IASLC早期肺成像联盟（ELIC）开源深度学习和定量测量倡议

SCI

13 September 2023

The IASLC Early Lung Imaging Confederation (ELIC) Open-Source Deep Learning and Quantitative Measurement Initiative

(Journal of Thoracic Oncology, IF: 20.4)

Stephen Lam, MD, Murry W. Wynes, PhD, Casey Connolly, MPH, Kazuto Ashizawa, MD, PhD, Sukhinder Atkar-Khattra, BSc., Chandra P. Belani, MD, Domenic DiNatale, Claudia I. Henschke, MD, PhD, Bruno Hochhegger, MD, PhD, Claudio Jacomelli, Małgorzata Jelitto, MD, Artit Jirapatnakul, PhD, Karen L. Kelly, MD, Karthik Krishnan, MS, Takeshi Kobayashi, MD, PhD, Jacqueline Logan, BaN, Juliane Mattos, PhD, John Mayo, MD, Annette McWilliams, MBBS, FRACP, Tetsuya Mitsudomi, MD, Ugo Pastorino, MD, Joanna Polańska, PhD, Witold Rzyman, MD, PhD, Ricardo Sales dos Santos, MD, PhD, Giorgio V. Scagliotti, MD, PhD, Heather Wakelee, MD, David F. Yankelevitz, MD, John K. Field, PhD, FRCPath, James L. Mulshine, MD, Ricardo Avila, MS

CORREPONDENCE To: Ricardo Avila

Background 背景

With global adoption of CT lung cancer screening, there is increasing interest to use artificial intelligence (AI) deep learning methods to improve the clinical management process. To enable AI research using an open source, cloud-based, globally distributed, screening CT imaging dataset and computational environment that are compliant with the most stringent international privacy regulations that also protects the intellectual properties of researchers, the International Association of the Study of Lung Cancer (IASLC) sponsored development of the Early Lung Imaging Confederation (ELIC) resource in 2018. The objective of this report is to describe the updated capabilities of ELIC and illustrate how this resource can be utilized for clinically relevant AI research.

随着CT肺癌筛查在全球的普及，使用人工智能（AI）深度学习方法来改进临床管理过程的趋势渐盛。为了使用开源、基于云、全球分布的筛查CT影像数据库和计算环境进行人工智能研究，这些数据集和计算环境符合最严格的国际隐私法规，也保护了研究人员的知识产权，国际肺癌研究协会（IASLC）于2018年赞助了早期肺成像联盟（ELIC）资源的开发。本报告的目的是描述ELIC的最新功能，并说明如何将该资源用于临床相关的人工智能研究。

Methods 方法

In this second Phase of the initiative, metadata and screening CT scans from two time points were collected from 100 screening participants in seven countries. An automated deep learning AI lung segmentation algorithm, automated quantitative emphysema metrics, and a quantitative lung nodule volume measurement algorithm were run on these scans.

在该倡议的第二阶段，从七个国家的100名筛查者处收集了两个时间点的元数据和CT筛查图像。在这些图像上运行了自动深度学习AI肺段算法、自动定量肺气肿指标和定量肺结节体积测量算法。

Results 后果

A total of 1,394 CTs were collected from 697 participants. The LAV950 quantitative emphysema metric was found to be potentially useful in distinguishing lung cancer from benign cases using a combined slice thickness ≥ 2.5 mm. Lung nodule volume change measurements had better sensitivity and specificity for classifying malignant from benign lung nodules when applied to solid lung nodules from high quality CT scans.

总计从697名参与者中收集了1394份胸部CT影像数据。LAV950定量肺气肿分析测量可通过层厚≥2.5 mm的胸部CT影像来区分癌症和良性病例。当应用高质量CT扫描识别实体肺结节时，肺结节体积变化测量对于区分恶性和良性肺结节具有更好的灵敏度和特异性。

Conclusion 结论

These initial experiments demonstrated that ELIC can support deep learning AI and quantitative imaging analyses on diverse and globally distributed cloud-based datasets.

这些初步实验表明，ELIC可在基于云的全球分布式数据集上支持深度学习人工智能和定量成像分析。

打赏