(77 days)
The ECHO-COM for MS-Windows Version 2.0 cardiac software program is recommended for quantitative analysis of M-mode, B-mode, and Doppler ultrasound data. The system may be employed adjunctively in the performance of stress echocardiography in patients with suspected coronary artery disease either on-line by direct connection to the video signal output port of a suitable diagnostic ultrasound system or off-line by retrieving and processing previously stored data. The system permits the use of either standard or customized examination protocols, data analysis formulae, and report formats.
The ECHO-COM device is a proprietary software program for cardiac data analysis designed for installation in an IBM-compatible personal computer (PC) and use with an MS-DOS Windows operating platform. Certain dedicated hardware items are provided with the software. The PC itself and the general-purpose peripheral hardware necessary to complete the system are supplied by or obtained from local vendors. The dedicated hardware supplied with the ECHO-COM system consists of a real-time frame grabber, ECG trigger kit, control panel, hardlock, and video cable. The digitizer card interface for the frame grabber has sufficient memory for the processing and storage of color images. This frame grabber captures all available information from the ultrasound system with which it is used during on-line operation including ECG trigger pulses and scale markers for distance measurements. The ECG trigger pulses permit time-gated cine loop recording, optimizing the system for the performance of stress-echo examinations. ECHO-COM software is designed to offer exceptional flexibility during either on-line or off-line operation. Images and cine loops may be archived in a variety of commonly used formats, including Macintosh, permitting integration into a wide range of networking systems such as InterNet and CompServe for convenient transfer of data to remote locations. A Review Module is provided that enables users to perform measurement, calculation, and documentation of stored images and loops at the workstation. An Examination Scheme Editor permits the user to supplement the pre-programmed protocols with customized examination approaches to conform with local requirements and preferences. Data from 2-D, M-mode, and Doppler examinations can be combined as desired. Levels and scanning positions for stress-echo examinations can be freely selected. The Report and Formula Editor enables the user to add custom-designed report formats and supplementary calculation formulae to the standard menu. A comprehensive range of cardiac ultrasound examination programs is provided including stress-echo test computations, semi-quantitative wall motion scoring, wall thickness evaluation, left ventricular volume studies, left ventricular function, color coding, M-mode functions, and Doppler functions. The disc summation method for quantitation of left ventricular volumes by two-dimensional echocardiography, as recommended by the American Society of Echocardiography, is employed. The tolerance levels utilized in the ECHO-COM program are based directly on the range of normal values adopted by the American Society of Echocardiography. All other equations and formulae included in the ECHO-COM program are based on accepted mathematical principles or derived from the published literature.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Acceptance Criteria and Device Performance
The provided document (K960665) is a 510(k) summary for a medical device submitted in 1996. The regulatory landscape and requirements for clinical studies, especially for AI/software as a medical device (SaMD), have evolved significantly since then. As a result, the "performance data" section in this summary is very brief and does not meet the standards or detail expected in modern submissions regarding acceptance criteria and studies.
Based on the information provided, the core "acceptance criteria" presented are related to the performance of the software's calculations compared to existing echocardiographic systems.
Table of Acceptance Criteria and Reported Device Performance:
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Numeric values calculated by ECHO-COM for MS-Windows program must align with corresponding values generated by the echocardiographic system itself. | "Software performance has been successfully validated through checks conducted in hospitals under typical clinical operating conditions in addition to internal testing throughout the software development cycle." |
| Adherence to quality standards for design, production, testing, and service procedures. | "All product design, production, testing, and service procedures are conducted in accordance with the following international quality standards: ISO 9001:1994 and ISO 9004-2:1991." |
Study Details Based on Provided Text:
Given the 1996 context, the "study" described is a very high-level validation of calculated values, not a rigorous clinical trial as would be expected today.
-
Sample size used for the test set and the data provenance:
- Sample Size: Not specified. The document only mentions "checks conducted in hospitals."
- Data Provenance: The "checks" were conducted "in hospitals under typical clinical operating conditions." This suggests prospective, real-world data collection during clinical use, but specific countries or retrospective/prospective details are not provided beyond this.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not specified. The ground truth for direct comparison appears to be the "corresponding values generated by the echocardiographic system itself." It is implicitly assumed that the operators/clinicians using the echocardiographic system are experts, but their number and specific qualifications are not detailed.
-
Adjudication method for the test set:
- Not applicable/Not specified. The validation method described is a direct comparison of numeric output from the ECHO-COM system to the source echocardiographic system. There is no mention of a human adjudication process to resolve discrepancies, though such a process might have implicitly occurred during "validation."
-
If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- No. This type of MRMC comparative effectiveness study was not conducted or reported. The device is a "Cardiac Data Analysis Program" and is intended to calculate metrics based on ultrasound data, not to assist human readers in, for example, image interpretation in a diagnostic setting that would typically warrant an MRMC study.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, effectively. The software's performance was evaluated by comparing its calculated numeric output directly against the numeric output of the echocardiographic system itself. This represents an algorithm-only performance evaluation in terms of its computational accuracy. Human interaction is involved in operating both systems, but the comparison is focused on the software's inherent ability to calculate correctly.
-
The type of ground truth used:
- Expert Consensus / Device Output: The ground truth for the software's calculations appears to be the "corresponding values generated by the echocardiographic system itself" (which itself provides quantitative measurements). The document also states that the device's "tolerance levels" and "equations and formulae" are "based directly on the range of normal values adopted by the American Society of Echocardiography" and "accepted mathematical principles or derived from the published literature." This suggests that the ground truth for the underlying methodology is established medical and scientific consensus, and the "validation" checks confirm the software accurately implements this consensus.
-
The sample size for the training set:
- Not specified. The document primarily discusses validation and compliance with established medical principles, not a machine learning training process with a specific training set size. This device pre-dates the widespread use of deep learning and large-scale training datasets for medical devices.
-
How the ground truth for the training set was established:
- Not applicable/Not specified in terms of a "training set" as understood in modern AI/ML development. The software's "knowledge" or "rules" are based on "accepted mathematical principles," "published literature," and "normal values adopted by the American Society of Echocardiography." These served as the foundation for the software's design and implementation, rather than a data-driven training process.
§ 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).