Overcoming limitations of smartphone sensors — Every year, 1.3 million people die because of car accidents. 20–50 million people suffer injuries. [WHO] Car accident can put the body under acceleration force with a power of more than 100g. [NHTSA] Nowadays, phones like Pixels from Google or the latest iPhones support car accident detection. Pixels uses a…

Building the fall detection app for Android — When the project started, I had little experience in Android app development. It was whooping four years ago at the time of writing this post. I built a couple of small apps for tracking sports activities before. However, they had not reached that level of maturity for releasing them publicly…

Testing fall detection model with real-life data and conversion of sklearn model to Android-compatible structure — The previous post used the data to create machine-learning models for fall detection. Random forest obtained an overall accuracy of 90.58%, with precision for the falls 79% and a sensitivity of 89% on the validation data. SVM obtained overall accuracy of 90.42%, with precision for the falls at 77% and…

Training, optimisation and evaluation of machine learning models with engineered features — In the previous post, we analysed all 3129 activities with 19 features. However, we do not need all of the features. They can be correlated or fully redundant for the prediction. Fall Detection App: Cleaning Data and Extracting Features Extracting features to differentiate falls among other activitiestomas-repcik.medium.com

Extracting features to differentiate falls among other activities — Filtering of the data To create a fall detection app, we need data. As in a previous post, we measured 3260 activities with smartphones. Volunteers were falling on the ground, walking, sitting, and lying on a bed. However, we need to overcome some pitfalls. Fall detection app: How I made people fall People falling, sitting, walking, and laying down to create the datasettomas-repcik.medium.com

How do the fall and other activities compare? — In the previous post, I described how I obtained more than 3000 samples of different activities of volunteers walking, sitting down, laying down and falling on the ground with help of the phone sensors. Measurements with adequate types can be previewed and interpreted with a small effort. To make it…

People falling, sitting, walking, and laying down to create the dataset — Defining issue During our existence in the real world, we do a lot of different movements, like sitting on the sofa, travelling on the bus, playing catch with kids and others. …

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