K Number
K970344
Date Cleared
1997-08-06

(189 days)

Product Code
Regulation Number
868.5630
Panel
AN
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The MDILog system monitors the compliance of asthma and chronic respiratory patients with metered dose inhalers (MDIs) and allows for the physician to assess patients' techniques in using MDIs. The MDILog is intended for outpatient use by a single individual under the care or treatment of a physician or licensed health care professional. The device is to be used whenever patient compliance monitoring is indicated. The health care professional prescribes the patient medication treatment plan and MDILog monitors and records MDI medication usage.

The MDILog is intended for use by a single patient under the care or treatment of a physician or licensed health care professional. The MDILog is prescribed by the doctor when MDI usage monitoring is indicated. The MDILog can be used by any patient who regularly uses MDIs as prescribed by a physician.

Device Description

The MDILog system monitors the compliance of asthma and chronic respiratory patients with metered dose inhalers (MDIs) and allows the physician to assesses patients' technique in using MDIs. The system consists of the MDLog electronic monitor, an adapter, a docking station to communicate to an IBM compatible personal computer, and software. The MDILog monitor is a small electronics module operating on a 3-volt battery that attaches to the dispensing boot of a metered dose inhaler (MDI). It records the time and date when a patient uses an MDI, measures certain properties of the patient's technique, and transmits these data to a docking station, which transmits the data to a computer.

Attached onto outside of dispenser body. Heated thermister used to detect air flow, mechanical beam with strain gage used to detect canister actuation and moving magnet used to detect shake. Data stored in device memory for later retrieval.

AI/ML Overview

This looks like a 510(k) premarket notification document for a medical device called MDILog, Model MDC-511. This type of document is submitted to the FDA to demonstrate that a device is substantially equivalent to a legally marketed predicate device, rather than proving its effectiveness through clinical trials with specific acceptance criteria in the same way a new drug or high-risk device might.

Therefore, the requested information about "acceptance criteria" and a "study that proves the device meets the acceptance criteria" in the traditional sense of a clinical trial with performance metrics like sensitivity, specificity, accuracy, and detailed statistical analysis is not present in this document. The submission focuses on demonstrating equivalency based on technological characteristics and performance in simulated-use conditions rather than a comparative effectiveness study involving human readers or a standalone algorithm performance study.

Here's a breakdown of the available information based on your request, highlighting what is not covered by this type of submission:

1. A table of acceptance criteria and the reported device performance:

  • Acceptance Criteria (Not applicable in the traditional sense for this 510(k)): This document does not define specific clinical performance "acceptance criteria" like sensitivity or specificity. Instead, the "acceptance" is based on demonstrating substantial equivalence to predicate devices through technical characteristic comparison and performance testing.

  • Reported Device Performance: The document describes "performance testing in a simulated-use condition" and "environmental tests," but it does not provide any specific quantitative results or metrics from these tests. The "Table of Comparison to Legally Marketed Devices" (excerpted below) highlights features and technical specifications rather than performance outcomes against predefined clinical criteria.

    CharacteristicNebulizer Chronolog System (K823423)PeakLog System (K940835)Doser (K93555)MDILog System
    Intended UseMonitor medication usage and compliance.Measures spirometric functions and monitors compliance.Monitor medication usage.Monitor medication usage and compliance.
    Principle of OperationMDI canister placed in Chronolog body. Mechanical switch used to detect canister actuation. Data stored in device memory for later retrieval.Hot wire technology used to detect air flow & firmware algorithms used to extrapolate measurements. Data stored in device memory for later retrieval.Attaches onto MDI canister in dispenser. Pressure activated when the Doser is pushed down thus pushing the MDI to dispense medication. Counts stored in device memoryAttached onto outside of dispenser body. Heated thermistor used to detect air flow, mechanical beam with strain gage used to detect canister actuation and moving magnet used to detect shake. Data stored in device memory for later retrieval.
    Data CollectionRecords date and time of each MDI canister actuation.Measures PEFR and FEV1. Records date and time of each measurement.Records number of MDI canister actuations.Records date and time of shake, canister actuation and inhale.
    Internal ClockYes (4 minute resolution)Yes (1 second resolution)Yes (30 minute resolution)Yes (1 second resolution)
    Maximum Maneuvers Stored256500300 (approximately)1300
    Battery LifeOver 1 year4 weeks1 yearSix months
    Patient ReminderNoneYes - BeepsYes - BeepsYes - Beeps (reminds patient when to take medication and when canister is empty)
    (and other characteristics listed in the provided document)

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective):

  • This information is not provided. The document mentions "performance testing in a simulated-use condition" but does not detail the sample size (e.g., number of uses, devices, or individuals involved in the simulation), data provenance, or whether the study was retrospective or prospective.

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 information is not applicable/provided. The 510(k) submission does not describe a test set with human-established ground truth in a clinical context. The performance testing was in a "simulated-use condition," which would likely involve engineering or technical verification rather than expert clinical assessment of patient data.

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

  • This information is not applicable/provided for the reason stated above.

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 is not applicable. The MDILog is a compliance monitoring device for MDIs, not an AI-powered diagnostic tool that assists human readers. No MRMC study or AI assistance is mentioned.

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

  • This is not applicable in the context of an "algorithm" as commonly understood in AI. The device itself is standalone in its monitoring function, but the "performance testing in a simulated-use condition" doesn't detail performance metrics that would typically apply to algorithm-only evaluations (e.g., specific accuracy, precision, recall for an AI task).

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

  • This information is not provided. Given the nature of "simulated-use condition" performance testing for a monitoring device, the "ground truth" would likely be based on controlled experimental conditions and known mechanical/electronic outputs, rather than clinical ground truth like pathology or expert consensus. For example, if the device is testing MDI actuation detection, the ground truth would be the actual actuation events under controlled settings.

8. The sample size for the training set:

  • This information is not provided. This device is a hardware monitoring system with embedded software, not a machine learning model that undergoes "training" on a dataset in the AI sense.

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

  • This information is not applicable for the reason stated above (not an AI training set).

In summary:

This 510(k) notification focuses on establishing substantial equivalence based on the device's technological characteristics and intended use compared to existing predicate devices. It briefly mentions "performance testing in a simulated-use condition" and "environmental tests" as support, but it does not provide the detailed study information, acceptance criteria, or quantitative performance results that would be expected for a device proving clinical effectiveness or an AI model's performance. The review process for such a device is primarily concerned with safety and the demonstration of equivalent functionality to already cleared devices.

§ 868.5630 Nebulizer.

(a)
Identification. A nebulizer is a device intended to spray liquids in aerosol form into gases that are delivered directly to the patient for breathing. Heated, ultrasonic, gas, venturi, and refillable nebulizers are included in this generic type of device.(b)
Classification. Class II (performance standards).