TY - JOUR
T1 - CLSSL-ResNet
T2 - Predicting malignancy of solitary pulmonary nodules from CT images by chimeric label with self-supervised learning
AU - Zhao, Tianhu
AU - Qi, Shouliang
AU - Yue, Yong
AU - Zhang, Baihua
AU - Li, Jingxu
AU - Wen, Yanhua
AU - Yao, Yudong
AU - Qian, Wei
AU - Guan, Yubao
N1 - Publisher Copyright:
© 2023 - IOS Press. All rights reserved.
PY - 2023
Y1 - 2023
N2 - Background: Pulmonary granulomatous nodules (GN) with spiculation or lobulation have a similar morphological appearance to solid lung adenocarcinoma (SADC) under computed tomography (CT). However, these two kinds of solid pulmonary nodules (SPN) have different malignancies and are sometimes misdiagnosed. Objective: This study aims to predict malignancies of SPNs by a deep learning model automatically. Methods: A chimeric label with self-supervised learning (CLSSL) is proposed to pre-train a ResNet-based network (CLSSL-ResNet) for distinguishing isolated atypical GN from SADC in CT images. The malignancy, rotation, and morphology labels are integrated into a chimeric label and utilized to pre-train a ResNet50. The pre-trained ResNet50 is then transferred and fine-tuned to predict the malignancy of SPN. Two image datasets of 428 subjects (Dataset1, 307; Dataset2, 121) from different hospitals are collected. Dataset1 is divided into training, validation, and test data by a ratio of 7:1:2 to develop the model. Dataset2 is utilized as an external validation dataset. Results: CLSSL-ResNet achieves an area under the ROC curve (AUC) of 0.944 and an accuracy (ACC) of 91.3%, which was much higher than that of the consensus of two experienced chest radiologists (77.3%). CLSSL-ResNet also outperforms other self-supervised learning models and many counterparts of other backbone networks. In Dataset2, AUC and ACC of CLSSL-ResNet are 0.923 and 89.3%, respectively. Additionally, the ablation experiment result indicates higher efficiency of the chimeric label. Conclusion: CLSSL with morphology labels can increase the ability of feature representation by deep networks. As a non-invasive method, CLSSL-ResNet can distinguish GN from SADC via CT images and may support clinical diagnoses after further validation.
AB - Background: Pulmonary granulomatous nodules (GN) with spiculation or lobulation have a similar morphological appearance to solid lung adenocarcinoma (SADC) under computed tomography (CT). However, these two kinds of solid pulmonary nodules (SPN) have different malignancies and are sometimes misdiagnosed. Objective: This study aims to predict malignancies of SPNs by a deep learning model automatically. Methods: A chimeric label with self-supervised learning (CLSSL) is proposed to pre-train a ResNet-based network (CLSSL-ResNet) for distinguishing isolated atypical GN from SADC in CT images. The malignancy, rotation, and morphology labels are integrated into a chimeric label and utilized to pre-train a ResNet50. The pre-trained ResNet50 is then transferred and fine-tuned to predict the malignancy of SPN. Two image datasets of 428 subjects (Dataset1, 307; Dataset2, 121) from different hospitals are collected. Dataset1 is divided into training, validation, and test data by a ratio of 7:1:2 to develop the model. Dataset2 is utilized as an external validation dataset. Results: CLSSL-ResNet achieves an area under the ROC curve (AUC) of 0.944 and an accuracy (ACC) of 91.3%, which was much higher than that of the consensus of two experienced chest radiologists (77.3%). CLSSL-ResNet also outperforms other self-supervised learning models and many counterparts of other backbone networks. In Dataset2, AUC and ACC of CLSSL-ResNet are 0.923 and 89.3%, respectively. Additionally, the ablation experiment result indicates higher efficiency of the chimeric label. Conclusion: CLSSL with morphology labels can increase the ability of feature representation by deep networks. As a non-invasive method, CLSSL-ResNet can distinguish GN from SADC via CT images and may support clinical diagnoses after further validation.
KW - Lung adenocarcinoma
KW - chimeric label
KW - classification
KW - deep learning
KW - granulomatous nodules
KW - self-supervised learning
UR - http://www.scopus.com/inward/record.url?scp=85171601190&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85171601190&partnerID=8YFLogxK
U2 - 10.3233/XST-230063
DO - 10.3233/XST-230063
M3 - Article
C2 - 37424490
AN - SCOPUS:85171601190
SN - 0895-3996
VL - 31
SP - 981
EP - 999
JO - Journal of X-Ray Science and Technology
JF - Journal of X-Ray Science and Technology
IS - 5
ER -