(90 days)
The Quantum Ventilation Module is intended for the continuous monitoring of critical clinical parameters during procedures that require extracorporeal circulation. The Quantum Ventilation Module is an accessory that only works with the Quantum Workstation. Parameters provided by the Quantum Ventilation Module include:
- · Measurement of up to three blood flow channels and arterial and venous flow differential
- Indication of gas bubbles
- · Extracorporeal gas flow measurements (02, CO2, gas flow, and CO2 removal)
- · Predicted PO2 and PCO2
- · Temperature
- · Up to three circuit pressure channels
- · Reservoir level indication
- Two channels of vacuum
- · Blend and control gas flow (air/O2/CO2)
The Quantum Ventilation Module is to only be used by an experienced and trained clinician. The device is not intended to be used by the patient or other untrained personnel.
The Quantum Ventilation Module provides gas blending and continuous non-invasive monitoring of critical clinical parameters in extracorporeal circuits used in cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) procedures. The Quantum Ventilation Module is an accessory to the Quantum Workstation or can be used in place of the Quantum Diagnostic Module as part of the Quantum Pump Console. When paired with the Quantum Workstation, the combination of the Quantum Workstation and Quantum Ventilation Module is known as the Quantum Ventilation System.
The Quantum Ventilation Module performs five functions:
-
- Provides measurements from embedded and attached sensors to monitor gases into and out of a blood oxygenator.
-
- Provides measurements from attached sensors for blood flow, bubble detection, pressure, level and temperature to monitor an extracorporeal blood loop.
-
- Provides gas blending to ensure the precision delivery of FiO₂ (21 to 100%), CO₂ and sweep flow rates.
-
- Provides regulation of vacuum supply to provide two channels of vacuum. One is flow-regulated to remove waste anesthesia gas, the other pressure-regulated for applications including Vacuum-Assisted Venous Drainage (VAVD) and hemoconcentration.
-
- Sends these physiological measurements to the Quantum Workstation for display to the user.
The Quantum Ventilation Module, with its attached sensors, can measure flow, pressure, reservoir level, temperature and gas diagnostics, in addition to performing electronic gas blending of up to three gases and built-in vacuum management for the removal of waste anesthetic gas. The primary interface for controlling and displaying measurements is the Quantum Workstation: however, the Quantum Ventilation Module also contains a display with control knobs. The Quantum Ventilation Module only works with the Quantum Workstation.
The provided document is a 510(k) Premarket Notification for a medical device called the Quantum Ventilation Module. This type of submission is for demonstrating "substantial equivalence" to a legally marketed predicate device, rather than proving safety and effectiveness through extensive clinical trials as would be required for a Premarket Approval (PMA) application.
Therefore, the document does not describe a study involving an AI algorithm or meeting the typical acceptance criteria for AI/ML device performance. Instead, it focuses on non-clinical performance data to demonstrate the device's functionality and safety as a medical instrument.
Given this context, I will address the questions to the best of what can be inferred from the provided text, noting where the information is absent due to the nature of a 510(k) submission for a non-AI hardware device.
Acceptance Criteria and Device Performance (Based on Non-Clinical Testing for a Hardware Device)
The document primarily discusses non-clinical performance data, which are typically tests to ensure the device performs as intended and is safe. The "acceptance criteria" here are implied by the successful completion of these tests.
1. A table of acceptance criteria and the reported device performance
Since this is a hardware device (Ventilation Module) and not an AI algorithm, the acceptance criteria are not in terms of common AI metrics like sensitivity, specificity, or AUC, nor is there comparative effectiveness data against human readers. The criteria are related to the device's physical and electronic performance, along with its software validation. The document states:
| Acceptance Criteria Category (Implied) | Reported Device Performance Summary (as per document) |
|---|---|
| Electrical Safety | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Electromagnetic Compatibility (EMC) | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Electrosurgery Interference | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Hardware Testing (Printed Circuit Boards) | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Software Verification and Validation | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Usability Validation | Non-clinical testing performed to support substantial equivalence. (Implies successful completion.) |
| Diagnostic Measurements Accuracy (e.g., flow, pressure, temperature, gas) | "Equivalent sensor performance" to the predicate device (Quantum Diagnostic Module K173591) is claimed, implying accuracy meets established standards for these parameters. |
| Gas Blending Precision (FiO₂, CO₂, sweep flow rates) | Claimed to "ensure the precision delivery" of these parameters. |
| Vacuum Regulation | Provides "regulation of vacuum supply." |
2. Sample size used for the test set and the data provenance
This information is not applicable in the context of an "AI test set" here. The "test set" would refer to the physical units of the device subjected to non-clinical tests. The tests performed ("Electrical safety", "Electromagnetic compatibility", "Electrosurgery interference", "Hardware testing of printed circuit boards", "Software verification and validation", "Usability validation") are typically laboratory-based engineering and software validation tests. The document does not specify the number of units tested, the conditions, or the specific "data provenance" (e.g., country of origin) beyond the manufacturer being in the UK. These are not data-driven performance studies on patient cohorts.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
This is not applicable as this is a hardware device. Ground truth, in the context of AI, refers to annotated data. For a hardware device, "ground truth" might refer to established measurement standards or known physical properties used for calibration and validation of sensors. The document does not specify the number or qualifications of experts involved in these engineering validation processes.
4. 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 establishing consensus among human readers for image labeling or clinical decision-making ground truth in AI studies. For hardware testing, performance is measured against engineering specifications and industry standards, not through expert adjudication in this manner.
5. 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 is a hardware medical device, not an AI-powered diagnostic or assistive tool. Therefore, an MRMC study is not relevant, and the concept of human readers improving with AI assistance does not apply.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
This refers to an AI algorithm's performance. The Quantum Ventilation Module is a hardware device with embedded software, but it's not an AI algorithm that makes diagnostic decisions or interpretations in the way this question implies. Its performance is the measurement and control capabilities of the instrument itself.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
Not applicable. For a hardware device, the "ground truth" for non-clinical performance validation would be derived from:
- Engineering specifications and design documents.
- International and national standards for electrical safety, EMC, and medical device performance (e.g., IEC standards).
- Calibration standards for sensors (e.g., precision gas mixes, flow simulators, temperature baths, pressure gauges).
- Bench testing and physical measurements.
8. The sample size for the training set
Not applicable. This device is not an AI/ML model trained on a dataset. It's a hardware device with firmware.
9. How the ground truth for the training set was established
Not applicable. As there is no "training set" in the AI sense, this question is not relevant. The "ground truth" for developing the device's functionality would stem from engineering principles, clinical requirements for extracorporeal circulation, and established medical device design practices.
§ 870.4300 Cardiopulmonary bypass gas control unit.
(a)
Identification. A cardiopulmonary bypass gas control unit is a device used to control and measure the flow of gas into the oxygenator. The device is calibrated for a specific gas.(b)
Classification. Class II (performance standards).