(58 days)
The Confirm Rx™ ICM is indicated for the monitoring and diagnostic evaluation of patients who experience unexplained symptoms such as: dizziness, palpitations, chest pain, syncope, and shortness of breath, as well as patients who are at risk for cardiac arrhythmias. It is also indicated for patients who have been previously diagnosed with atrial fibrillation or who are susceptible to developing atrial fibrillation.
The Confirm RxTM ICM has not been specifically tested for pediatric use.
The Confirm Rx™ Insertable Cardiac Monitor (ICM) Model DM3500 is intended as a minimally invasive, implantable diagnostic monitoring device, with subcutaneous electrodes, looping memory, and automatic as well as patient-activated EGM storage capability, which help physicians monitor, diagnose, and document patients who are susceptible to cardiac arrhythmias. It has a two-year longevity, MR conditional labeling, sensing and detection algorithms, and Bluetooth communication. Specific features include:
- Patient-initiated triggering of EGM storage using the myMerlin™™ Patient App for mobile devices. This includes capability for the patient to identify symptoms, which are stored with the EGM for physician review.
- Automated triggering of EGM storage when tachycardia, bradycardia, or pauses are detected; with physician-programmable values for pause duration, bradycardia rate, tachycardia rate, and number of tachycardia intervals.
- Automated triggering of EGM storage when atrial fibrillation (AF) is detected, with physician programmable values for AF duration.
- The ability to identify EGM anomalies as a consequence of noise or vigorous activity and inhibit EGM storage as applicable.
- Remote care monitoring.
The provided document is a 510(k) premarket notification for the Abbott Confirm Rx™ Insertable Cardiac Monitor (ICM) System, Model DM3500, with updated firmware. The purpose of this submission is to demonstrate substantial equivalence to a predicate device (K182981).
The document details the device description, technological characteristics, and the testing conducted to support the substantial equivalence claim. However, it does NOT contain a detailed study proving the device meets specific acceptance criteria for AI/algorithm performance as would be typically found for new AI-driven diagnostic devices. Instead, the focus is on demonstrating that the updated firmware, particularly its detection algorithms, functions as intended without raising new issues of safety and effectiveness compared to the predicate device.
Specifically, the acceptance criteria and performance data are described in terms of verification and validation activities showing the device meets its "predetermined design and performance specifications" and "functions in accordance with product specifications." The changes in the firmware relate to "detection algorithms for the diagnosis of bradycardia, asystole (pause), and atrial fibrillation (AF)," including "second pass undersensing discriminators" and a "P-wave detection discriminator."
Given the information provided, it's not possible to present a table of acceptance criteria and reported device performance in the context of an AI study as you described.
However, I can extract the relevant information regarding the firmware update validation and
address the closest approximations to your questions for this type of medical device submission.
Analysis of the Provided Document Regarding Acceptance Criteria and Study:
The document describes an update to the firmware of an existing, already cleared device. The "study" here is the validation of the firmware update to ensure it doesn't negatively impact safety or effectiveness and that the new algorithms perform as intended.
1. A table of acceptance criteria and the reported device performance
The document states that the "Completion of all verification and validation activities demonstrated that the device with updated firmware meets its predetermined design and performance specifications and that the product is substantially equivalent to the predicate Confirm Rx™ ICM device (K182981)." It also states, "The results of the testing show that the candidate Confirm RxTM ICM performs as intended and is safe for its intended use."
The updated firmware includes changes to detection algorithms for bradycardia, asystole (pause), and atrial fibrillation (AF). The acceptance criteria would inherently be related to the accuracy and reliability of these detections compared to established benchmarks or the predicate device, although specific numerical performance metrics (e.g., sensitivity, specificity, accuracy percentages) are not detailed in this summary.
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Device meets predetermined design and performance specifications for detection algorithms (bradycardia, asystole, AF) including new discriminators. | Device with updated firmware meets its predetermined design and performance specifications. |
| Device performs as intended for its specified use (monitoring and diagnostic evaluation of arrhythmias). | Device performs as intended and is safe for its intended use. |
| Updated firmware does not raise new issues of safety and effectiveness compared to the predicate device. | The minor differences (firmware update) do not raise new issues of safety and effectiveness. |
| Maintenance of device longevity with updated firmware. | Accounted for the current drain and time used to execute the undersensing and P-wave discriminators within the battery longevity calculation. |
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
The document does not specify the sample size (number of patients, number of EGM recordings) used for the verification and validation (V&V) testing. It also does not specify the data provenance (country of origin, retrospective/prospective). This type of detail is typically found in the full testing report, not usually summarized in this section of a 510(k) summary.
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)
The document does not specify the number or qualifications of experts used to establish ground truth. For embedded device algorithm testing, ground truth might often be established through simulated signals, pre-recorded clinical data with confirmed diagnoses, or expert review of EGMs, but this detail is not provided.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
The document does not specify any adjudication method for ground truth establishment.
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, an MRMC comparative effectiveness study was not conducted or described. This is a device firmware update, not a new AI-assisted diagnostic tool where human reader performance would be compared. The focus is on the device's internal algorithm performance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the testing described appears to be a standalone validation of the algorithm's performance within the device, given the context of "Software/Firmware Verification and System Verification" and "Design Validation." The purpose is to confirm the algorithm's internal logic and detection capabilities. The document states:
- "Second pass undersensing discriminators added to the Asystole and Bradycardia detection algorithms to reject false Pause and Bradycardia detections."
- "Second pass P-wave detection discriminator to reject false AF detections."
This implies that the algorithm itself was tested for its ability to correctly identify and reject false detections, which is a standalone performance metric.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The type of ground truth is not explicitly stated. For cardiac rhythm detection algorithms, ground truth typically involves a highly accurate reference ECG/EGM interpretation, which could be established via:
- Expert Consensus: Review by multiple cardiologists or electrophysiologists.
- Manual Annotation: Beat-by-beat or episode-by-episode annotation of EGM data by trained personnel, often adhering to specific criteria.
- Validation against high-fidelity recordings: Comparing device output to a gold-standard recording from a different, validated system.
Given the focus on "reject[ing] false detections," it implies that there was a reference against which the algorithm's detection (true positive/false positive) was measured.
8. The sample size for the training set
The document does not provide any information about a training set size. This indicates that this submission is about validating a firmware update for an already cleared device, not seeking de novo clearance for a new machine learning algorithm that typically requires a distinct training phase. While the algorithms were likely "trained" or designed using data at some point in their development, that information is not part of this 510(k) summary for a firmware update.
9. How the ground truth for the training set was established
Since there is no mention of a training set in this document, there is no information on how its ground truth was established.
§ 870.2800 Medical magnetic tape recorder.
(a)
Identification. A medical magnetic tape recorder is a device used to record and play back signals from, for example, physiological amplifiers, signal conditioners, or computers.(b)
Classification. Class II (performance standards).