Search Results

The AirPurge™ System is intended for detection and automatic removal of air in intravenous (I.V.) lines during administration of intravenous solutions, blood and blood products. It is indicated for use in the Operating Room and post anesthesia care areas. The AirPurge™ System is placed distal to I.V. bags using gravity feed or pressure, and may be used with or without fluid warmers.

The Airpurge System (AirPurge) is intended to detect and automatically remove air from an IV infusion line. Airpurge consists of two main components, the control unit and the disposable unit. The disposable unit (Figure-1a) contains the system fluid path and is a sterile, single-use component. This disposable unit is loaded onto the control unit (Figure-1b). The door of the control unit is then closed. Once the device is connected to the infusion line and primed, it uses an ultrasonic sensor to detect air in the infusion line. When air is detected, the device system closes the patient line and diverts the fluid flow to the waste collection bag. A second ultrasonic sensor detects when the air has been removed from the IV line and causes the controller to return the infusion flow back to the patient.

Here's an analysis of the AirPurge System's acceptance criteria and the studies proving it meets them, based on the provided text:

Acceptance Criteria and Device Performance (Table 1 and 5)

Study Information

The provided document describes non-clinical/bench studies to demonstrate the AirPurge System's performance and safety. It does not detail any clinical studies, MRMC studies, or standalone algorithm performance studies in the way typically associated with AI/software-as-a-medical-device (SaMD). The device itself appears to be a physical device with embedded software and sensors, not an AI diagnostic algorithm.

Here's the breakdown of the available information:

1. Sample size used for the test set and the data provenance:

   • Test Set Sample Size: Not explicitly stated for each specific performance characteristic (e.g., how many air bubbles of 25 microliters were tested, or how many purge cycles were performed at each flow rate). The studies are referred to as "non-clinical/bench studies," suggesting laboratory-based testing rather than patient data.
   • Data Provenance: The data is generated from non-clinical/bench testing conducted by Anesthesia Safety Products, LLC. There is no mention of country of origin of data or retrospective/prospective human data, as these were not clinical trials.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • This is not applicable to the described studies. The "ground truth" for the performance characteristics (e.g., actual air volume, actual flow rate, actual fluid loss, actual response time) would be established by the precise instrumentation and methodologies used in the bench testing, not by expert human interpretation.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • Not applicable. Adjudication methods like 2+1 or 3+1 are used for expert review of human image data or clinical cases to establish a consensus ground truth. The studies described are bench tests with objective, measurable parameters.

4. 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 MRMC comparative effectiveness study was done or described. This device is an automated physical system for air removal, not an AI diagnostic tool designed to assist human readers/clinicians in interpretation.

5. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:

   • Yes, in the context of the device's function. The device (which includes embedded software) performs its air detection and removal function "standalone" as an automated system. The non-clinical performance tests ("Limits of Air Detection," "Fluid Loss," "Flow Rate," "Response Time") directly assess this standalone performance of the device system. However, this is not a standalone AI algorithm in the typical sense of a diagnostic or predictive model.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

   • The ground truth for the non-clinical performance studies was established through controlled laboratory measurements and characterization. For example:
     • Air volume: Precisely measured air injection.
     • Flow rate: Precisely controlled fluid flow rates.
     • Fluid loss/response time: Measured by instruments during testing.
   • For other aspects like biocompatibility, sterilization, and electrical safety, the ground truth refers to conformance with established standards and the results of laboratory tests.

7. The sample size for the training set:

   • Not applicable. This device is not an AI/machine learning model in the typical sense that would require a "training set" of data. The software within the device is likely more rule-based or deterministic, rather than a learned model.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no "training set" for an AI model.

Summary of Studies and Evidence:

The AirPurge System's regulatory approval is based on a comprehensive set of non-clinical/bench studies that verify its physical and software performance, safety controls, biocompatibility, sterility, electrical safety, and human factors. These studies, rather than clinical trials with human data, directly address the acceptance criteria related to its technical specifications (air detection, flow rates, fluid loss, response time) and hazard mitigation. The "ground truth" for these studies is derived from controlled laboratory conditions and measurements against predefined technical specifications and regulatory standards.

Ask a specific question about this device