The Journal of Medical Imaging allows for the peer-reviewed communication and archiving of fundamental and translational research, as well as applications, focused on medical imaging, a field that continues to benefit from technological improvements and yield biomedical advancements in the early detection, diagnostics, and therapy of disease as well as in the understanding of normal conditions.
Deep learning systems have emerged as powerful mechanisms for learning domain translation models. However, in many cases, complete information in one domain is assumed to be necessary for sufficient cross-domain prediction. In this work, we motivate a formal justification for domain-specific information separation in a simple linear case and illustrate that a self-supervised approach enables domain translation between data domains while filtering out domain-specific data features. We introduce a novel approach to identify domainspecific information from sets of unpaired measurements in complementary data domains by considering a deep learning cross-domain autoencoder architecture designed to learn shared latent representations of data while enabling domain translation. We introduce an orthogonal gate block designed to enforce orthogonality of input feature sets by explicitly removing non-sharable information specific to each domain and illustrate separability of domain-specific information on a toy dataset.
Pathologists rely on clinical information, tissue morphology, and sophisticated molecular diagnostics to accurately infer the metastatic origin of secondary liver cancer. In this paper, we introduce a deep learning approach to identify spatially localized regions of cancerous tumor within hematoxylin and eosin stained tissue sections of liver cancer and to generate predictions of the cancer’s metastatic origin. Our approach achieves an accuracy of 90.2% when classifying metastatic origin of whole slide images into three distinct classes, which compares favorably to an established clinical benchmark by three board-certified pathologists whose accuracies ranged from 90.2% to 94.1% on the same prediction task. This approach illustrates the potential impact of deep learning systems to leverage morphological and structural features of H&E stained tissue sections to guide pathological and clinical determination of the metastatic origin of secondary liver cancers.