K Number
K231274
Device Name
Natural Cycles
Date Cleared
2023-08-24

(114 days)

Product Code
Regulation Number
884.5370
Panel
OB
Reference & Predicate Devices
Predicate For
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

Natural Cycles is a stand-alone software application, intended for women 18 years and older, to monitor their fertility. Natural Cycles can be used for preventing a pregnancy (contraception) or planning a pregnancy (conception).

Device Description

Natural Cycles is an over-the-counter web and mobile-based standalone software application that monitors a woman's menstrual cycle using information entered by the user and informs the user about her past, current, and future fertility status. The following information is used by the Natural Cycles software:

  • daily body temperature measurements
  • information about the user's menstruation cycle (i.e., start date, number of days)
  • optional ovulation or pregnancy test results
    A proprietary algorithm evaluates the data and returns the user's fertility status.
    Natural Cycles is available in three modes: Contraception (NC° Birth Control), Conception (NC° Plan Pregnancy), and Pregnancy (NCº Follow Pregnancy). For NCº Birth Control mode, the device provides predictions of "not fertile," shown as green days, and "use protection," shown as red days, that allow the user to determine the days on which her risk of conception is highest, and then make choices about either abstaining from sex or using a barrier method of contraception to prevent pregnancy.
    In addition to measuring daily basal body temperature with an oral thermometer with two decimal points, the predicate submission cleared the Oura Ring for automatic temperature input to the Natural Cycles algorithm, and the current submission expands to allow the device to utilize automatic temperature inputs from the Apple Watch.
    Natural Cycles can be used by women 18 years and older. Women who have been on hormonal birth control within 60 days prior to using Natural Cycles have a higher risk of becoming pregnant when compared to women who have not been on hormonal birth control within the 12 months prior to using the device may not be right for women who have a medical condition where pregnancy would be associated with a significant risk to the mother or the fetus.
AI/ML Overview

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) summary:

This device, Natural Cycles, is a software application for contraception. The 510(k) submission seeks to add the Apple Watch as an accepted source for daily basal body temperature (BBT) input, to be used by the existing Natural Cycles fertility algorithm. The core algorithm remains unchanged.

1. Table of Acceptance Criteria and Reported Device Performance

The core acceptance criterion for this submission revolves around the comparability of the algorithm's performance when using Apple Watch temperature data versus a traditional oral thermometer, specifically in the context of identifying ovulation and providing "green days" (not fertile). The study did not define explicit acceptance criteria in terms of specific sensitivity/specificity thresholds, but rather focused on demonstrating that the Apple Watch input did not negatively impact the safety and effectiveness as compared to the predicate device.

Acceptance Criteria (Implicit)Reported Device Performance
Algorithm's ability to identify ovulation with Apple Watch temperature data is comparable to oral thermometer data."The results of the clinical study demonstrated that when temperature was inputted from either the Apple Watch or the two-decimal place oral thermometer, the Natural Cycles algorithm was able to identify that ovulation had occurred, which was confirmed in the study by comparison to the 153 positive LH test results."
Use of Apple Watch temperature data does not increase the risk of unintended pregnancy (e.g., by incorrectly assigning "green days" in the fertile window)."Compared to the two-decimal place oral thermometer, the Natural Cycles algorithm provided 0.6 fewer green days (not fertile) in the luteal phase of the menstrual cycle when the input temperature was from the Apple Watch without increasing the risk of unintended pregnancy." (Note: A reduction in "green days" in the luteal phase, post-ovulation, would generally be considered a more cautious or neutral outcome in terms of pregnancy risk, as the luteal phase, while generally infertile, should not be incorrectly identified as fertile.)
The change in temperature input source does not raise different questions of safety and effectiveness.The summary concludes: "A comparison of intended use and technological characteristics combined with performance data demonstrates that Natural Cycles is as safe and effective as the predicate device and supports a determination of substantial equivalence." The 0.6 fewer green days with Apple Watch vs. 1.6 additional green days with Oura Ring (predicate) are noted, but explicitly stated: "however, this difference does not impact the safety and effectiveness of Natural Cycles for its intended use." This implicitly means the performance variations were within acceptable safety margins for contraceptive effectiveness.

