(196 days)
The SonarMed AirWave Airway Monitoring System is used to assist in verifying placement of the endotracheal tube (ETT), to assist in detecting movement of the ETT tip, to assist in detecting obstruction of the ETT, and to assist in listening to breath sounds.
The SonarMed AirWave Airway Monitoring System is intended for use by qualified personnel to assist with artificial airway management for patients in an in-hospital setting (intensive care, operating room, and emergency department settings, as well as intra-hospital transport).
The SonarMed AirWave Airway Monitoring System is to be used as an adjunct to normal clinical practice and is a not a stand-alone diagnostic system.
It is intended for use with neonates, infants, children, adolescents, and adults (sizes 2.5 mm to 9.0 mm).
The SonarMed AirWave is comprised of a SonarMed Monitor (Monitor) that is used in conjunction with a single-patient use SonarMed Sensor (Sensor) and software that operates the Monitor and Sensor. The Monitor is powered from an external power supply and has a battery backup. When in use, the SonarMed Sensor is placed in-line between the ventilator circuit and the proximal end of the endotracheal tube (ETT) of a patient who is connected to a ventilator.
Using acoustic reflection technology, signals from the Sensor are displayed on the Monitor showing the clinician:
- The baseline location of the ETT tip as established by the clinician
- Estimation of passageway around the tipof the ETT, relative to the . ETT diameter
- ETT movement relative to the baseline location ●
- . ETT occlusion / obstruction information including percent obstructed and location of the obstruction
- The clinician can choose whether to view information about the . patient's airway in either a waveform or graphic on the Monitor's LCD. Additionally, the clinician can use the microphones to listen to breath sounds. The information provided by the device is to be used in conjunction with normal clinical practice to assist with management of the artificial airway of the patient.
The provided text describes the SonarMed AirWave Airway Monitoring System and its substantial equivalence determination by the FDA. However, it does not explicitly detail the acceptance criteria and a study proving the device meets these criteria in the typical quantitative manner (e.g., sensitivity, specificity, AUC values) usually associated with AI/ML device performance.
The document focuses on demonstrating substantial equivalence to a predicate device (K143042) through various bench tests and by addressing modifications. The "acceptance criteria" appear to be implicit in the comparative performance against the predicate device, aiming for "as accurate or more accurate."
Based on the provided text, here's an attempt to answer your questions to the extent possible, acknowledging the limitations of the information provided for a typical AI/ML device performance study:
1. A table of acceptance criteria and the reported device performance
The document doesn't provide explicit quantitative acceptance criteria or corresponding reported performance metrics (like percentage accuracy thresholds) in a tabular format. Instead, it describes the conclusion of the bench testing:
| Performance Aspect | Implied Acceptance Criterion / Comparison | Reported Device Performance |
|---|---|---|
| Passageway Detection Accuracy | Accurately estimates diameter of passageway around ETT tip (compared to predicate) | "The Passageway Detection Study documented that the subject device accurately estimates the diameter of the passageway around the tip of the ETT for all patient populations." |
| ETT Movement Detection | Accurately detects distance and direction of ETT tip movement (compared to predicate) | "The Movement Study documented that the subject device accurately detects the distance and direction of the ETT tip movement both upward and downward for all patient populations." |
| Obstruction Detection | Accurately detects percentage of obstruction in ETT (compared to predicate) | "The Obstruction Detection Study documented that the subject device accurately detects the percentage of obstruction in the ETT for all patient populations." |
| Overall Performance | As accurate or more accurate than the predicate. | Stated under the "Performance" section of the comparison table: "As accurate or more accurate than predicate." |
2. Sample size used for the test set and the data provenance
The document refers to "bench testing" and "studies" (Passageway Detection Study, Movement Study, Obstruction Detection Study) but does not specify the sample sizes used for these tests.
The data provenance is not mentioned. It is implied these are pre-clinical/bench test data, not clinical patient data, given the description of testing on the device itself. Therefore, concepts like "country of origin of the data," "retrospective," or "prospective" as applied to patient cohorts are not applicable here.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
The document does not specify the number of experts or their qualifications for establishing ground truth. Given that the testing is described as "bench testing" on the device's accuracy for physical measurements (ETT movement, obstruction, passageway diameter), it's more likely that the "ground truth" was established by calibrated instruments or known physical alterations to the test setup, rather than human expert interpretation of images/data.
4. Adjudication method for the test set
The document does not mention any adjudication method, as it does not describe a study involving multiple human readers or interpretations. The "studies" appear to be objective measurements from the device against a known physical ground truth during bench testing.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done
No, an MRMC comparative effectiveness study involving human readers is not mentioned in the provided text. The evaluation focuses on the device's intrinsic measurement capabilities in a bench test setting, not on how human readers' performance with or without the device.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
The "bench testing" described appears to be a standalone performance evaluation of the device's sensors and algorithms, given that it aims to demonstrate the device "accurately detects" or "accurately estimates" various parameters.
7. The type of ground truth used
The ground truth used appears to be objective, known physical parameters established by the bench test setup for ETT tip location, obstruction percentage, and airway diameter. This is inferred from the description of "Passageway Detection Study," "Movement Study," and "Obstruction Detection Study." It is not based on expert consensus, pathology, or outcomes data in a clinical sense.
8. The sample size for the training set
The document does not mention a training set size. This device relies on acoustic reflection technology and embedded processing, which are described as being similar to the predicate device. It's not explicitly stated as an AI/ML device that requires a large, dedicated "training set" in the common sense of deep learning. Updates were made to software (e.g., O2 ranges, waveform screen functionality), but the underlying "technology" (acoustic reflectometry) remains the same as the predicate.
9. How the ground truth for the training set was established
As no training set is explicitly mentioned for a machine learning model, the method for establishing ground truth for such a set is not described.
§ 868.5730 Tracheal tube.
(a)
Identification. A tracheal tube is a device inserted into a patient's trachea via the nose or mouth and used to maintain an open airway.(b)
Classification. Class II (performance standards).