Accuracy of an Overnight Axillary-Temperature Sensor for Ovulation Detection: Validation in 194 Cycles

Yaniv Shpaichler; Alicia Thompson; Benedicte Fromager; Michael Vardi; Rene Ecochard

doi:10.3390/s25206327

Fertility Awareness · Effectiveness

Accuracy of an Overnight Axillary-Temperature Sensor for Ovulation Detection: Validation in 194 Cycles

Shpaichler Y, Thompson A, Fromager B, Vardi M, Ecochard R

Published October 13, 2025 Sensors, 25(20), 6327

DOI 10.3390/s25206327 PMC PMC12567647

RRM Academy Synopsis

Armpit Temperature Sensor Matched Urine Ovulation Tests in 194 Cycles

The Tempdrop armband identified fertile days with 96.8% sensitivity and 99.1% specificity in 194 cycles from 125 women in the United States. Urine ovulation tests were the reference. The armband's ovulation day fell within two days of the urine-test day in about 93 out of 100 cycles. Two authors work for the device maker.

Key Findings

The armband's estimated ovulation day fell within two days of the urine-test day in 181 of 194 cycles (93.3%, 95% CI: 89–96%).
The armband's ovulation day averaged 1.02 (s.d. 1.7) days after the LH surge day. Seven cycles had a delayed estimate and six an early one.
Across 5394 days, the armband flagged 1138 fertile days correctly and missed 38. It flagged 38 non-fertile days as fertile.
Sensitivity was 96.8% (95% CI 95.6; 97.7) and specificity 99.1% (98.8; 99.4). Accuracy was 98.6% (98.2; 98.9).
Positive predictive value was 96.8% (95.6; 97.7) and negative predictive value 99.1% (98.8; 99.4).

Interpretation

The study is a retrospective validation. The algorithm ran after each cycle ended, and the test cycles were separate from its training data. The reference was a urine LH test, so the results describe agreement with that test. Of 258 women recruited, 125 met the criteria, which required cycle length of 24 to 35 days and temperatures on at least 80% of days. The authors name the lack of prospective evaluation as the main limitation. Two authors work for Tempdrop Ltd. The authors suggest the temperature curve could help evaluate the luteal phase. This study did not test that.

RRM Context

Fertility awareness methods read a temperature rise as a sign that ovulation happened. Serial ultrasound or a pregnancy confirms that an egg was released. Cycle-charting-informed care starts from daily data on the ovulatory cycle. Wearable sensors offer one way to gather the temperature record. The authors suggest that a high LH sometimes seen days before ovulation might explain the rare disagreements between the two markers.

Our editorial summary of this paper, not the article's abstract.

Abstract

Several studies have evaluated the reliability of using temperature sensors placed in different locations on the body to identify the day of ovulation. However, such demonstrations are lacking for axillary temperature wearable devices. This study aimed to evaluate the accuracy with which an axillary temperature armband sensor (Tempdrop) identifies the day of ovulation and the fertile window, using the Clearblue Connected Ovulation Test System as the reference method. A total of 194 cycles were analyzed from 125 women that participated in the study between April 2023 and June 2024. The performance parameters were high: the sensitivity (96.8% (95% CI 95.6; 97.7)), specificity (99.1% (98.8; 99.4)), accuracy (98.6% (98.2; 98.9)), positive predictive value (96.8% (95.6; 97.7)) and negative predictive value (99.1% (98.8; 99.4)). Furthermore, the results revealed a remarkably clear and better-than-expected change in temperature around the time of ovulation. This axillary temperature wearable sensor is an effective alternative to urine ovulation tests for determining the timing of ovulation. Another advantage is that it provides a clear temperature curve that can be used to evaluate the quality of the luteal phase.

Topics

Fertility Awareness › Effectiveness › Comparative Effectiveness · Research Methods › Measurement and Statistics › Instrument Development and Validation · Menstrual Cycle › Cycle Biomarkers › Basal Body Temperature

Yaniv Shpaichler, Alicia Thompson, Benedicte Fromager, Michael Vardi, Rene Ecochard

Y Shpaichler, A Thompson, B Fromager, Mike Vardi, M Vardi, R Ecochard

DOI 10.3390/s25206327 10.3390/s25206327 Shpaichler et al. 2025, Shpaichler 2025

Accuracy of an Overnight Axillary-Temperature Sensor for Ovulation Detection: Validation in 194 Cycles

Armpit Temperature Sensor Matched Urine Ovulation Tests in 194 Cycles

Key Findings

Interpretation

RRM Context

Abstract

Topics

By this author

Related research