Authors: Negin Ghamsarian, Javier Gamazo Tejero, Pablo Márquez Neila, Sebastian Wolf, Martin Zinkernagel, Klaus Schoeffmann, and Raphael Sznitman
26th Medical Image Computing and Computer-Assisted Intervention 2023 (MICCAI 2023), Vancouver, Canada, 8-12 October 2023
Abstract: Models capable of leveraging unlabelled data are crucial in overcoming large distribution gaps between the acquired datasets across different imaging devices and configurations. In this regard, self-training techniques based on pseudo-labeling have been shown to be highly effective for unsupervised domain adaptation. However, the unreliability of pseudo labels can hinder the capability of self-training techniques to induce abstract representation from the unlabeled target dataset, especially in the case of large distribution gaps. Since the neural network performance should be invariant to image transformations, we look to this fact to identify uncertain pseudo labels. Indeed, we argue that transformation invariant detections can provide more reasonable approximations of ground truth. Accordingly, we propose an unsupervised domain adaptation strategy termed transformation-invariant self-training (TI-ST) to assess pixel-wise pseudo-labels’ reliability and filter out unreliable detections during self-training. We perform comprehensive evaluations for domain adaptation using three different modalities of medical images, two different network architectures, and several alternative state-of-the-art domain adaptation methods. Experimental results confirm the superiority of our proposed method in mitigating the lack of target domain annotation and boosting segmentation performance in the target domain.