HMM-based estimation of menstrual cycle from skin temperature during sleep

Ultradian rhythms in heart rate variability and distal body temperature anticipate onset of the luteinizing hormone surge

The menstrual cycle is characterized by predictable patterns of physiological change across timescales. Although patterns of reproductive hormones across the menstrual cycle, particularly ultradian rhythms, are well described, monitoring these measures repeatedly to predict the preovulatory luteinizing hormone (LH) surge is not practical. In the present study, we explored whether non-invasive measures coupled to the reproductive system: high frequency distal body temperature (DBT), sleeping heart rate (HR), sleeping heart rate variability (HRV), and sleep timing, could be used to anticipate the preovulatory LH surge in women. To test this possibility, we used signal processing to examine these measures in 45 premenopausal and 10 perimenopausal cycles alongside dates of supra-surge threshold LH and menstruation. Additionally, urinary estradiol and progesterone metabolites were measured daily surrounding the LH surge in 20 cycles. Wavelet analysis revealed a consistent pattern of DBT and HRV ultradian rhythm (2-5 h) power that uniquely enabled anticipation of the LH surge at least 2 days prior to its onset in 100% of individuals. Together, the present findings reveal fluctuations in distal body temperature and heart rate variability that consistently anticipate the LH surge, suggesting that automated ultradian rhythm monitoring may provide a novel and convenient method for non-invasive fertility assessment.

Thermometry and interpretation of body temperature

This article reviews the historical development and up-to-date state of thermometric technologies for measuring human body temperature (BT) from two aspects: measurement methodology and significance interpretation. Since the first systematic and comprehensive study on BT and its relation to human diseases was conducted by Wunderlich in the late 19th century, BT has served as one of the most fundamental vital signs for clinical diagnosis and daily healthcare. The physiological implication of BT set point and thermoregulatory mechanisms are briefly outlined. Influential determinants of BT measurement are investigated thoroughly. Three types of BT measurement, i.e., core body temperature, surface body temperature and basal body temperature, are categorized according to its measurement position and activity level. With the comparison of temperature measurement in industrial fields, specialties in technological and biological aspects in BT measurement are mentioned. Methodologies used in BT measurement are grouped into instrumental methods and mathematical methods. Instrumental methods utilize results of BT measurements directly from temperature-sensitive transducers and electronic instrumentations by the combination of actual and predictive measurement, invasive and noninvasive measurement. Mathematical methods use several numerical models, such as multiple regression model, autoregressive model, thermoregulatory mechanism-based model and the Kalman filter-based method to estimate BT indirectly from some relevant vital signs and environmental factors. Thermometry modalities are summarized on the dichotomies into invasive and noninvasive, contact and noncontact, direct and indirect, free and restrained, 1-D and n-D. Comprehensive interpretation of BT has an equal importance as the measurement of BT. Two modes to apply BT are classified into real-time applications and long-term applications. With rapid advancement in IoT infrastructure, big data analytics and AI platforms, prospects for future development in thermometry and interpretation of BT are discussed.