Download This Paper
Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
DeepLN: An Artificial Intelligence Tool to Predict Pathological Subtypes in CT-Detected Pulmonary Nodules
27 Pages Posted: 26 Feb 2021
More...Abstract
Background: Distinction of malignant pulmonary nodules from the benign ones based on computed tomography (CT) images can be time-consuming but significant in routine clinical management. The advent of artificial intelligence (AI) has provided the opportunity to improve the accuracy of cancer risk prediction.
Methods: A total of 8950 detected pulmonary nodules with complete pathological results were enrolled retrospectively. The different radiological manifestations were identified mainly as various tumor density and morphological features. Then, these nodules were classified into benign and malignant groups, both of which were subdivided into finer specific types. Here, we proposed the deep convolutional neural network named DeepLN to identify the radiological features and predict the pathologic subtypes in lung nodules.
Findings: In identification of malignant nodules, our DeepLN algorithm achieved an AUC of 0.8461 in the test set. Regarding predicting the pathological subtypes in the test set, the AUCs in the benign group were 0.8784 for neoplasm, 0.8200 for inflammation, and 0.8183 for other benign ones, while AUCs were 0.8603 for LUAD, 0.8710 for LUSC, 0.7421 for other malignant ones respectively in the malignant group. In terms of density, the AUCs were 0.9713, 0.7879, and 0.9027 for the pGGO, mGGO and solid nodules. As for the morphological features, the AUCs were 0.8284 and 0.9064 for spiculation and lung cavity respectively.
Interpretation: Our deep learning algorithm represented a competitive performance to predict the pathologic subtypes on the basis of non-invasive CT images, and thus was with great possibility to be utilized in later routine clinical workflow
Funding: National Natural Science Foundation of China, National Major Science and Technology Projects of China, Science and Technology Project of Chengdu, and Science and Technology Project of Sichuan.
Declaration of Interest: None to declare.
Ethical Approval: Ethics approval was obtained from the ethics committee of West China
Hospital(Project No. 2019-195).
Suggested Citation: Suggested Citation