Handcrafted Histological Transformer (H2T): Unsupervised Representation of Whole Slide Images
This work addresses the need for transparent AI in clinical pathology, though it appears incremental as it adapts existing Transformer insights to a new domain.
The authors tackled the trade-off between predictive power and interpretability in unsupervised whole-slide image (WSI) representations for cancer analysis by developing a handcrafted framework called H2T, which achieved competitive performance on 5,306 WSIs and was up to 14 times faster than Transformer models.
Diagnostic, prognostic and therapeutic decision-making of cancer in pathology clinics can now be carried out based on analysis of multi-gigapixel tissue images, also known as whole-slide images (WSIs). Recently, deep convolutional neural networks (CNNs) have been proposed to derive unsupervised WSI representations; these are attractive as they rely less on expert annotation which is cumbersome. However, a major trade-off is that higher predictive power generally comes at the cost of interpretability, posing a challenge to their clinical use where transparency in decision-making is generally expected. To address this challenge, we present a handcrafted framework based on deep CNN for constructing holistic WSI-level representations. Building on recent findings about the internal working of the Transformer in the domain of natural language processing, we break down its processes and handcraft them into a more transparent framework that we term as the Handcrafted Histological Transformer or H2T. Based on our experiments involving various datasets consisting of a total of 5,306 WSIs, the results demonstrate that H2T based holistic WSI-level representations offer competitive performance compared to recent state-of-the-art methods and can be readily utilized for various downstream analysis tasks. Finally, our results demonstrate that the H2T framework can be up to 14 times faster than the Transformer models.