GitHub - AnginoA/Aircatch: This study addresses the quantification of gaseous microemboli during cardiac interventions using a machine learning approach with real-time echography data. A 2.5D U-Net model segments GME across temporal ultrasound images, enabling precise pixel-based quantification of GME size.

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
images		images
results		results
trained_nns		trained_nns
Dataset_Generator.py		Dataset_Generator.py
Image_DF_history_1_main_frame_0.h5		Image_DF_history_1_main_frame_0.h5
figure.png		figure.png
functions.py		functions.py
mask_generator.py		mask_generator.py

About

This study addresses the quantification of gaseous microemboli during cardiac interventions using a machine learning approach with real-time echography data. A 2.5D U-Net model segments GME across temporal ultrasound images, enabling precise pixel-based quantification of GME size.