Combining Heterogeneously Labeled Datasets For Training Segmentation Networks

TL;DR

This paper introduces a novel cost function that enables effective training of segmentation networks using heterogeneously labeled datasets, significantly improving performance over naive methods in medical imaging tasks.

Contribution

The authors propose a new cost function that allows combining datasets with different label sets for training segmentation networks, addressing heterogeneity issues.

Findings

The proposed cost function outperforms naive masking approaches.

Results are close to using fully annotated datasets.

Effective integration of heterogeneous labels improves segmentation accuracy.

Abstract

Accurate segmentation of medical images is an important step towards analyzing and tracking disease related morphological alterations in the anatomy. Convolutional neural networks (CNNs) have recently emerged as a powerful tool for many segmentation tasks in medical imaging. The performance of CNNs strongly depends on the size of the training data and combining data from different sources is an effective strategy for obtaining larger training datasets. However, this is often challenged by heterogeneous labeling of the datasets. For instance, one of the dataset may be missing labels or a number of labels may have been combined into a super label. In this work we propose a cost function which allows integration of multiple datasets with heterogeneous label subsets into a joint training. We evaluated the performance of this strategy on thigh MR and a cardiac MR datasets in which we artificially merged labels for half of the data. We found the proposed cost function substantially outperforms a naive masking approach, obtaining results very close to using the full annotations.