arXiv 2303.119352023-03-18eess.IVcs.CVcs.LG

基于 Vision Transformer 的模型利用胸部 X 光图像量化肺炎严重程度

Vision Transformer-based Model for Severity Quantification of Lung Pneumonia Using Chest X-ray Images

University of the Basque Country（巴斯克大学）；IKERBASQUE；Ho Chi Minh City Open University（胡志明市开放大学）；University of Strasbourg（斯特拉斯堡大学）；INSERM（法国国家健康与医学研究院）；Université de Haute-Alsace（上阿尔萨斯大学）

Bouthaina Slika, Fadi Dornaika, Hamid Merdji, Karim Hammoudi

备注：: This work has been submitted to the IEEE for possible publication

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出 ViTReg-IP，一种由 ViT 与回归头构成的轻量模型，用于根据胸部 X 光量化 COVID-19 等肺病严重程度，并在多来源数据集上验证其低成本、高性能与泛化性。

摘要：

为了开发通用且可靠的方法，从胸部 X 光（CXR）中诊断和评估 COVID-19 的严重程度，需要大量维护良好的 COVID-19 数据集。现有的严重程度量化架构需要昂贵的训练计算才能获得最佳结果。为了让医疗专业人员能够快速、自动地识别 COVID-19 患者并预测相关严重程度指标，需要计算机工具。在本研究中，我们提出一种基于 Vision Transformer（ViT）的神经网络模型，该模型依赖少量可训练参数来量化 COVID-19 和其他肺部疾病的严重程度。我们提出一种可行的 CXR 严重程度量化方法，称为 Vision Transformer Regressor Infection Prediction（ViTReg-IP），它由一个 ViT 和一个回归头派生而来。我们使用来自不同公开来源的多种额外胸部 X 光测试数据集，研究了模型的泛化潜力。在此背景下，我们与若干具有竞争力的深度学习分析方法进行了比较研究。实验结果表明，我们的模型能够以相对较低的计算成本，在严重程度量化方面提供峰值性能，并具有较高的泛化能力。本研究使用的源代码已公开发布于 https://github.com/bouthainas/ViTReg-IP。

英文摘要：

To develop generic and reliable approaches for diagnosing and assessing the severity of COVID-19 from chest X-rays (CXR), a large number of well-maintained COVID-19 datasets are needed. Existing severity quantification architectures require expensive training calculations to achieve the best results. For healthcare professionals to quickly and automatically identify COVID-19 patients and predict associated severity indicators, computer utilities are needed. In this work, we propose a Vision Transformer (ViT)-based neural network model that relies on a small number of trainable parameters to quantify the severity of COVID-19 and other lung diseases. We present a feasible approach to quantify the severity of CXR, called Vision Transformer Regressor Infection Prediction (ViTReg-IP), derived from a ViT and a regression head. We investigate the generalization potential of our model using a variety of additional test chest radiograph datasets from different open sources. In this context, we performed a comparative study with several competing deep learning analysis methods. The experimental results show that our model can provide peak performance in quantifying severity with high generalizability at a relatively low computational cost. The source codes used in our work are publicly available at https://github.com/bouthainas/ViTReg-IP.

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。