MesoGraph: automatic profiling of mesothelioma subtypes from histological images
Article
Eastwood, M., Sailem, H., Marc, S., Gao, X., Offman, J., Karteris, E., Fernandez, A., Jonigk, D., Cookson, W., Moffatt, M., Popat, S., Minhas, F. and Robertus, J. 2023. MesoGraph: automatic profiling of mesothelioma subtypes from histological images. Cell Reports Medicine. 4 (10). https://doi.org/10.1016/j.xcrm.2023.101226
Type | Article |
---|---|
Title | MesoGraph: automatic profiling of mesothelioma subtypes from histological images |
Authors | Eastwood, M., Sailem, H., Marc, S., Gao, X., Offman, J., Karteris, E., Fernandez, A., Jonigk, D., Cookson, W., Moffatt, M., Popat, S., Minhas, F. and Robertus, J. |
Abstract | Mesothelioma is classified into three histological subtypes, epithelioid, sarcomatoid, and biphasic, according to the relative proportions of epithelioid and sarcomatoid tumor cells present. Current guidelines recommend that the sarcomatoid component of each mesothelioma is quantified, as a higher percentage of sarcomatoid pattern in biphasic mesothelioma shows poorer prognosis. In this work, we develop a dual-task graph neural network (GNN) architecture with ranking loss to learn a model capable of scoring regions of tissue down to cellular resolution. This allows quantitative profiling of a tumor sample according to the aggregate sarcomatoid association score. Tissue is represented by a cell graph with both cell-level morphological and regional features. We use an external multicentric test set from Mesobank, on which we demonstrate the predictive performance of our model. We additionally validate our model predictions through an analysis of the typical morphological features of cells according to their predicted score. |
Keywords | graph; neural networks; multiple instance learning; mesothelioma; cancer; subtyping; digital pathology |
Sustainable Development Goals | 3 Good health and well-being |
Middlesex University Theme | Health & Wellbeing |
Research Group | Artificial Intelligence group |
Publisher | Elsevier |
Journal | Cell Reports Medicine |
ISSN | 2666-3791 |
Publication dates | |
Online | 09 Oct 2023 |
17 Oct 2023 | |
Publication process dates | |
Submitted | 12 Jan 2023 |
Accepted | 14 Sep 2023 |
Deposited | 17 Nov 2023 |
Output status | Published |
Publisher's version | License |
Digital Object Identifier (DOI) | https://doi.org/10.1016/j.xcrm.2023.101226 |
Scopus EID | 2-s2.0-85174075475 |
Web of Science identifier | WOS:001102113300001 |
Related Output | |
Is new version of | MesoGraph: automatic profiling of malignant mesothelioma subtypes from histological images |
Language | English |
https://repository.mdx.ac.uk/item/v7y06
Download files
74
total views23
total downloads3
views this month3
downloads this month