Edit this page

NA-MIC Project Weeks

Back to Projects List

Interactive Deep Learning for Congenital Heart Disease Segmentation

Key Investigators

Project Description


  1. Aim: segment all cardiac chambers and great vessels from cardiac MRI, for children with congenital heart disease.
  2. 20 training cases + large anatomical variability - remains a challenge for automatic segmentation.
  3. Approach: Integrate some interaction from the user, e.g. scribbles or landmarks.

Approach and Plan

  1. Already have framework for interactive segmentation. Currently testing using scribbles for aorta segmentation.
  2. Investigate data augmentation to prevent overfitting - noise / slight intensity changes / small deformations.
  3. Parameter tuning.

Progress and Next Steps

  1. Implemented on-the-fly data augmentation, including (1) random affine transformations constrained by a user-specified maximum rotation, translation, scale and shear, and (2) random elastic deformation.
  2. Currently running trials to measure impact and tune parameters.


Example multi-chamber segmentation

Background and References