Adversarial Domain Adaptation for Classification of Prostate Histopathology Whole-Slide Images
This work addresses domain adaptation for prostate cancer diagnosis, which is incremental as it applies adversarial training to a known bottleneck in medical imaging.
The paper tackled the problem of Gleason grading for prostate cancer diagnosis by addressing domain shift between histopathology slides from different institutions, achieving significant classification improvement compared to baseline models.
Automatic and accurate Gleason grading of histopathology tissue slides is crucial for prostate cancer diagnosis, treatment, and prognosis. Usually, histopathology tissue slides from different institutions show heterogeneous appearances because of different tissue preparation and staining procedures, thus the predictable model learned from one domain may not be applicable to a new domain directly. Here we propose to adopt unsupervised domain adaptation to transfer the discriminative knowledge obtained from the source domain to the target domain without requiring labeling of images at the target domain. The adaptation is achieved through adversarial training to find an invariant feature space along with the proposed Siamese architecture on the target domain to add a regularization that is appropriate for the whole-slide images. We validate the method on two prostate cancer datasets and obtain significant classification improvement of Gleason scores as compared with the baseline models.