Combining graph neural networks and computer vision methods for cell nuclei classification in lung tissue

• Published in: Heliyon 2024-04, Vol.10 (7), p.e28463-e28463, Article e28463
• Main Authors: Pérez-Cano, Jose, Sansano Valero, Irene, Anglada-Rotger, David, Pina, Oscar, Salembier, Philippe, Marques, Ferran
• Format: Article
• Language: English
• Subjects: Convolutional neural networks; Graph neural networks; Histology; Lung cancer
• ISSN: 2405-8440; 2405-8440
• DOI: 10.1016/j.heliyon.2024.e28463

The detection of tumoural cells from whole slide images is an essential task in medical diagnosis and research. In this article, we propose and analyse a novel approach that combines computer vision-based models with graph neural networks to improve the accuracy of automated tumoural cell detection in lung tissue. Our proposal leverages the inherent structure and relationships between cells in the tissue. Experimental results on our own curated dataset show that modelling the problem with graphs gives the model a clear advantage over just working at pixel level. This change in perspective provides extra information that makes it possible to improve the performance. The reduction of dimensionality that comes from working with the graph also allows us to increase the field of view with low computational requirements. Code is available at https://github.com/Jerry-Master/lung-tumour-study, models are uploaded to https://huggingface.co/Jerry-Master/Hovernet-plus-Graphs, and the dataset is published on Zenodo https://zenodo.org/doi/10.5281/zenodo.8368122. •Graph-based tumoural cell detection improves over previous state of the art.•Batch normalisation usefulness depends on graph architecture.•Preparing a dataset of 85 images from 9 patients with lung tumour disease.•Open-source code and pre-trained models available for reproducible research.