Evaluation of the accuracy of wrist skin temperature measured using an infrared sensor for prediction ovulation date

Technological advancements have made it possible to measure vital signs through wearable devices. This study evaluated the accuracy of wrist skin temperature (WST) measured by a continuous monitoring biosensor in the Galaxy Watch 5 for predicting ovulation dates. We recruited 106 women to measure WST during sleep, oral temperature upon waking, and perform ovulation tests around expected ovulation dates. Ovulation was confirmed via progesterone blood test. We analyzed 225 cycles (85 women) in the full analysis set (FAS), and 156 cycles (52 women) in the per-protocol (PP) set. WST predicted ovulation dates within 2 days in 64.0% (FAS) and 64.1% (PP) of cycles, and within 3 days in 77.8% (FAS) and 78.9% (PP) of cycles. Next menstrual cycle prediction accuracy was 68.2% (FAS) and 69.9% (PP) within 2 days, and 77.8% (FAS) and 84.1% (PP) within 3 days. WST and oral temperature showed moderate correlation (FAS: r = 0.423; PP: r = 0.448). While predicting ovulation within 2 days wasn't statistically significant, prediction within 3 days showed significant results. In conclusion, WST measurement using a continuous monitoring biosensor during sleep may be helpful in predicting ovulation date and next menstruation dates, offering another useful application for wearable devices in reproductive health monitoring.