An inexpensive smartphone-based device for point-of-care ovulation testing
Potluri et al.
The finding, in our words
The authors developed an inexpensive smartphone-based device that uses artificial intelligence to detect fern patterns in self-collected saliva samples, achieving over 99% accuracy in predicting ovulation. This approach offers a low-cost, patient-friendly alternative to urine hormone tests and basal body temperature monitoring for natural family planning.
A paraphrase to the Library’s standard, never the abstract. The source is one link away and is always the authority.
Self-collected salivary progesterone and the ratio of estradiol to progesterone showed strong agreement with serum levels, supporting the use of saliva as a noninvasive matrix for monitoring the menstrual cycle.
In a study of 46 perinatal nicotine users, dried blood spots and saliva were used weekly to measure reproductive hormones, finding that lower postpartum estradiol and greater peripartum declines in oxytocin were associated with increased nicotine craving and use. This demonstrates the feasibility of decentralised microsampling and remote surveys for longitudinal hormone tracking in vulnerable populations.
Self-collected oral and stool samples revealed distinct microbiome compositions in patients with endometriosis compared to controls, with Fusobacterium enrichment specifically observed in oral samples from moderate to severe cases. These results support the feasibility of decentralised self-collection for non-invasive biomarker screening in reproductive health.
In a feasibility pilot of 29 perimenopausal women using daily saliva progesterone measured at home, a significant association was seen between short luteal phases and having more cycles under 23 days, but this did not remain after adjusting for observation count and no differences were found for other cycle characteristics, suggesting short luteal phases may not reliably indicate cycle irregularity in this population.
Only 24% of participants completed this two-cycle feasibility study of AI-interpreted salivary ferning for ovulation prediction in females with irregular cycles, with 44% withdrawing citing cycle irregularity, pregnancy, or time constraints. The authors concluded that decentralised fertility trials must streamline procedures and provide ovulatory health education to improve retention and reduce burden before smartphone-based diagnostics can be scaled for this population.