While men are more likely to be diagnosed with diabetes at a younger age, women face higher risks of diabetes as they reach 65 and up. This is primarily because after menopause, women lose the potency of estradiol, the primary form of estrogen, making them more vulnerable to insulin resistance and diabetes.  Given these hormonal differences, there is a need to better understand how sex differences affect diabetes.

Most datasets do not have labels to indicate if a women has a menstrual cycle or if she doesn’t have one. Despite this, previous research has established that other physiological metrics, particularly basal body temperature, have a strong correlation with menstrual cyclicity. The figure below shows that temperature follows a sinusoidal pattern that aligns with different phases of the menstrual cycle. Our group’s idea was to apply this concept of identifying cyclicity to label-free glucose data. We expected for there to be some influence of menstrual cycles on glucose levels.