Predicting labor onset relative to the estimated date of delivery using smart ring physiological data

Abstract The transition from pregnancy into parturition is physiologically directed by maternal, fetal and placental tissues. We hypothesize that these processes may be reflected in maternal physiological metrics. enrolled pregnant participants the third-trimester ( n = 118) to study continuously worn smart ring devices monitoring heart rate, rate variability, skin temperature, sleep physical activity negative temperature coefficient, 3-D accelerometer infrared photoplethysmography sensors. Weekly surveys assessed labor symptoms, pain, fatigue mood. estimated association between each metric, gestational age, likelihood of a participant’s beginning prior (versus after) clinical delivery date (EDD) 40.0 weeks with mixed effects regression. A boosted random forest was trained on metrics predict pregnancies naturally passed EDD versus undergoing onset EDD. Here we report many raw sleep, activity, symptom are correlated age. As age advances, individuals have lower resting 0.357 beats/minute/week, 0.84 higher variability (milliseconds) shorter durations sleep. Further, predictions determine would pass accuracy 0.71 (area under receiver operating curve). Self-reported symptoms correlate increased not timing (relative EDD) or spontaneous labor. use ring-derived data improve prediction natural duration relative