(84 days)
The AeroNOx 2.0 is intended to provide a constant controlled concentration of nitric oxide in breathing gas by delivering a constant controlled flow of nitric oxide into the inspiratory limb of a mechanical ventilator that operates using a continuous constant flow of fresh gas into the inspiratory limb of the ventilator. The AeroNOx 2.0 is also intended to be used with a flow inflating manual ventilator (an AeroNOx 2.0 accessory), by introducing controlled flows of nitric oxide into the fresh gas flow to the manual ventilator. It is also intended to monitor nitric oxide, nitrogen dioxide, and oxygen concentrations in the breathing gas. The AeroNOx 2.0 is intended to be used within a hospital or during air or ground transport outside the hospital.
The AeroNOx 2.0 nitric oxide delivery system is specifically designed for the transmission and control of gaseous nitric oxide (NO) in parts per million (ppm) concentrations. It is divided into two main components: the delivery component and an analysis component. These two components are mounted into a single enclosure. The delivery component is designed to deliver a constant flow of nitric oxide gas into the main gas stream of a constant flow ventilator or a modified flow inflating manual resuscitator (see "AeroNOx Bagger" and "INOstat Bagger" below). The nitric oxide gas is titrated into the ventilator's gas delivery tubing at a point at least 30cm upstream of the gas analysis port to ensure adequate gas mixing prior to patient delivery. The analysis component of the AeroNOx 2.0 is designed to measure nitric oxide, nitrogen dioxide and oxygen from the ventilator tubing or flow inflating manual resuscitator on the inspiratory side, prior to patient administration. The analysis system consists of one nitric oxide electrochemical cell, one nitrogen dioxide electrochemical cell, one galvanic oxygen sensor, and a pump to draw the gas sample from the ventilator's bulk gas flow for analysis within the block that mounts sensors.
The provided text is a 510(k) summary for the AeroNOx 2.0 Nitric Oxide Titration & Monitoring System. It describes a device modification submission, specifically adding compatibility with more ventilators. This is not a submission for a new AI/ML-driven medical device, and as such, the typical acceptance criteria and study designs associated with AI/ML performance (like MRMC studies, ground truth establishment with experts, data provenance for AI training/test sets) are not applicable to this document.
The "acceptance criteria" here refer to the device meeting performance requirements with the newly compatible ventilators, ensuring substantial equivalence to the predicate device.
Here's how to address the request based on the provided document:
1. A table of acceptance criteria and the reported device performance:
The document broadly states that the device "met all the performance requirements as outlined above" but does not detail specific quantitative acceptance criteria (e.g., accuracy, precision ranges for gas concentrations) or the numerical results of those tests. This type of information would typically be in a more detailed test report, not usually fully replicated in a 510(k) summary.
Acceptance Criteria (Implied from the document):
| Parameter | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| Ventilator Compatibility | Device functions correctly and safely with additional ventilators. | "The AeroNox 2.0 met all the performance requirements as outlined above." |
| Performance (e.g., NO, NO2, O2 monitoring, NO titration) | Maintain established performance from the predicate device. | "The AeroNox 2.0 met all the performance requirements as outlined above." |
| Safety | No new safety concerns introduced with the modification. | Implied by "found to be substantially equivalent to the predicate devices." |
2. Sample size used for the test set and the data provenance:
- Sample Size: Not explicitly stated. The document only mentions "Ventilator Compatibility Testing." It does not specify how many different ventilators were tested, nor the duration or number of trials for each.
- Data Provenance: This is a physical device test, not a data-driven AI test. Therefore, data provenance in the context of "country of origin of the data, retrospective or prospective" is not directly applicable. The testing was prospective, as it was conducted to support this specific 510(k) submission.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):
This question is not applicable. This is a hardware/system performance test, not a clinical study involving human expert interpretation of medical images or data. "Ground truth" for this device would refer to known gas concentrations in the test setup, measured by calibrated reference instruments. Experts are not involved in establishing this type of ground truth.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
This question is not applicable. Adjudication methods like 2+1 or 3+1 are used in clinical studies where human readers provide interpretations (e.g., radiologists interpreting images) and consensus is needed to establish ground truth. This is a bench test of a medical device's functional performance.
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:
This question is not applicable. MRMC studies are specific to evaluating the impact of AI/ML algorithms on human reader performance, typically in diagnostic imaging. This submission concerns a physical medical device (nitric oxide delivery and monitoring system), not an AI algorithm assisting human readers.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
This question is not applicable. The AeroNOx 2.0 is a physical device, not a standalone algorithm. While it contains internal algorithms for controlling gas delivery and monitoring, the "standalone performance" concept usually refers to an AI algorithm's diagnostic or predictive accuracy when operating independently of human clinicians. The document describes bench testing of the system's overall functionality.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
For this device, the "ground truth" would be established by reference instruments or known gas mixtures used during the bench testing. For example, if testing NO concentration accuracy, the "ground truth" would be the precisely known concentration of NO in the test gas, measured by a highly accurate, calibrated lab-grade gas analyzer. It is not expert consensus, pathology, or outcomes data.
8. The sample size for the training set:
This question is not applicable, and potentially misleading. The term "training set" refers to data used to train AI/ML models. This device is a traditional electromechanical medical device, not an AI/ML product. It does not have a "training set" in the machine learning sense. Any calibration data used by the sensors or control systems would be part of the device's inherent design and calibration process, not a "training set."
9. How the ground truth for the training set was established:
This question is not applicable for the same reasons as #8.
§ 868.5165 Nitric oxide administration apparatus.
(a)
Identification. The nitric oxide administration apparatus is a device used to add nitric oxide to gases that are to be breathed by a patient. The nitric oxide administration apparatus is to be used in conjunction with a ventilator or other breathing gas administration system.(b)
Classification. Class II. The special control for this device is FDA's “Guidance Document for Premarket Notification Submissions for Nitric Oxide Administration Apparatus, Nitric Oxide Analyzer, and Nitrogen Dioxide Analyzer.”