Women's basal body temperature (BBT) shows a periodic pattern that associates with menstrual cycle. Although this fact suggests a possibility that daily BBT time series can be useful for estimating the underlying phase state as well as for predicting the length of current menstrual cycle, little attention has been paid to model BBT time series. In this study, we propose a state‐space model that involves the menstrual phase as a latent state variable to explain the daily fluctuation of BBT and the menstruation cycle length. Conditional distributions of the phase are obtained by using sequential Bayesian filtering techniques. A predictive distribution of the next menstruation day can be derived based on this conditional distribution and the model, leading to a novel statistical framework that provides a sequentially updated prediction for upcoming menstruation day. We applied this framework to a real data set of women's BBT and menstruation days and compared prediction accuracy of the proposed method with that of previous methods, showing that the proposed method generally provides a better prediction. Because BBT can be obtained with relatively small cost and effort, the proposed method can be useful for women's health management. Potential extensions of this framework as the basis of modeling and predicting events that are associated with the menstrual cycles are discussed. © 2017 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.
This paper proposes a method to estimate a woman's menstrual cycle based on the hidden Markov model (HMM). A tiny device was developed that attaches around the abdominal region to measure cutaneous temperature at 10-min intervals during sleep. The measured temperature data were encoded as a two-dimensional image (QR code, i.e., quick response code) and displayed in the LCD window of the device. A mobile phone captured the QR code image, decoded the information and transmitted the data to a database server. The collected data were analyzed by three steps to estimate the biphasic temperature property in a menstrual cycle. The key step was an HMM-based step between preprocessing and postprocessing. A discrete Markov model, with two hidden phases, was assumed to represent higher- and lower-temperature phases during a menstrual cycle. The proposed method was verified by the data collected from 30 female participants, aged from 14 to 46, over six consecutive months. By comparing the estimated results with individual records from the participants, 71.6% of 190 menstrual cycles were correctly estimated. The sensitivity and positive predictability were 91.8 and 96.6%, respectively. This objective evaluation provides a promising approach for managing premenstrual syndrome and birth control.
An HMM-based method is proposed to estimate biphasic property in female menstrual cycle. A tiny device is developed to measure skin temperature change during sleep. Data are collected from 30 female participants for 6 months. Raw data are preprocessed to remove obvious outliers and clamped between 34 and 42 degree Celsius. A two hidden states HMM-based algorithm was applied to estimate the biphasic property in menstrual cycle. The results showed that the number of correctly detected menstrual cycle is 159 among 173 in 30 participants during 6 months. Overall sensitivity reaches 92.0%.
The menstrual cycle is divided into hypothermic and hyperthermic phases based on the periodic shift in the basal body temperature (BBT), reflecting events occurring in the ovary. In the present study, we proposed a state-space model that explicitly incorporates the biphasic nature of the menstrual cycle, in which the probability density distributions for the advancement of the menstrual phase and that for the BBT switch depending on a latent state variable. Our model derives the predictive distribution of the day of the next menstruation onset that is adaptively adjusted by accommodating new observations of the BBT sequentially. It also enables us to obtain conditional probabilities of the woman being in the early or late stages of the cycle, which can be used to identify the duration of hypothermic and hyperthermic phases, possibly as well as the day of ovulation. By applying the model to real BBT and menstruation data, we show that the proposed model can properly capture the biphasic characteristics of menstrual cycles, providing a good prediction of the menstruation onset in a wide range of age groups. The application of the proposed model to a large data set containing 25 622 cycles provided by 3533 women further highlighted the between-age differences in the population characteristics of menstrual cycles, suggesting its wide applicability. KEYWORDS menstrual cycle length (MCL), ovulation, periodic phenomena, phase identification, sequential Bayesian filtering and prediction Statistics in Medicine. 2019;38:2157-2170.wileyonlinelibrary.com/journal/sim
In this study, we focused on the uctuation of the abdominal skin temperature (AST) during sleep as a second marker for determining the biphasic menstrual cycle, alongside the basal body temperature. The nocturnal AST was measured every 10 min using a wearable device mounted on the abdominal wall. With this system, the AST time-series data were recorded for a total of 1667, 1035, and 1690 days from seven participants for the menstrual/follicular, ovulatory, and luteal phases, respectively. First, the AST uctuation was evaluated by plotting the cumulative probability distribution (CPD) of changes in AST every 10 min from 0 to 0.7 C. The results showed that the CPD tted well with an exponential attenuation curve. Second, the mean attenuation coef cients obtained by exponential regression from the CPD data were compared among the three phases. For regular menstrual cycles, the attenuation coef cient was the highest in the menstrual/follicular phase (8.57; 95% con dence interval 8.44-8.70; R 2 = 0.983; P < 0.001), followed by the ovulatory phase (7.80; 95% con dence interval 7.65-7.96; R 2 = 0.985; P < 0.001) and then the luteal phase (7.24; 95% con dence interval 7.12-7.36; R 2 = 0.985; P < 0.001). Finally, we examined whether the attenuation coef cients can be used as an index to classify the three phases by long short-term memory (LSTM)-based deep learning. Consequently, the attenuation coef cient affected the prediction of the menstrual/follicular, ovulatory, and luteal phases with signi cantly higher F-measures of 0.603, 0.328, and 0.660, respectively. These results suggest that the thermoregulatory system may increase the AST uctuation in healthy women during the transition from the follicular phase to the ovulatory phase and then to the luteal phase.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.