A cross-modal 3D deep learning for accurate lymph node metastasis prediction in clinical stage T1 lung adenocarcinoma

Xingyu Zhao; Xiang Wang; Wei Xia; Qiong Li; Liu Zhou; Qingchu Li; Rui Zhang; Jiali Cai; Junming Jian; Li Fan; Wei Wang; Honglin Bai; Zhen Li; Yi Xiao; Yuguo Tang; Xin Gao; Shiyuan Liu

doi:10.1016/j.lungcan.2020.04.014

A cross-modal 3D deep learning for accurate lymph node metastasis prediction in clinical stage T1 lung adenocarcinoma

Lung Cancer. 2020 Jul:145:10-17. doi: 10.1016/j.lungcan.2020.04.014. Epub 2020 Apr 25.

Authors

Xingyu Zhao¹, Xiang Wang², Wei Xia³, Qiong Li², Liu Zhou³, Qingchu Li², Rui Zhang³, Jiali Cai², Junming Jian¹, Li Fan², Wei Wang², Honglin Bai¹, Zhen Li⁴, Yi Xiao², Yuguo Tang³, Xin Gao⁵, Shiyuan Liu⁶

Affiliations

¹ University of Science and Technology of China, Hefei, China; Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China.
² Department of Radiology, Changzheng Hospital of the Navy Medical University, Shanghai, China.
³ Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China.
⁴ Department of Intervention Therapy, Zhengzhou University First Affiliated Hospital, China.
⁵ Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China. Electronic address: xingaosam@yahoo.com.
⁶ Department of Radiology, Changzheng Hospital of the Navy Medical University, Shanghai, China. Electronic address: liushiyuan@smmu.edu.cn.

PMID: 32387813
DOI: 10.1016/j.lungcan.2020.04.014

Abstract

Objectives: The evaluation of lymph node (LN) status by radiologists based on preoperative computed tomography (CT) lacks high precision for early lung cancer patients; erroneous evaluations result in inappropriate therapeutic plans and increase the risk of complications. This study aims to develop a cross-modal 3D neural network based on CT images and prior clinical knowledge for accurate prediction of LN metastasis in clinical stage T1 lung adenocarcinoma.

Patients and methods: Five hundred one lung adenocarcinoma patients with clinical stage T1 were enrolled. Data including: corresponding 3D nodule-centered patches of CT; prior clinical features; and pathological labels of LN status were obtained. We proposed a cross-modal deep learning system, which can successfully incorporate prior clinical knowledge and CT images into a 3D neural network to predict LN metastasis. We trained and validated our system with 401 cases and tested its performance with 100 cases. The result was compared with that of the logistic regression integration model, the single deep learning model without prior clinical knowledge integration, radiomics method, and manual evaluation by radiologists.

Results: The model proposed DensePriNet achieved an AUC of 0.926, which is significantly higher than the logistic regression integration model (0.904) single deep learning model (0.880), and radiomics method (0.891). The Matthews Correlation Coefficient (MCC) of DensePriNet (0.705) was significantly higher than manual classification by one senior radiologist (0.534) and one junior radiologist (0.416), respectively.

Conclusion: The performance of the single deep learning method is significantly higher than the radiomics method and the radiologists, and integration of prior clinical knowledge into the deep learning model enhance the diagnostic precision of LN status and facilitate the application of precision medicine.

Keywords: CT; Deep learning; Knowledge; Lung adenocarcinoma; Lymph node; Metastasis.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Adenocarcinoma of Lung* / diagnosis
Deep Learning*
Humans
Lung Neoplasms* / diagnosis
Lymph Nodes
Lymphatic Metastasis
Tomography, X-Ray Computed