Fall detection app: the beginning of journey

The story contains the creation of data set, forming of an algorithm and the building of an Android app for fall detection.

Tomáš Repčík
3 min readSep 17, 2022
Main screens of the fall detection app — BeSafeBox
Overview of the image.

The first part of the series is about starting the whole project and giving a bit of the background. Feel free to skip to the next part if you are interested in the technicalities. 💻

How did I get to it?

I was finishing my biomedical bachelor's degree in 2018-2019. Firstly we had to get 3-month experience in some company related to medicine / biomedical engineering. Secondly, we had to write a thesis on the picked topic.

I was trying to combine my hobbyist passion for mobile development with my school, but with no success. Luckily, one of my teachers suggested reaching out to one company. I reached out to the creators of an app called Záchranka (in explicit translation Ambulance). Their app makes calling the emergency much faster and more precise with one button. Their app brings the most valuable asset, the ability to save a life. Visit them here. 🚑

At that time, they were not able to give me the job. Although, after lengthy discussions, we were able to come up with the idea for the app and create the topic for my thesis: Detection of a car accident and collapse by Android smartphones.

If you have issue to find the right idea, search for working people in area of your interest. They are much more eager to share ideas, then you think.

I will dedicate one post also for car accident detection, but fall detection had become the core of my work.

What was my starting point?

My study was not focused on programming purely, not to mention programming for Android or data science. I had rough principles of making working code and process signal, but nothing specific about the required tools to make such an app. I had no idea how to use Python tools to analyse data or create models. The same went for coding the Android app.

To overcome a lack of experience, I had to learn a lot about Android development and data science research. Spending time on Youtube, Stackoverflow, Udemy, and other sites has become daily bread. 📚

As you can assume, the app does not demonstrate current trends or best practices in mobile development. However, it provides insight into how to solve problems. I would bet every programmer would like to come back in time from time to time. Perfection is the enemy of good enough.

What is the result?

The main screen of the app

The process of the app development was:

  1. Definition of the problem
  2. Gathering of the data and analysis
  3. Creation of the algorithm/model
  4. Integration into the Android app
  5. Real-life testing

I invite you to my journey of creating this app called BeSafeBox, from the beginning to the end. I hope it will help you to overcome some of your difficulties. If not, at least help you find a way how to build your next big app/idea 😃.

My GitHub repositories links: 👇

The whole series will be stored here:

Fall Detection Android App - BeSafeBox

8 stories



Tomáš Repčík

https://tomasrepcik.dev/ - Flutter app developer with experience in native app development and degree in biomedical engineering.