Domain Adaptation for Deviating Acquisition Protocols in CNN-based Lesion Classification on Diffusion-Weighted MR Images

TL;DR

This paper introduces a domain adaptation method for CNN-based lesion classification on DWI MR images, enabling models to handle varying input protocols across clinical sites without retraining.

Contribution

It proposes a model-based domain adaptation technique that restores training inputs from altered channels, reducing the need for re-training across different scanning protocols.

Findings

Significant performance improvement with the proposed method

Outperforms implicit domain adaptation schemes

Enables CNN deployment across heterogeneous clinical data

Abstract

End-to-end deep learning improves breast cancer classification on diffusion-weighted MR images (DWI) using a convolutional neural network (CNN) architecture. A limitation of CNN as opposed to previous model-based approaches is the dependence on specific DWI input channels used during training. However, in the context of large-scale application, methods agnostic towards heterogeneous inputs are desirable, due to the high deviation of scanning protocols between clinical sites. We propose model-based domain adaptation to overcome input dependencies and avoid re-training of networks at clinical sites by restoring training inputs from altered input channels given during deployment. We demonstrate the method's significant increase in classification performance and superiority over implicit domain adaptation provided by training-schemes operating on model-parameters instead of raw DWI images.