In today's fast-paced world, obesity and lack of physical activity have become growing concerns. With sedentary lifestyles, unhealthy diets, and increasing stress levels, many people struggle to maintain a healthy balance. These issues not only affect individual well-being but also put a strain on healthcare systems worldwide.
To raise awareness and help individuals assess their personal situation, we have developed a small web app. This tool allows users to quickly evaluate their own status quo regarding weight and fitness levels, providing them with insights and motivation to act towards a healthier lifestyle.
Methods
The data science application is based on Streamlit and offers an interactive and elegant web interface. First, the user data was queried with st.number_input(), whereby in some cases predefined categories were specified. Subplot from the matplotlib.pyplot package was used to display the body dimensions in a 6x2 grid, as well as the section below on habits and lifestyle. In the background, the NHANES data set was loaded from a csv file that we had cleaned up in advance. A subset was formed from the user input on gender and age. The histograms were created with the seaborn package and calculated according to fixed specifications. The user inputs were sorted into the corresponding bin as input_bin and displayed as plt.bar above the seaborn histogram. The text (e.g. on the smoker status) was inserted with plt.text() and the axes switched off so that a clear arrangement was achieved.
Results / Learnings
Throughout the development of our web app, we gained a deeper understanding of working with scientific datasets. Handling missing values, ensuring data integrity, and making informed decisions on data cleaning proved to be crucial challenges. Additionally, we learned how complex it is to create meaningful comparison groups that provide relevant and accurate benchmarks. And, of course, we experienced firsthand that working in a team can be exhausting—especially when trying to reach a consensus. However, through plenty of discussions (and some patience), we improved our communication skills and, in the end, are proud of the result.
Team & Rollen
Julian Repke
Involved in all steps of the project.
Samuel
Involved in all steps of the project.
Zubeda Rahmah Nankabirwa
Involved in all steps of the project.
Sebastian Fischer
Involved in all steps of the project.
Mentor:in
Julia Norget
