LDCT image biomarkers that matter most for the deep learning classification of indeterminate pulmonary nodules

Author:

Masquelin Axel H.1,Cheney Nick2,José Estépar Raúl San3,Bates Jason H.T.4,Kinsey C. Matthew5

Affiliation:

1. Electrical and Biomedical Engineering, University of Vermont, Burlington, VT, USA

2. Computer Science, University of Vermont, Burlington, VT, USA

3. Department of Radiology, Brigham and Women’s Hospital, Somerville, MA, USA

4. Department of Medicine, College of Medicine, University of Vermont, Burlington, VT, USA

5. Department of Medicine, Pulmonary and Critical Care, College of Medicine, University of Vermont, Burlington, VT, USA

Abstract

BACKGROUND: Continued improvement in deep learning methodologies has increased the rate at which deep neural networks are being evaluated for medical applications, including diagnosis of lung cancer. However, there has been limited exploration of the underlying radiological characteristics that the network relies on to identify lung cancer in computed tomography (CT) images. OBJECTIVE: In this study, we used a combination of image masking and saliency activation maps to systematically explore the contributions of both parenchymal and tumor regions in a CT image to the classification of indeterminate lung nodules. METHODS: We selected individuals from the National Lung Screening Trial (NLST) with solid pulmonary nodules 4–20 mm in diameter. Segmentation masks were used to generate three distinct datasets; 1) an Original Dataset containing the complete low-dose CT scans from the NLST, 2) a Parenchyma-Only Dataset in which the tumor regions were covered by a mask, and 3) a Tumor-Only Dataset in which only the tumor regions were included. RESULTS: The Original Dataset significantly outperformed the Parenchyma-Only Dataset and the Tumor-Only Dataset with an AUC of 80.80 ± 3.77% compared to 76.39 ± 3.16% and 78.11 ± 4.32%, respectively. Gradient-weighted class activation mapping (Grad-CAM) of the Original Dataset showed increased attention was being given to the nodule and the tumor-parenchyma boundary when nodules were classified as malignant. This pattern of attention remained unchanged in the case of the Parenchyma-Only Dataset. Nodule size and first-order statistical features of the nodules were significantly different with the average malignant and benign nodule maximum 3d diameter being 23 mm and 12 mm, respectively. CONCLUSION: We conclude that network performance is linked to textural features of nodules such as kurtosis, entropy and intensity, as well as morphological features such as sphericity and diameter. Furthermore, textural features are more positively associated with malignancy than morphological features.

Publisher

IOS Press

Reference41 articles.

1. Y. Wu et al., Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation, ArXiv160908144 Cs (2016). http://arxiv.org/abs/1609.08144 (accessed November 2, 2021).

2. Deep Residual Learning for Image Recognition

3. Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups

4. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography;Ardila;Nat. Med.,2019

5. B. Kim et al., Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV), ArXiv171111279 Stat. (2018). http://arxiv.org/abs/1711.11279 (accessed May 13, 2021).

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3