Preprocessing Data

The preprocessing of imaging data is essential to effective training and testing of ML models. In these notebooks we review how to remove outliers, baseline wander, and complete per-lead normalization. The first file is the most important. Rolling averages and flatline removals are non-essential preprocessing steps but may be beneficial in other use cases.

1. Waveform_Array_Generation_Truncation_Normalization.ipynb

Truncation, Baseline Wander Removal, and Per-Lead Normalization

2. Visualizing_ECGs_Rolling_Average.ipynb

Denoise ECGs using rolling average screening techniques

3. Finding_Flatlines.ipynb

Find flatlines in ECG data of various length for additional filtering