Typically, if women want to know if they're ovulating, they have to perform either a urine test or a basal body temperature analysis. Scientists at Harvard-affiliated Brigham and Women's Hospital are developing what they believe may be a better alternative, in the form of a smartphone-based saliva test.

According to the researchers, the cost of single-use home testing kits can add up over time, plus it may be difficult for users to interpret the results. By contrast, their experimental new system can be reused indefinitely, and it displays unambiguous results on the smartphone's screen.

To use it, women smear a saliva sample across a glass slide. If that woman is ovulating, then as the sample air-dries, it will crystallize into a structure that resembles fern leaves – the process is known as salivary ferning. If an untrained layperson tries checking the dried sample, though, they can easily misinterpret any patterns that may be present.

For that reason, the slide is slotted into a 3D-printed optical device which is in turn placed over a smartphone's camera. An artificial intelligence-based app then visually analyzes the sample, determining if ferning is indeed occurring. That app was trained using over 1,500 images of both ovulating and non-ovulating saliva samples.

When tested on samples from six women – obtained during both ovulating and non-ovulating phases of their menstrual cycle, as determined by urine testing – the system was 99 percent accurate at identifying ovulation, and 100 percent accurate at identifying non-ovulation. That said, as is the case with other forms of female fertility testing, the technology currently doesn't work on samples from women with estrogen imbalances, cysts in the ovaries, or who are taking fertility medication.

Additional testing on a larger population will be required before the system can be commercialized.

"When we published last year on a technology for analyzing sperm to detect male infertility, we were approached by those who had read about our work and were wondering if we could develop a smartphone-based system to provide ovulation testing at home," says lead scientist Dr. Hadi Shafiee. "Our study indicates that an accurate, automated and low-cost test is indeed possible."

A paper on the research was recently published in the journal Lab on a Chip.

View gallery - 2 images