Ovulation Detection as a Digital Biomarker

Ovulation detection is the identification of the point in the menstrual cycle at which ovulation occurred, and by extension the classification of days into follicular and luteal phases. The physiological basis is well established: progesterone rises after ovulation and produces a small sustained increase in body temperature that persists through the luteal phase.

Basal body temperature has been used to detect that shift for decades, requiring an oral measurement immediately on waking before any movement, which is demanding enough that adherence has always been the limiting factor. Continuous wrist temperature removes that burden entirely by measuring through the night.

This is the one entry in this library where the digital measure substitutes for the manual method rather than complementing an instrument. It has no questionnaire counterpart, and that is the point: it replaces a self reported diary rather than sitting alongside one.

A wrist worn sensor records skin temperature continuously overnight, and an algorithm looks for the sustained shift in the nocturnal baseline that follows ovulation. Because the shift is a few tenths of a degree, it is detectable only against a stable personal baseline built over previous cycles, not from a single night.

Research has compared wrist skin temperature against basal body temperature for ovulation detection directly, and earlier work demonstrated that wrist wearables capture the temperature changes associated with the cycle. Multi parameter approaches add resting heart rate, heart rate variability and respiratory rate, all of which shift with cycle phase, and machine learning models combining them have been used to predict the fertile window and menstruation.

Detection is usually retrospective. Confirming ovulation after it has happened is considerably more reliable than predicting it in advance.

The research uses are the defensible ones. Phase classification lets a study control for cycle phase when interpreting other wearable measures, which is the argument made throughout this category. It supports research into cycle disorders and into conditions where ovulation is disrupted, and it provides an objective marker in trials where restoring ovulatory function is the outcome.

This page deliberately does not address consumer fertility use. Using wearable temperature to decide when to try to conceive, and especially to avoid conceiving, is a different claim carrying a different regulatory bar, and consumer products in this space have repeatedly presented algorithmic predictions with more confidence than their validation supports.

The distinction is not pedantic. A method that is adequate for classifying phase in a research analysis can be entirely inadequate for a decision with a pregnancy at stake.

Wrist temperature is skin temperature, so everything that affects it affects this measure: ambient room temperature, bedding, alcohol, illness, disrupted sleep and shift work. Alcohol in particular reliably distorts the nocturnal temperature profile.

Detection requires several cycles of baseline before it becomes reliable, which rules out short studies, and it is retrospective, so it confirms rather than predicts.

Anovulatory cycles produce no temperature shift, and an absent detection can mean either no ovulation or a failure to detect it, which are very different findings and cannot be distinguished from the signal alone. Hormonal contraception removes the shift entirely. And no wearable in this library is cleared as a contraceptive or conception aid, which is a statement about scope rather than about accuracy.

• Shilaih M, et al. Modern fertility awareness methods: wrist wearables capture the changes in temperature associated with the menstrual cycle. Biosci Rep. 2018. pubmed.ncbi.nlm.nih.gov
• Zhu TY, et al. The accuracy of wrist skin temperature in detecting ovulation compared to basal body temperature. J Med Internet Res. 2021. pubmed.ncbi.nlm.nih.gov
• Goodale BM, et al. Wearable sensors reveal menses-driven changes in physiology and enable prediction of the fertile window. J Med Internet Res. 2019. pubmed.ncbi.nlm.nih.gov
• Luo C, et al. Prediction of the fertile window and menstruation with a wearable device via machine-learning algorithms. Reprod Biomed Online. 2025. pubmed.ncbi.nlm.nih.gov