arXiv 2302.126532023-02-23cs.CVcs.LG

MesoGraph：从组织学图像自动分析恶性间皮瘤亚型

MesoGraph: Automatic Profiling of Malignant Mesothelioma Subtypes from Histological Images

University of Warwick（华威大学）；University of Oxford（牛津大学）；King’s College London（伦敦国王学院）；University of Middlesex（密德萨斯大学）；Queen Mary University of London（伦敦玛丽女王大学）；Brunel University London（伦敦布鲁内尔大学）；Manchester University（曼彻斯特大学）；Medizinische Hochschule Hannover（汉诺威医学院）；Imperial College London（帝国理工学院）

Mark Eastwood, Heba Sailem, Silviu Tudor, Xiaohong Gao, Judith Offman, Emmanouil Karteris, Angeles Montero Fernandez, Danny Jonigk, William Cookson, Miriam Moff… 展开作者

Mark Eastwood, Heba Sailem, Silviu Tudor, Xiaohong Gao, Judith Offman, Emmanouil Karteris, Angeles Montero Fernandez, Danny Jonigk, William Cookson, Miriam Moffatt, Sanjay Popat, Fayyaz Minhas, Jan Lukas Robertus

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

摘要：

根据上皮样和肉瘤样肿瘤细胞的相对比例，恶性间皮瘤分为三种组织学亚型：上皮样、肉瘤样和双相。双相肿瘤显示两种细胞类型的显著群体。这种亚型是主观的，并受到当前诊断指南的限制，甚至在使用三级系统表征上皮样和肉瘤样组分的相对比例的连续体时，专家胸部病理学家之间也可能存在差异。在这项工作中，我们开发了一种新的双任务图神经网络（GNN）架构，具有排名损失，以学习能够将组织区域评分到细胞分辨率的模型。这允许根据样品中所有细胞的合计肉瘤样缔合评分对肿瘤样品进行定量分析。所提出的方法只使用核心级别的标签和框架的预测任务作为一个双重多实例学习（MIL）的问题。组织由具有细胞水平形态学和区域特征的细胞图表示。我们使用来自Mesobank的外部多中心测试集，在此基础上，我们展示了我们模型的预测性能。我们通过分析细胞的典型形态学特征来验证我们的模型预测，发现我们的模型识别的一些形态学差异与病理学家使用的已知差异相匹配。我们进一步表明，模型评分预测患者生存率的风险比为2.30。所提出的方法的代码以及数据集可在https://github.com/measty/MesoGraph上获得。

英文摘要：

Malignant mesothelioma is classified into three histological subtypes, Epithelioid, Sarcomatoid, and Biphasic according to the relative proportions of epithelioid and sarcomatoid tumor cells present. Biphasic tumors display significant populations of both cell types. This subtyping is subjective and limited by current diagnostic guidelines and can differ even between expert thoracic pathologists when characterising the continuum of relative proportions of epithelioid and sarcomatoid components using a three class system. In this work, we develop a novel 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 of all the cells in the sample. The proposed approach uses only core-level labels and frames the prediction task as a dual multiple instance learning (MIL) problem. Tissue is represented by a cell graph with both cell-level morphological and regional features. We use an external multi-centric test set from Mesobank, on which we demonstrate the predictive performance of our model. We validate our model predictions through an analysis of the typical morphological features of cells according to their predicted score, finding that some of the morphological differences identified by our model match known differences used by pathologists. We further show that the model score is predictive of patient survival with a hazard ratio of 2.30. The code for the proposed approach, along with the dataset, is available at: https://github.com/measty/MesoGraph.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。