Identifying one's fertility window is critical for many women trying to get pregnant – or for those wanting to avoid pregnancy – but it can be a complex, inconvenient and often inaccurate process. Inspired by personal experience, Vanessa Xi, founder of Silicon Valley-based Yono Labs, has developed a hearing-aid-like wearable which takes the data collection and analysis off your hands, while adding significant improvements in accuracy.
The detection of ovulation can be undertaken in a number of ways – blood progesterone level, endometrial biopsy, ultrasound, ovulation predictor test and by charting basal body temperature (BBT), the last two being the most commonly used methods. BBT is the lowest body temperature reached over a 24-hour period, and most commonly occurs between 2 am and 6 am. Ovulation is associated with a 0.5° to 1° F (0.3° - 0.6° C) rise in BBT, and it's this slight rise, charted carefully, daily, over a long period which is used to flag one's fertility window.
The problem is, traditional methods of measuring BBT are often inconvenient, incomplete and prone to error. To get an accurate picture of one's fertility window, BBT has to be measured at the same time, every morning, before moving, before getting out of bed, before anything. Add to this the difficulty of consistently getting a reliable reading from skin-contact thermometers, the need to record your readings in the same place and the fact that you then need to analyze your data yourself, and you can see why the Yono was invented.
Vanessa Xi was determined to simplify this process, and using knowledge derived from statistics and machine learning, she and her team developed a two-part solution. The first part is the Yono itself, a small ear-bud-sized sensor which takes 70 to 120 readings throughout the night – as opposed to the single data point of the traditional method. These core-body temperature readings are far more accurate than skin-contact thermometers, which can be thrown off by ambient temperatures. The device keeps all of this data onboard, eliminating the need for a Wi-Fi or Bluetooth connection.
The second part of the solution is the Yono base station, which syncs the data with the Yono smartphone app (iOS only at this stage). The app employs machine learning algorithms to plot a monthly fertility chart, which in turn, predicts the fertility window.
The accuracy of the Yono is set to improve even further, with the company stating that it's working with testing users to train its algorithms to make more accurate predictions. The next step in the development of those algorithms is in the hands of Peter Song, a professor of biostatistics at the University of Michigan. The Yono collects a lot of data, but it can be a little noisy as readings can be affected by head movements or if the earbud falls out during the night.
In a paper published in the journal IEEE Transactions on Biomedical Engineering, Song shows how his algorithms filter out unhelpful readings and identify the most relevant data points. "For example, we got rid of data below 32° C, because that's not biologically possible," says Song, who hopes this work can benefit other wearable data-gathering products. "We want to put the data to better use, so we can build a little more intelligence into these devices."
The Yono is available from US$145.99 and is FSA (Flexible Spending Account)/HSA (Health Savings Account)-eligible. International shipping is unavailable. You can watch the Yono explanatory video below.
Sources: Yono Labs/IEEE Spectrum
