Abstract

Lactate is a by-product of the body's metabolic process and lactate levels can be monitored to track metabolism rates. It can have useful applications especially in the health and sports industry, since lactate levels can be used to find an individual's ideal exercise level and detect medical conditions such as sepsis and heart failures.

Currently, the process of lactate testing is invasive; athletes need to have their ears or fingers pricked by lab technicians every five minutes. After interviewing UPenn Rowing's head coach and the US National team's head coach, we heard positive feedback about our non-invasive approach of lactate testing and are excited to share with them the progress we have made on this product.

Last semester, we had some issues with the sensor. We considered using an electrochemical sensor to detect lactate in the blood; however, we soon realized that there would be too much noise. Before classes went online, we instead planned to fabricate an electromagnetic sensor and test it with a VNA, but after the transition, we decided to develop a machine learning model which takes in gender, age, height, weight, blood glucose level, oxygen saturation level, and heart rate as inputs and predicts a lactate threshold. The prediction can be used to recommend workout routines based on the user's fitness goal.

Share this project:

Updates