A vast amount of people around the world menstruate, but with such a common issue there still is not a menstrual tracking device that meets all user's needs. Current tracking devices require a lot of user input which can be difficult, especially for people with cognitive disabilities that may cause issues with memory. Another common issue with the current tracking devices is that they do not apply to people with irregular cycles. Data safety is another concern for people who want to track their cycles, especially with the overturning of Roe v. Wade.

We decided on a bracelet that measures three vital signs (skin conductivity, heart rate, and temperature) and analyzes them in a machine-learning model to predict menstrual cycles. We chose vital signs because they fluctuate according to what stage of the menstrual cycle a person is currently in for both regular and irregular menstrual cycles. The machine learning model will be able to adapt to each person and provide more accurate predictions as the device continues to be used.

The current prototype connects an Arduino UNO to three vital sign sensors that measure skin conductivity, heart rate, and temperature. All the sensors are attached to a silicon bracelet. The Arduino UNO stores the vital sign readings for heart rate, temperature, and skin conductivity, constantly updates the average values for each sensor, and stores the present-day value as a different variable. These six values are sent to the Raspberry Pi via bidirectional serial communication. The machine learning model is deployed to the Raspberry Pi and executes the Decision Tree model to predict days until the next cycle start date, which is displayed on an LED screen.