Samuel Pröll - Personal homepage

Facial landmark detection is still easy with MediaPipe (2023 update)

In 2023, MediaPipe has seen a major overhaul and now provides various new features in addition to a more versatile API. While code from my older post still works (as of writing - November 2023, mediapipe==0.10.7), I want to briefly take a look at the new API and recreate the rotating face using it. MediaPipe preview Note that as of November 2023, MediaPipe is still in preview and the API could change again in coming versions.

ECG R peak detection in Python: a comparison of libraries

For decades now, electrocardiography (ECG) has been a crucial tool in medicine. And with wearable ECG devices making their way into clinical settings, the amount of ECG data available will continue to increase1. Automatic detection of heart beats (R peaks, QRS complexes) is an important step in ECG analysis. In this post, we will compare some of the libraries you may come across when looking for a ready-to-use solution in Python.

Managing multi-dimensional datasets with metadata in Python

What many datasets for machine learning tutorials are lacking is metadata. Example datasets are, with good reason, kept as simple as possible and only capture the core of the problem. In the real world however, a business goal very rarely comes as one neat pack of uniform data. You may have CT images obtained with different scanners, manufacturing data from different facilities, sales data including different marketing campaigns… When working on a particular task, you want to be aware of such information, as disregarding it may lead to unexpected model behavior.