The final, formatted version of the article will be published soon.
ORIGINAL RESEARCH article
Front. Oncol.
Sec. Radiation Oncology
Volume 14 - 2024 |
doi: 10.3389/fonc.2024.1402994
Deep learning-based image analysis predicts PD-L1 status from 18 F-FDG PET/CT images in Non-small-cell lung cancer
Provisionally accepted
chen liang
1*
meiyu zheng
1*
han zou
2*
yu han
1*
yingying zhan
1*
yu xing
1*
chang liu
1*
chao zuo
1*
Jinhai Zou
1*- 1 Cangzhou Central Hospital, Cangzhou, Hebei, China
- 2 Lanzhou University Second Hospital, Lanzhou, Gansu Province, China
Background: There is still a lack of clinically validated biomarkers to screen lung cancer patients suitable for programmed dead cell-1 (PD-1)/programmed dead cell receptor-1 (PD-L1) immunotherapy. Detection of PD-L1 expression is invasively operated and some PD-L1-negative patients can also benefit from immunotherapy, thus the joint modeling of both Deep learning images and clinical features was used to improve the prediction performance of PD-L1 expression in non-small cell lung cancer(NSCLC).: Retrospective collection of 101 patients diagnosed with pathology in our hospital who underwent 18F FDG PET/CT scans, with lung cancer tissue Tumor Propulsion Score (TPS) ≥ 1% as a positive expression. Extract lesions after preprocessing PET/CT images and using deep learning 3D DenseNet121 to learn lesions in PET, CT, and PET/CT images, fully connected 1024 features are extracted, and clinical features (age, gender, smoking/no smoking history, lesion diameter, lesion volume, maximum standard uptake value of lesions[SUVmax], mean standard uptake value of lesions[SUVmean], total lesion glycolysis[TLG]) are combined for joint modeling based on structured data Category Embeding Model.Results: area under a receiver operating characteristic (ROC) curve (AUC) and accuracy of predicting PD-L1 positive for PET, CT, and PET/CT test groups were 0.814 ± 0.0152, 0.7212 ± 0.0861, and 0.90 ± 0.0605, 0.806 ± 0.023, 0.70 ± 0.074, and 0.950 ± 0.0250, respectively; After joint clinical feature modeling, the AUC and accuracy of predicting PD-L1 positive for PET/CT were 0.96 ± 0.00905 and 0.950 ± 0.0250, respectively.This study combines the features of 18 F-FDG PET/CT images with clinical features using deep learning to predict the expression of PD-L1 in NSCLC, suggesting that 18 F-FDG PET/CT images can be conducted as biomarkers for PD-L1 expression.
Keywords: NSCLC, 18 F-FDG PET/CT, deep learning, PD-L1, Joint modeling
Received: 18 Mar 2024; Accepted: 25 Jul 2024.
Copyright: © 2024 liang, zheng, zou, han, zhan, xing, liu, zuo and Zou. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
chen liang, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
meiyu zheng, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
han zou, Lanzhou University Second Hospital, Lanzhou, Gansu Province, China
yu han, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
yingying zhan, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
yu xing, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
chang liu, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
chao zuo, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
Jinhai Zou, Cangzhou Central Hospital, Cangzhou, 061001, Hebei, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.