Published — Heavy Data Augmentation Boosts Cranial Defect Reconstruction

Marek Wodzinski · September 11, 2024

DeepImplant

This paper has been published in Computers in Biology and Medicine (2024). It demonstrates that strong, diverse augmentation strategies - from registration-based transforms to latent-diffusion augmentations-— substantially improve reconstruction robustness for rare and noisy defects. The result is faster, more reliable automatic implant modeling that generalizes better to real clinical scans.

Read the paper

Twitter, Facebook