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

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.

Pulmonary Adenofibromas: A Clinicopathologic Correlation of 13 Cases.

Thirteen cases of primary pulmonary adenofibromas are presented. The patients are 8 women and 5 men between the ages of 41 and 73 years (average: 57 y). The patients presented with nonspecific symptomatology or their tumor was identified during routine chest films. A wedge resection was performed in all cases with lymph node sampling. Grossly, the tumors varied in size from 1 to 2.5 cm in greatest dimension. The entire tumor was histologically evaluated in all cases. All the tumors shared similar histologic features namely leaf-like/phyllodes-like growth patterns with varying areas of sclerosis, focal inflammation, and entrapped epithelium. A wide panel of immunohistochemical studies was performed including epithelial, neural, muscle, and vascular markers, all of which showed negative staining. The tumors were positive only for vimentin in the stroma and keratin in the entrapped epithelium. Further evaluation in 6 cases using in situ hybridization for the solitary fibrous tumor was performed and was negative. Clinical follow-up in all the patients showed no evidence of recurrence or metastatic disease, during a period of 12 to 36 months. The current cases highlight the unusual occurrence of pulmonary adenofibromas and the importance of separating these tumors from other tumors that may have the potential to recur or metastasize. The use of proper immunohistochemical stains/molecular analysis aids in the proper classification of these tumors.