
|---|---|---|
| Reader Study I: Improvement in Agreement Among General Radiologists | (Implicit: Demonstrates improved agreement with AI assistance) | Demonstrated improvement of agreement among general radiologists with the assistance of the RV/LV output report. |
| Reader Study II: Accuracy of RV/LV Diameter Ratios | (Implicit: Demonstrates accuracy compared to radiologists' measurements) | Demonstrated accuracy of RV/LV diameter ratios compared to radiologists' measurement of the RV/LV diameter ratio. |
2. Sample Size Used for the Test Set and Data Provenance
The document states: "Anonymized CTPA datasets were utilized in the reader study." However, it does not specify the exact sample size of the test set nor the country of origin of the data. It does indicate that the study was retrospective as it used "anonymized CTPA datasets."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document states: "The second test (Reader Study-II) will demonstrated the accuracy of RVLV diameter ratios compared to radiologist's measurement of the RVLV diameter ratio." This implies that radiologists' measurements were used as a reference for accuracy. However, the document does not specify the number of radiologists or their specific qualifications (e.g., years of experience, subspecialty) used to establish this ground truth.
4. Adjudication Method for the Test Set
The document does not describe any specific adjudication method (e.g., 2+1, 3+1) for establishing the ground truth or resolving discrepancies in Reader Study II. It only mentions "radiologist's measurement," implying individual radiologist readings were used as comparison. For Reader Study I, which focused on "agreement among general radiologists," the specific adjudication or consensus method to determine the "improved agreement" is not detailed.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
Yes, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was performed in the form of Reader Study I. This study "demonstrated the improvement of the agreement among general radiologist with the assistance of the RVLV output report."
Effect size of how much human readers improve with AI vs. without AI assistance: The document states that Reader Study I "demonstrated the improvement of the agreement among general radiologist with the assistance of the RVLV output report." However, it does not provide quantitative effect size results (e.g., specific statistical metrics like AUC improvement, sensitivity/specificity changes, or inter-reader agreement measures like kappa values) directly in this summary.
6. Standalone (Algorithm Only) Performance Study
Yes, a standalone performance evaluation was also done, implicitly referred to as Reader Study II. This study "demonstrated the accuracy of RVLV diameter ratios compared to radiologist's measurement of the RVLV diameter ratio." This test assesses the algorithm's direct output (RV/LV diameter ratios) against a human reference, indicating a standalone performance evaluation.
7. Type of Ground Truth Used
The ground truth used for performance evaluation (specifically for Reader Study II, or for comparison in general) was based on expert (radiologist) measurements. The document states, "accuracy of RVLV diameter ratios compared to radiologist's measurement of the RVLV diameter ratio."
8. Sample Size for the Training Set
The document does not provide any information regarding the sample size used for the training set of the AI algorithm.
9. How the Ground Truth for the Training Set Was Established
The document does not provide any information on how the ground truth for the training set was established.
§ 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).