Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning

Clear cell renal cell carcinoma (ccRCC) is the most common type of renal cancer. Extensive efforts have been made to utilize radiomics from computed tomography (CT) imaging to predict tumor immune microenvironment (TIME) measurements. This study proposes a Green Learning (GL) framework for approxima...

Full description

Saved in:
Bibliographic Details
Main Authors: Yixing Wu, Alexander Shieh, Steven Cen, Darryl Hwang, Xiaomeng Lei, S. J. Pawan, Manju Aron, Inderbir Gill, William D. Wallace, C.-C. Jay Kuo, Vinay Duddalwar
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/11/6/191
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850168396490997760
author Yixing Wu
Alexander Shieh
Steven Cen
Darryl Hwang
Xiaomeng Lei
S. J. Pawan
Manju Aron
Inderbir Gill
William D. Wallace
C.-C. Jay Kuo
Vinay Duddalwar
author_facet Yixing Wu
Alexander Shieh
Steven Cen
Darryl Hwang
Xiaomeng Lei
S. J. Pawan
Manju Aron
Inderbir Gill
William D. Wallace
C.-C. Jay Kuo
Vinay Duddalwar
author_sort Yixing Wu
collection DOAJ
description Clear cell renal cell carcinoma (ccRCC) is the most common type of renal cancer. Extensive efforts have been made to utilize radiomics from computed tomography (CT) imaging to predict tumor immune microenvironment (TIME) measurements. This study proposes a Green Learning (GL) framework for approximating tissue-based biomarkers from CT scans, focusing on the PD-L1 expression and CD68 tumor-associated macrophages (TAMs) in ccRCC. Our approach includes radiomic feature extraction, redundancy removal, and supervised feature selection through a discriminant feature test (DFT), a relevant feature test (RFT), and least-squares normal transform (LNT) for robust feature generation. For the PD-L1 expression in 52 ccRCC patients, treated as a regression problem, our GL model achieved a 5-fold cross-validated mean squared error (MSE) of 0.0041 and a Mean Absolute Error (MAE) of 0.0346. For the TAM population (CD68+/PanCK+), analyzed in 78 ccRCC patients as a binary classification task (at a 0.4 threshold), the model reached a 10-fold cross-validated Area Under the Receiver Operating Characteristic (AUROC) of 0.85 (95% CI [0.76, 0.93]) using 10 LNT-derived features, improving upon the previous benchmark of 0.81. This study demonstrates the potential of GL in radiomic analyses, offering a scalable, efficient, and interpretable framework for the non-invasive approximation of key biomarkers.
format Article
id doaj-art-9e401fa311ca4c8890fac7ac30ae3abe
institution OA Journals
issn 2313-433X
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-9e401fa311ca4c8890fac7ac30ae3abe2025-08-20T02:20:58ZengMDPI AGJournal of Imaging2313-433X2025-06-0111619110.3390/jimaging11060191Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green LearningYixing Wu0Alexander Shieh1Steven Cen2Darryl Hwang3Xiaomeng Lei4S. J. Pawan5Manju Aron6Inderbir Gill7William D. Wallace8C.-C. Jay Kuo9Vinay Duddalwar10Media Communications Lab, University of Southern California, Los Angeles, CA 90089, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USADepartment of Pathology, University of Southern California, Los Angeles, CA 90033, USAInstitute of Urology, University of Southern California, Los Angeles, CA 90033, USADepartment of Pathology, University of Southern California, Los Angeles, CA 90033, USAMedia Communications Lab, University of Southern California, Los Angeles, CA 90089, USARadiomics Lab, Department of Radiology, University of Southern California, Los Angeles, CA 90033, USAClear cell renal cell carcinoma (ccRCC) is the most common type of renal cancer. Extensive efforts have been made to utilize radiomics from computed tomography (CT) imaging to predict tumor immune microenvironment (TIME) measurements. This study proposes a Green Learning (GL) framework for approximating tissue-based biomarkers from CT scans, focusing on the PD-L1 expression and CD68 tumor-associated macrophages (TAMs) in ccRCC. Our approach includes radiomic feature extraction, redundancy removal, and supervised feature selection through a discriminant feature test (DFT), a relevant feature test (RFT), and least-squares normal transform (LNT) for robust feature generation. For the PD-L1 expression in 52 ccRCC patients, treated as a regression problem, our GL model achieved a 5-fold cross-validated mean squared error (MSE) of 0.0041 and a Mean Absolute Error (MAE) of 0.0346. For the TAM population (CD68+/PanCK+), analyzed in 78 ccRCC patients as a binary classification task (at a 0.4 threshold), the model reached a 10-fold cross-validated Area Under the Receiver Operating Characteristic (AUROC) of 0.85 (95% CI [0.76, 0.93]) using 10 LNT-derived features, improving upon the previous benchmark of 0.81. This study demonstrates the potential of GL in radiomic analyses, offering a scalable, efficient, and interpretable framework for the non-invasive approximation of key biomarkers.https://www.mdpi.com/2313-433X/11/6/191radiomicsclear cell renal cell carcinomaPD-L1CD68Green Learningfeature selection
spellingShingle Yixing Wu
Alexander Shieh
Steven Cen
Darryl Hwang
Xiaomeng Lei
S. J. Pawan
Manju Aron
Inderbir Gill
William D. Wallace
C.-C. Jay Kuo
Vinay Duddalwar
Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
Journal of Imaging
radiomics
clear cell renal cell carcinoma
PD-L1
CD68
Green Learning
feature selection
title Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
title_full Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
title_fullStr Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
title_full_unstemmed Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
title_short Prediction of PD-L1 and CD68 in Clear Cell Renal Cell Carcinoma with Green Learning
title_sort prediction of pd l1 and cd68 in clear cell renal cell carcinoma with green learning
topic radiomics
clear cell renal cell carcinoma
PD-L1
CD68
Green Learning
feature selection
url https://www.mdpi.com/2313-433X/11/6/191
work_keys_str_mv AT yixingwu predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT alexandershieh predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT stevencen predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT darrylhwang predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT xiaomenglei predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT sjpawan predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT manjuaron predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT inderbirgill predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT williamdwallace predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT ccjaykuo predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning
AT vinayduddalwar predictionofpdl1andcd68inclearcellrenalcellcarcinomawithgreenlearning