K Number
K101801
Device Name
MONICA AN24
Manufacturer
Date Cleared
2011-02-03

(220 days)

Product Code
Regulation Number
884.2720
Panel
OB
Reference & Predicate Devices
N/A
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Monica AN24 is an intrapartum maternal-fetal monitor that non-invasively measures and displays fetal heart rate (FHR) and uterine activity (UA). The AN24 acquires and displays the FHR tracing from abdominal surface electrodes that pick up the fetal ECG (fECG) signal. Using the same surface electrodes, the AN24 also acquires and displays the UA tracing from the uterine electromyography (EMG) signal. The AN24 is indicated for use on women who are at term (>36 completed weeks), in labor, with singleton pregnancies, using surface electrodes on the maternal abdomen.

The AN24 maternal-fetal monitor is intended for use by healthcare professionals in a clinical setting.

Device Description

The Monica AN24™ is a small, battery-powered device for L&D surveillance of fetal well-being. The AN24™ is designed to passively monitor Fetal Heart Rate (FHR) from the fetal electrocardiogram (fECG) and Uterine Activity (UA) from the Electrohysterogram (EHG) during pregnancy and can be used at any time from > 36 completed weeks gestation in laboring patients. The AN24™ is suitable for singleton pregnancies only.

The Monica AN24 is a small, battery powered electrophysiological monitor (specifically fetal ECG and uterine EMG). The electrical signals are passively monitored on three channels using five electrodes placed on the pregnant abdomen in specific locations. From these electrical signals the Fetal Heart Rate (FHR) and Uterine Activity (UA) are continuously extracted and displayed in the same standard format as the predicate device.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study details for the Monica AN24 device, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly stated as distinct thresholds in the document, but rather implied by the non-inferiority comparison to the predicate device and the reported outcomes from the clinical study. The study aimed to demonstrate that the Monica AN24 is at least as accurate and reliable as the predicate device. Therefore, the "acceptance criteria" can be inferred as performance greater than or equal to the predicate device for each metric.

MetricImplied Acceptance Criteria (relative to Predicate)Reported Monica AN24 PerformanceReported Predicate Performance (Philips 50XM)
Fetal Heart Rate (FHR) Interpretability (PPA)≥ 72%85%72%
FHR Accuracy (RMS Error)≤ 12 bpm5 bpm12 bpm
Uterine Activity (UA) Interpretability (PPA)≥ 67%97%67%
UA Sensitivity Accuracy (Sensitivity for Contractions)≥ 55%89%55%
UA Timing Accuracy (Mean Timing Difference for Contractions)≤ 3.3 seconds lag2.5 seconds lag3.3 seconds lag

2. Sample Sizes and Data Provenance

  • Test Set (Clinical Study):
    • Total Subjects Enrolled: 60 women.
    • Evaluable Data Subjects: 34 subjects.
    • FHR Study Subjects: 33 women.
    • UA Study Subjects: 30 women.
    • Data Provenance: The study was conducted at two clinical sites: QHC, New York, and Temple University, Philadelphia. This indicates prospective data collection in the USA.
    • Training Set Sample Size: Not explicitly stated. The document describes a "clinical equivalence trial" and a "subsequent Multi-Reader-Multi-Case (MRMC) study," which are generally used for validation/testing.

3. Number of Experts and Qualifications for Ground Truth (Test Set)

  • Number of Experts: Four.
  • Qualifications of Experts: Board Certified Obstetricians. (Specific years of experience are not mentioned).

4. Adjudication Method (Test Set)

  • Uterine Activity (UA): The uterine activity data was "independently reviewed in a Multi-Reader Study by four Board Certified Obstetricians who independently indicated on randomized trace segments." The results were "averaged" to establish the ground truth for UA. This aligns with a form of consensus-based adjudication, likely an average or 4/4 agreement model, though the exact decision rule isn't specified beyond "average of 4 board certified obstetricians."
  • Fetal Heart Rate (FHR): The ground truth for FHR was established directly by the "Direct fetal Scalp ECG (second Phillips Series 50XM device)," which served as the gold standard. No expert adjudication was mentioned for FHR.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

  • Was an MRMC study done? Yes, a "Multi-Reader-Multi-Case (MRMC) study" was integral to the evaluation of Uterine Activity (UA).
  • Effect Size of Human Readers' Improvement with AI vs. without AI assistance: The document focuses on the performance of the device itself (Monica AN24 or predicate) compared to the gold standard, with human readers involved in establishing the ground truth for UA and in reviewing "False Positives." It does not describe a comparative effectiveness study where human readers interpret tracings with and without AI assistance to measure reader improvement. The MRMC study in this context refers to multiple readers evaluating the output of different devices (Monica AN24, TOCO, IUPC) to derive ground truth and compare device performance.

6. Standalone (Algorithm Only) Performance Study

  • Yes, the study primarily assessed the standalone performance of the Monica AN24 device (algorithm only), comparing its output directly against established gold standards (fetal scalp ECG for FHR, IUPC for UA). The metrics (PPA, RMS error, sensitivity, timing accuracy) are all measures of the device's inherent performance.

7. Type of Ground Truth Used

  • Fetal Heart Rate (FHR): Direct fetal Scalp ECG (from a second Philips Series 50XM device). This is a direct physiological measurement considered the gold standard.
  • Uterine Activity (UA): Intrauterine Pressure Catheter (IUPC) (from a Philips Series 50XM device), with expert consensus from four Board Certified Obstetricians used to interpret the tracings and mark contractions. This is a combination of direct physiological measurement and expert consensus/adjudication.

8. Sample Size for the Training Set

  • The document does not explicitly state the sample size used for training the Monica AN24 device's algorithms. The described study is a clinical equivalence trial used for validation of the device's performance against a gold standard and a predicate, not for training.

9. How Ground Truth for the Training Set was Established

  • The document does not provide information on how ground truth was established for the training set, as the focus is on the clinical validation study for regulatory submission.

§ 884.2720 External uterine contraction monitor and accessories.

(a)
Identification. An external uterine contraction monitor (i.e., the tokodynamometer) is a device used to monitor the progress of labor. It measures the duration, frequency, and relative pressure of uterine contractions with a transducer strapped to the maternal abdomen. This generic type of device may include an external pressure transducer, support straps, and other patient and equipment supports.(b)
Classification. Class II (performance standards).