Sparse annotation strategies for segmentation of short axis cardiac MRI

TL;DR

This study explores sparse annotation strategies for cardiac MRI segmentation, demonstrating that annotating fewer slices, especially from the middle of volumes, can achieve high accuracy with significantly less labeling effort.

Contribution

It identifies the most informative slices for annotation and shows that sparse slice annotations can match full dataset performance in cardiac MRI segmentation.

Findings

Annotating fewer slices from the middle of volumes yields high segmentation accuracy.

Training on 48 annotated volumes can achieve Dice scores above 0.85.

Focusing on slice annotations is more beneficial than increasing the number of annotated volumes.

Abstract

Short axis cardiac MRI segmentation is a well-researched topic, with excellent results achieved by state-of-the-art models in a supervised setting. However, annotating MRI volumes is time-consuming and expensive. Many different approaches (e.g. transfer learning, data augmentation, few-shot learning, etc.) have emerged in an effort to use fewer annotated data and still achieve similar performance as a fully supervised model. Nevertheless, to the best of our knowledge, none of these works focus on which slices of MRI volumes are most important to annotate for yielding the best segmentation results. In this paper, we investigate the effects of training with sparse volumes, i.e. reducing the number of cases annotated, and sparse annotations, i.e. reducing the number of slices annotated per case. We evaluate the segmentation performance using the state-of-the-art nnU-Net model on two public datasets to identify which slices are the most important to annotate. We have shown that training on a significantly reduced dataset (48 annotated volumes) can give a Dice score greater than 0.85 and results comparable to using the full dataset (160 and 240 volumes for each dataset respectively). In general, training on more slice annotations provides more valuable information compared to training on more volumes. Further, annotating slices from the middle of volumes yields the most beneficial results in terms of segmentation performance, and the apical region the worst. When evaluating the trade-off between annotating volumes against slices, annotating as many slices as possible instead of annotating more volumes is a better strategy.