(67 days)
The Reveal LINQ ICM is an insertable automatically-activated monitoring system that records subcutaneous ECG and is indicated in the following cases:
· patients with clinical syndromes or situations at increased risk of cardiac arrhythmias
· patients who experience transient symptoms such as dizziness, palpitation, syncope, and chest pain that may suggest a cardiac arrhythmia
The device has not been tested specifically for pediatric use.
The Reveal LINQ ICM (Model LNQ11) in association with the LINQ Mobile Manager Application (Model MSW001 or MSW002) and patient connector (Model 24965 and 24967) (referred to as the LINQ Mobile Manager system) are substantially equivalent to the following predicate device: Reveal LINQ ICM (Model LNO11) cleared via K160689 on April 22, 2016.
This document, K163460, is a 510(k) Premarket Notification for the Medtronic Reveal LINQ Insertable Cardiac Monitor, Model LNQ11. It's a submission to demonstrate substantial equivalence to a predicate device, not necessarily to prove the device meets new acceptance criteria for novel functionality with detailed performance data typically associated with AI/ML devices.
Therefore, many of the requested points regarding acceptance criteria, training/test sets, expert adjudication, MRMC studies, and standalone performance data for AI are not present in this document, as this device primarily relies on proving equivalence to an existing, already approved medical device, and the changes appear to be in a patient connector component, rather than core AI algorithms.
However, based on the information provided, here's what can be extracted and inferred:
1. A table of acceptance criteria and the reported device performance
The document does not provide a table of new acceptance criteria and their corresponding performance metrics because the submission is for showing substantial equivalence to a predicate device, not for a novel device with new performance claims. The performance data provided are generally related to safety and functionality to ensure the new component (patient connector Model 24967) does not negatively impact the existing device's performance.
The "acceptance criteria" here are implicitly that the new device component performs the same as the predicate device component and meets relevant safety standards.
-
Implicit Acceptance Criteria (for the new CareLink SmartSync Device Manager, patient connector, Model 24967):
- Biocompatibility: Must pass cytotoxicity, sensitization, and irritation tests.
- Electrical Safety and EMC: Must comply with IEC 60601-1 and IEC 60601-1-2 standards.
- Software: Verification and validation testing must be passed, with the software considered a "major" level of concern.
- Mechanical: Must pass inspections for design features, function, workmanship, labeling, control activation forces, chemical resistance, environmental/drop testing, and button/contact reliability.
- No negative impact on existing Reveal LINQ ICM functionalities (e.g., R-wave sensing, sampling rate, detection algorithms).
-
Reported Device Performance (from the document):
- Biocompatibility: "The battery of testing for materials used in the Model 24967 included the following tests: Cytotoxicity, Sensitization, Irritation." (Implies successful completion for acceptance).
- Electrical safety and electromagnetic compatibility (EMC): "The system complies with the IEC 60601-1, standards for safety and the IEC 60601-1-2 third and fourth edition versions of the standard for EMC." (Implies successful completion for acceptance).
- Software Verification and Validation Testing: "Software verification and validation testing were conducted and documentation was provided as recommended by FDA's Guidance..." (Implies successful completion for acceptance).
- Mechanical Testing: "The following is a list of testing performed: Inspection of the required mechanical design features and function...Workmanship inspection...Product labeling inspection...Forces required to activate controls...Chemical resistance testing...Environmental and drop testing...Reliability testing of buttons, electrical contacts, user connector insertions, and replaceable or moving mechanical components." (Implies successful completion for acceptance).
- Functional Equivalence: The comparison table (page 4) explicitly states "Same" for virtually all functional characteristics between the new configuration and the predicate device (e.g., R Wave Sensing, Sampling Rate, Storage Time, Noise Reversion, Brady Detection, Asystole Detection, Ventricular Tachycardia Detection, Afib Detection, Detection Algorithms, etc.). This implies performance is identical to the predicate which would have already met its own performance criteria.
2. Sample sizes used for the test set and the data provenance
The document describes non-clinical performance testing (biocompatibility, electrical safety, mechanical, software V&V) of a new component (patient connector) that connects to an existing, approved device. These tests typically involve lab-based testing of the specific component or system, not "test sets" in the sense of patient data for evaluating an algorithm's performance on clinical outcomes.
- Sample sizes: Not specified in terms of patient data. The tests performed are engineering and lab-based (e.g., number of test units for mechanical testing, number of samples for biocompatibility tests).
- Data provenance: Not applicable in the context of clinical data for AI/ML validation. These are engineering test results.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
Not applicable. The ground truth for biocompatibility is established by standard ISO tests. For electrical and mechanical safety, it's defined by engineering standards (e.g., IEC 60601 series). For software, it's based on verification and validation against requirements. There's no indication of clinical expert review for "ground truth" establishment in this type of submission for a system where the core algorithmic functionality is unchanged and equivalence is being sought for a specific component.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable, as noted above, this is about engineering and safety testing, not clinical data adjudication.
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 such study was performed or is mentioned. This submission is about a device (Insertable Cardiac Monitor) that records ECG and performs automated detection. It does not describe an AI that assists human readers in interpreting images or complex signals that would warrant an MRMC study.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The device itself includes detection algorithms (e.g., Brady, Asystole, VT, Afib). The document states these are "Same" as the predicate device. For the predicate device, or this device in itself, the performance of these automated detection algorithms would be considered a form of standalone performance. However, this submission specifically highlights that the algorithms themselves are unchanged from the predicate device (K160689). The focus of this submission is on the safety and performance of a new patient connector model that interfaces with the existing device.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
For the device's core functionality (i.e., its ability to detect arrhythmias), the ground truth in its initial approval would have likely been established through clinical trials comparing its detections against adjudicated ECGs, perhaps from expert cardiologists or verified clinical events. However, this information pertains to the predicate device's original approval and is not detailed in this 510(k) summary, as the algorithms are stated to be "Same."
For the new component (patient connector Model 24967) being approved in this submission, the "ground truth" for its acceptance criteria are established by:
- Biocompatibility standards: ISO 10993 series.
- Electrical safety standards: IEC 60601-1, IEC 60601-1-2.
- Software V&V best practices: FDA guidance for software in medical devices.
- Mechanical engineering standards: Internal Medtronic standards and potentially relevant industry standards for durability, force, etc.
8. The sample size for the training set
Not applicable. This document is not describing the development or validation of a new AI algorithm where a training set would be used. It's about demonstrating substantial equivalence for a hardware component.
9. How the ground truth for the training set was established
Not applicable, as there is no training set discussed for a new AI algorithm.
§ 870.1025 Arrhythmia detector and alarm (including ST-segment measurement and alarm).
(a)
Identification. The arrhythmia detector and alarm device monitors an electrocardiogram and is designed to produce a visible or audible signal or alarm when atrial or ventricular arrhythmia, such as premature contraction or ventricular fibrillation, occurs.(b)
Classification. Class II (special controls). The guidance document entitled “Class II Special Controls Guidance Document: Arrhythmia Detector and Alarm” will serve as the special control. See § 870.1 for the availability of this guidance document.