AI Article Synopsis

  • Artificial intelligence, particularly deep neural networks (DNN), can classify tumors from histology samples quickly and accurately, often matching or surpassing human pathologists' abilities.
  • There is a challenge in understanding how these neural networks make their predictions, but new explainability tools are being developed, including the use of synthetic histology created by conditional generative adversarial networks (cGAN).
  • The synthetic histology not only helps visualize key histologic features linked to tumor molecular types but also enhances the training of pathologists by providing intuitive visual aids for better understanding tumor biology.

Article Abstract

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.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10227067PMC
http://dx.doi.org/10.1038/s41698-023-00399-4DOI Listing

Publication Analysis

Top Keywords

synthetic histology
16
neural networks
8
histologic features
8
dnn models
8
histology
6
synthetic
5
deep learning
4
learning generates
4
generates synthetic
4
synthetic cancer
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!