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

  • Sample Size:
    • 272 users (de-identified information) met criteria for the main study.
    • A subset of 104 users was assessed for the ovulation Key Performance Indicator (KPI), requiring a positive luteinizing hormone (LH) test.
    • Data were collected over a total of 1918 menstrual cycles.
    • 505 of these cycles were considered complete and met study criteria.
    • 153 of the complete cycles had at least one positive LH test.
  • Data Provenance:
    • Country of Origin: The majority (58%) of the women were located in the European Union (EU) and the United Kingdom (UK), with the remainder (42%) located in the US.
    • Retrospective or Prospective: Prospective. The study involved women actively wearing the Apple Watch nightly, recording temperatures, and providing LH test results over time.

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

This summary does not specify the number or qualifications of experts used. The ground truth for ovulation was established through Luteinizing Hormone (LH) test results. These are over-the-counter medical tests used by individuals to detect the LH surge that precedes ovulation, and their interpretation is generally self-reported by the user.

4. Adjudication Method for the Test Set

No explicit adjudication method is mentioned. The ground truth appears to be based on user-reported LH test results, which are biochemical markers rather than subjective expert interpretations requiring adjudication. The comparison was statistical, between the algorithm's output using Apple Watch data versus oral thermometer data, benchmarked against LH test results.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, an MRMC study was not performed. This device is a standalone software application, and the study focused on the performance of the algorithm with different temperature inputs, not on human reader performance or improvement with AI assistance.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) Was Done

Yes, this was effectively a standalone algorithm performance evaluation. The study compared how the Natural Cycles algorithm performed in identifying ovulation and predicting "green days" when fed with temperatures from the Apple Watch versus an oral thermometer. The algorithm's output was then confirmed against LH test results. Humans were involved in providing the data (wearing the watch, taking oral temperature, performing LH tests) but not in interpreting images or making diagnoses that the AI would then assist with.

7. The Type of Ground Truth Used

The primary ground truth used for validating the ovulation detection performance was Luteinizing Hormone (LH) test results. This is a form of biochemical marker or outcomes-related data (in the sense that an LH surge is a biological event preceding ovulation).

8. The Sample Size for the Training Set

The document does not explicitly state the sample size for the training set. It refers to the Natural Cycles algorithm and states "There have been no changes to how the Natural Cycles algorithm determines the daily fertility status." This implies the algorithm was already trained and validated as part of prior 510(k) submissions (e.g., K202897, the predicate device). The current study is an evaluation of new input data rather than a re-training effort.

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

Since the training set details are not provided in this specific document (as the algorithm itself wasn't re-trained for this submission), this information cannot be definitively extracted from the provided text. However, given that Natural Cycles has been cleared for contraception previously, its algorithm would have been extensively validated with large datasets of menstrual cycle data, BBT readings, and confirmed ovulation events (likely through similar biochemical markers like LH, or possibly other clinical methods).

§ 884.5370 Software application for contraception.

(a)
Identification. A software application for contraception is a device that provides user-specific fertility information for preventing a pregnancy. This device includes an algorithm that performs analysis of patient-specific data (e.g., temperature, menstrual cycle dates) to distinguish between fertile and non-fertile days, then provides patient-specific recommendations related to contraception.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Clinical performance testing must demonstrate the contraceptive effectiveness of the software in the intended use population.
(2) Human factors performance evaluation must be provided to demonstrate that the intended users can self-identify that they are in the intended use population and can correctly use the application, based solely on reading the directions for use for contraception.
(3) Software verification, validation, and hazard analysis must be performed. Documentation must include the following:
(i) A cybersecurity vulnerability and management process to assure software functionality; and
(ii) A description of the technical parameters of the software, including the algorithm used to determine fertility status and alerts for user inputs outside of expected ranges.
(4) Labeling must include:
(i) The following warnings and precautions:
(A) A statement that no contraceptive method is 100% effective.
(B) A statement that another form of contraception (or abstinence) must be used on days specified by the application.
(C) Statements of any factors that may affect the accuracy of the contraceptive information.
(D) A warning that the application cannot protect against sexually transmitted infections.
(ii) Hardware platform and operating system requirements.
(iii) Instructions identifying and explaining how to use the software application, including required user inputs and how to interpret the application outputs.
(iv) A summary of the clinical validation study and results, including effectiveness of the application as a stand-alone contraceptive and how this effectiveness compares to other forms of legally marketed contraceptives.