Search Results

InSee is an accessory to the Vyaire Medical AirLife Incentive Spirometer (2500 mL and 4000 mL). InSee is indicated to flag and digitally record successful incentive spirometry attempts, use-related data and provide reminders to patients who are prescribed the Vyaire Medical AirLife Incentive Spirometer. It is intended to be used in a hospital environment by clinicians and patients.

Device Description

InSee is an accessory to the Vyaire Medical AirLife incentive spirometer. InSee consists of a plastic enclosure which attaches to the base of the AirLife incentive spirometer and records the following parameters: the number of attempts above 250 ml (AT), the Target Volume (TA), the number of successful attempts (SU) and the maximum target volume (MX). The InSee device also provide patient reminders. This device is only compatible with the Vyaire AirLife Volumetric Incentive Spirometer- 2500 & 4000 mL. InSee is intended for use in the Hospital environment by clinicians and patients. The device is non-invasive, reusable, and intended for use by one patient at a time.

AI/ML Overview

The InSee device, an accessory for incentive spirometers, underwent rigorous performance evaluation to demonstrate its safety and effectiveness. The following details outline the acceptance criteria and the study that validated the device's performance:

1. Table of Acceptance Criteria and the Reported Device Performance

Performance Metric	Acceptance Criteria	Reported Device Performance
Accuracy in detecting when the incentive spirometer (IS) piston crosses the target mark	$\geq$ 90% agreement with blinded physician observation	100% agreement with blinded physician observation
Repeatability across devices and target volumes	Not explicitly stated, but clinical study aimed to demonstrate high repeatability	100%
Absence of adverse events	Not explicitly stated, but expected for clinical validation	No adverse events reported
Absence of device deficiencies	Not explicitly stated, but expected for clinical validation	No device deficiencies reported
Absence of protocol deviations	Not explicitly stated, but expected for clinical validation	No protocol deviations reported
Absence of missing data	Not explicitly stated, but expected for clinical validation	No missing data reported

2. Sample Size Used for the Test Set and Data Provenance

Sample Size for Test Set: 30 subjects
Data Provenance: The study was a "prospective, IRB-approved performance validation study," implying the data was collected specifically for this study. The location of the study (e.g., country of origin) is not explicitly stated, but an IRB approval suggests it was conducted in a regulated clinical setting.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

Number of Experts: Not explicitly stated, but the ground truth was established by "blinded physician observation." It is implied that at least one, and likely multiple, physicians were involved in the observation to ensure blinding and robust ground truth establishment.
Qualifications of Experts: "Physician" is stated, but further specific qualifications (e.g., medical specialty, years of experience) are not provided in the document. The blinding suggests an objective assessment.

4. Adjudication Method for the Test Set

The document states "100% agreement with blinded physician observation." This implies that the physician's observations served as the direct ground truth for comparison, and no specific multi-reader adjudication method (e.g., 2+1, 3+1) beyond this single (or implicit consensus of multiple) blinded observation is detailed.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not conducted. The study focused on the accuracy of the InSee device's detection capabilities against human observation, rather than comparing human reader performance with and without AI assistance. The InSee device is an accessory that records data and provides reminders, not an AI interpretation tool that assists human readers in making diagnoses.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

Yes, the performance validation study evaluated the InSee device's ability to "accurately detect when the IS piston crossed the target mark, activated its audiovisual alert, and logged the inspiratory volume" in an automated fashion. The "blinded physician observation" served as the independent ground truth to verify the device's standalone accuracy in these functions.

7. The Type of Ground Truth Used

Expert Consensus/Observation: The ground truth was established by "blinded physician observation" of the incentive spirometry attempts, serving as the gold standard for verifying the InSee device's accuracy in detecting successful attempts and logging inspiratory volume.

8. The Sample Size for the Training Set

The document does not provide information regarding a distinct training set sample size. The description of the performance evaluation focuses on the "prospective ... performance validation study" of the device, which is typically a test or validation set for a final product. Given the nature of this device (an accessory that records physical parameters rather than an AI model learning from vast datasets), a separate, explicit "training set" might not be applicable in the traditional machine learning sense, or its "training" might be based on engineering principles and calibration rather than data-driven model training.

9. How the Ground Truth for the Training Set Was Established

As the document does not mention a distinct training set in the context of an algorithm or AI model, there is no information provided on how ground truth for such a set would have been established. The device appears to rely on sensors and predefined logic to detect spirometer events, rather than a learned AI model requiring labeled training data.

Ask a Question

Ask a specific question about this device

Page 1 of 1