The inspiration comes from two places. People want to know more and more about their body and about their activities throughout the day. A metric that hasn't been used before is perspiration. Sweat often has a negative connotation. It can provide many useful metrics to the user if they have the right instruments to measure it. One of the key things that measuring the perspiration rate can help is more accurately track the the user's hydration levels.

Secondly, people have to use more and more invasive methods to get accurate blood glucose measurements. Some people have to have it implanted under their skin to get continuous readings to help them monitor their blood glucose more carefully. This method of continuous measurements can cause a lot of issues and discomfort for the user. They can even fall off leading to an infection risk. If it could be easier to get an accurate blood glucose reading it would help a lot of people around the globe. Non-invasive methods have been extensively studied but they mainly use radio frequencies to measure the blood glucose levels. This really started our search for other non-invasive methods for measuring blood glucose.

What it does

It is a watch that enables you to finally know how hard you really worked in that last workout. Paired with your other wearables such as a Garmin Fenix 6 pro or other similar watch, you can finally get the full picture of how well you did and even a glimpse into metrics such as blood glucose and perspiration rate. Keeping track of your blood glucose levels can be helpful to anyone even those who don't have diabetes. If can even help prevent diabetes as whole. The are quite a few studies on non-invasive methods of measuring blood glucose. We focused on two studies that outlined the methods we use to measure glucose from sweat. The links are posted below in the sources section. These papers provided the strong background to our idea. One of the papers gave the equation that we used to get the blood glucose measurements. This came from the 5th paper in the sources.


How we built it

We used an ESP32 S2 QTPY from Adafruit for the brains of the watch which also allows for connectivity. It also has built-in touch support. This makes it easier to measure small changes in voltage. We use two pogo pins to ensure gentle contact with the skin at all times to measure conductance of the skin reliably. Secondly, we use two high quality environmental sensors to ensure accuracy of the readings. These sensors are the BME688 and BME280. These both have the same accuracy which is vital for a task like this. The case for the electronics is 3d printed so we could have a small enough case.


We used CockroachDB to store all of our data from the Garmin device.


We use Lambda Function

Mobile App

We used React Native to create a dashboard for the user to see all of their metrics all in one place. While most wearables come with a companion app, they lack the additional data we provide.

Challenges we ran into

The biggest challenge that we ran into is the sensors that we initial picked for their small size didn't work as intended. Luckily, we have other sensors on hand that fit the bill so with quickly switched to use those. We had to redesign the enclosure for them though so that took some time away from creating other features.

Accomplishments that we're proud of

We are proud that were we able to pack a lot capability and power into a small size. The ESP32 that we are using runs at 240 Mhz but it's the size of a postage stamp.


Built With

Share this project: