Predicting Athleticism from Exercise Physiology Factors
We used machine learning techniques to predict athleticism using SPO2 measurements.
VO2 is a vital health measurement that can be used as an indicator to predict health conditions. However, VO2 measurement is complex and requires a lot of technical resources. Thus, it will be useful to predict VO2 from relatively easy to obtain metrics, such as age, weight, sex, and heart rate, especially in low resource areas where access to VO2 measurements is not as feasible. Literature suggests that there is a relationship between VO2 and cardiorespiratory fitness. In this project, our goal is to build a model to predict VO2 based on features that are easy to collect, and use that prediction to classify an individual's cardiorespiratory fitness into athletic or non-athletic categories. The categories were made based on the threshold value for athleticism we obtained from literature. Based on literature review, we believe that we will be able to create supervised learning models that use easy to obtain metrics to predict VO2, and thus athleticism. We can evaluate the success of our models by comparison to labels in our dataset. To make an actionable solution, we will perform cross validation on a variety of models.
To learn more: