Lister Hill National Center for Biomedical Communications, National Library of Medicine, Bethesda, Maryland, United States of America.
Department of Electrical and Computer Engineering, Missouri University of Science and Technology, Rolla, Missouri, United States of America.
PLoS One. 2020 Nov 12;15(11):e0242301. doi: 10.1371/journal.pone.0242301. eCollection 2020.
Data-driven deep learning (DL) methods using convolutional neural networks (CNNs) demonstrate promising performance in natural image computer vision tasks. However, their use in medical computer vision tasks faces several limitations, viz., (i) adapting to visual characteristics that are unlike natural images; (ii) modeling random noise during training due to stochastic optimization and backpropagation-based learning strategy; (iii) challenges in explaining DL black-box behavior to support clinical decision-making; and (iv) inter-reader variability in the ground truth (GT) annotations affecting learning and evaluation. This study proposes a systematic approach to address these limitations through application to the pandemic-caused need for Coronavirus disease 2019 (COVID-19) detection using chest X-rays (CXRs). Specifically, our contribution highlights significant benefits obtained through (i) pretraining specific to CXRs in transferring and fine-tuning the learned knowledge toward improving COVID-19 detection performance; (ii) using ensembles of the fine-tuned models to further improve performance over individual constituent models; (iii) performing statistical analyses at various learning stages for validating results; (iv) interpreting learned individual and ensemble model behavior through class-selective relevance mapping (CRM)-based region of interest (ROI) localization; and, (v) analyzing inter-reader variability and ensemble localization performance using Simultaneous Truth and Performance Level Estimation (STAPLE) methods. We find that ensemble approaches markedly improved classification and localization performance, and that inter-reader variability and performance level assessment helps guide algorithm design and parameter optimization. To the best of our knowledge, this is the first study to construct ensembles, perform ensemble-based disease ROI localization, and analyze inter-reader variability and algorithm performance for COVID-19 detection in CXRs.
基于数据驱动的深度学习（DL）方法使用卷积神经网络（CNNs）在自然图像计算机视觉任务中表现出很有前景的性能。然而，它们在医学计算机视觉任务中的应用面临着几个限制，即：（i）适应不同于自然图像的视觉特征；（ii）由于随机优化和基于反向传播的学习策略，在训练过程中建模随机噪声；（iii）在支持临床决策方面，解释 DL 黑盒行为的挑战；以及（iv）地面实况（GT）注释中的读者间变异性影响学习和评估。本研究提出了一种系统的方法，通过应用于使用胸部 X 射线（CXRs）检测 2019 年冠状病毒病（COVID-19）的大流行需求来解决这些限制。具体来说，我们的贡献强调了通过以下方式获得的重要益处：（i）针对 CXRs 的预训练在转移和微调所学知识方面，以提高 COVID-19 检测性能；（ii）使用微调模型的集合进一步提高个体组成模型的性能；（iii）在各个学习阶段进行统计分析以验证结果；（iv）通过基于类别选择相关性映射（CRM）的感兴趣区域（ROI）定位解释所学的个体和集合模型行为；以及（v）使用同时真实和性能水平估计（STAPLE）方法分析读者间变异性和集合本地化性能。我们发现，集合方法显著提高了分类和定位性能，而读者间变异性和性能水平评估有助于指导算法设计和参数优化。据我们所知，这是首次构建集合、进行基于集合的疾病 ROI 定位以及分析 CXR 中 COVID-19 检测的读者间变异性和算法性能的研究。
