Deep Learning Generates Synthetic Cancer Histology for Explainability and Education
This work addresses explainability and education in AI-driven cancer diagnosis, though it is incremental as it applies existing generative methods to a specific domain.
The authors tackled the challenge of explaining deep neural network predictions in cancer histology by generating synthetic histology images using a conditional GAN, which exposed histologic features linked to molecular tumor subtypes and improved pathologist training.
Artificial intelligence methods including deep neural networks (DNN) can provide rapid molecular classification of tumors from routine histology with accuracy that matches or exceeds human pathologists. Discerning how neural networks make their predictions remains a significant challenge, but explainability tools help provide insights into what models have learned when corresponding histologic features are poorly defined. Here, we present a method for improving explainability of DNN models using synthetic histology generated by a conditional generative adversarial network (cGAN). We show that cGANs generate high-quality synthetic histology images that can be leveraged for explaining DNN models trained to classify molecularly-subtyped tumors, exposing histologic features associated with molecular state. Fine-tuning synthetic histology through class and layer blending illustrates nuanced morphologic differences between tumor subtypes. Finally, we demonstrate the use of synthetic histology for augmenting pathologist-in-training education, showing that these intuitive visualizations can reinforce and improve understanding of histologic manifestations of tumor biology.