Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images

Abstract Radiomics refers to the utilization of automated or semi-automated techniques to extract and analyze numerous quantitative features from medical images, such as computerized tomography (CT) or magnetic resonance imaging (MRI) scans. This study aims to develop a deep learning radiomics (DLR)...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiameng Lu, Xinyi Liu, Xiaoqing Ji, Yunxiu Jiang, Anli Zuo, Zihan Guo, Shuran Yang, Haiying Peng, Fei Sun, Degan Lu
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-91575-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849726205186539520
author Jiameng Lu
Xinyi Liu
Xiaoqing Ji
Yunxiu Jiang
Anli Zuo
Zihan Guo
Shuran Yang
Haiying Peng
Fei Sun
Degan Lu
author_facet Jiameng Lu
Xinyi Liu
Xiaoqing Ji
Yunxiu Jiang
Anli Zuo
Zihan Guo
Shuran Yang
Haiying Peng
Fei Sun
Degan Lu
author_sort Jiameng Lu
collection DOAJ
description Abstract Radiomics refers to the utilization of automated or semi-automated techniques to extract and analyze numerous quantitative features from medical images, such as computerized tomography (CT) or magnetic resonance imaging (MRI) scans. This study aims to develop a deep learning radiomics (DLR)-based approach for predicting programmed death-ligand 1 (PD-L1) expression in patients with non-small cell lung cancer (NSCLC). Data from 352 NSCLC patients with known PD-L1 expression were collected, of which 48.29% (170/352) were tested positive for PD-L1 expression. Tumor regions of interest (ROI) were semi-automatically segmented based on CT images, and DL features were extracted using Residual Network 50. The least absolute shrinkage and selection operator (LASSO) algorithm was used for feature selection and dimensionality reduction. Seven algorithms were used to build models, and the most optimal ones were identified. A combined model integrating DLR with clinical data was also developed. The predictive performance of each model was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve analysis. The DLR model, based on CT images, demonstrated an AUC of 0.85 (95% confidence interval (CI), 0.82–0.88), sensitivity of 0.80 (0.74–0.85), and specificity of 0.73 (0.70–0.77) for predicting PD-L1 status. The integrated model exhibited superior performance, with an AUC of 0.91 (0.87–0.95), sensitivity of 0.85 (0.82–0.89), and specificity of 0.75 (0.72–0.80). Our findings indicate that the DLR model holds promise as a valuable tool for predicting the PD-L1 status in patients with NSCLC, which can greatly assist in clinical decision-making and the selection of personalized treatment strategies.
format Article
id doaj-art-9409f3b812d34675994d7a80a4daae72
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-9409f3b812d34675994d7a80a4daae722025-08-20T03:10:14ZengNature PortfolioScientific Reports2045-23222025-04-0115111210.1038/s41598-025-91575-yPredicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT imagesJiameng Lu0Xinyi Liu1Xiaoqing Ji2Yunxiu Jiang3Anli Zuo4Zihan Guo5Shuran Yang6Haiying Peng7Fei Sun8Degan Lu9Department of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Nursing, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan HospitalDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineDepartment of Respiratory and Critical Care Medicine, The Second People’s Hospital of Yibin CityDepartment of Respiratory and Critical Care Medicine, Jining No.1 People’s HospitalDepartment of Respiratory, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Respiratory Diseases, Shandong Institute of Anesthesia and Respiratory Critical MedicineAbstract Radiomics refers to the utilization of automated or semi-automated techniques to extract and analyze numerous quantitative features from medical images, such as computerized tomography (CT) or magnetic resonance imaging (MRI) scans. This study aims to develop a deep learning radiomics (DLR)-based approach for predicting programmed death-ligand 1 (PD-L1) expression in patients with non-small cell lung cancer (NSCLC). Data from 352 NSCLC patients with known PD-L1 expression were collected, of which 48.29% (170/352) were tested positive for PD-L1 expression. Tumor regions of interest (ROI) were semi-automatically segmented based on CT images, and DL features were extracted using Residual Network 50. The least absolute shrinkage and selection operator (LASSO) algorithm was used for feature selection and dimensionality reduction. Seven algorithms were used to build models, and the most optimal ones were identified. A combined model integrating DLR with clinical data was also developed. The predictive performance of each model was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve analysis. The DLR model, based on CT images, demonstrated an AUC of 0.85 (95% confidence interval (CI), 0.82–0.88), sensitivity of 0.80 (0.74–0.85), and specificity of 0.73 (0.70–0.77) for predicting PD-L1 status. The integrated model exhibited superior performance, with an AUC of 0.91 (0.87–0.95), sensitivity of 0.85 (0.82–0.89), and specificity of 0.75 (0.72–0.80). Our findings indicate that the DLR model holds promise as a valuable tool for predicting the PD-L1 status in patients with NSCLC, which can greatly assist in clinical decision-making and the selection of personalized treatment strategies.https://doi.org/10.1038/s41598-025-91575-yNon-small cell lung cancer (NSCLC)Deep learningProgrammed death-ligand 1(PD-L1).
spellingShingle Jiameng Lu
Xinyi Liu
Xiaoqing Ji
Yunxiu Jiang
Anli Zuo
Zihan Guo
Shuran Yang
Haiying Peng
Fei Sun
Degan Lu
Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
Scientific Reports
Non-small cell lung cancer (NSCLC)
Deep learning
Programmed death-ligand 1(PD-L1).
title Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
title_full Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
title_fullStr Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
title_full_unstemmed Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
title_short Predicting PD-L1 status in NSCLC patients using deep learning radiomics based on CT images
title_sort predicting pd l1 status in nsclc patients using deep learning radiomics based on ct images
topic Non-small cell lung cancer (NSCLC)
Deep learning
Programmed death-ligand 1(PD-L1).
url https://doi.org/10.1038/s41598-025-91575-y
work_keys_str_mv AT jiamenglu predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT xinyiliu predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT xiaoqingji predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT yunxiujiang predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT anlizuo predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT zihanguo predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT shuranyang predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT haiyingpeng predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT feisun predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages
AT deganlu predictingpdl1statusinnsclcpatientsusingdeeplearningradiomicsbasedonctimages