Knowledge distillation for semi-supervised domain adaptation

TL;DR

This paper introduces a knowledge distillation approach for semi-supervised domain adaptation in medical image segmentation, outperforming adversarial methods without requiring dataset-specific hyperparameter tuning.

Contribution

It proposes a novel KD-based method for domain adaptation that is simpler and more generally applicable than ADA, eliminating the need for hyperparameter tuning.

Findings

KD method achieves higher WMH dice scores than baseline and ADA.

The approach reduces overfitting and improves generalization to unseen MRI scanners.

It simplifies semi-supervised domain adaptation without dataset-specific tuning.

Abstract

In the absence of sufficient data variation (e.g., scanner and protocol variability) in annotated data, deep neural networks (DNNs) tend to overfit during training. As a result, their performance is significantly lower on data from unseen sources compared to the performance on data from the same source as the training data. Semi-supervised domain adaptation methods can alleviate this problem by tuning networks to new target domains without the need for annotated data from these domains. Adversarial domain adaptation (ADA) methods are a popular choice that aim to train networks in such a way that the features generated are domain agnostic. However, these methods require careful dataset-specific selection of hyperparameters such as the complexity of the discriminator in order to achieve a reasonable performance. We propose to use knowledge distillation (KD) -- an efficient way of transferring knowledge between different DNNs -- for semi-supervised domain adaption of DNNs. It does not require dataset-specific hyperparameter tuning, making it generally applicable. The proposed method is compared to ADA for segmentation of white matter hyperintensities (WMH) in magnetic resonance imaging (MRI) scans generated by scanners that are not a part of the training set. Compared with both the baseline DNN (trained on source domain only and without any adaption to target domain) and with using ADA for semi-supervised domain adaptation, the proposed method achieves significantly higher WMH dice scores.