Cargando…

Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies

Achieving high feature reproducibility while preserving biological information is one of the main challenges for the generalizability of current radiomics studies. Non-clinical imaging variables, such as reconstruction kernels, have shown to significantly impact radiomics features. In this study, we...

Descripción completa

Detalles Bibliográficos
Autores principales: Yoon, Jin H., Sun, Shawn H., Xiao, Manjun, Yang, Hao, Lu, Lin, Li, Yajun, Schwartz, Lawrence H., Zhao, Binsheng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8707549/
https://www.ncbi.nlm.nih.gov/pubmed/34941646
http://dx.doi.org/10.3390/tomography7040074
_version_ 1784622463649316864
author Yoon, Jin H.
Sun, Shawn H.
Xiao, Manjun
Yang, Hao
Lu, Lin
Li, Yajun
Schwartz, Lawrence H.
Zhao, Binsheng
author_facet Yoon, Jin H.
Sun, Shawn H.
Xiao, Manjun
Yang, Hao
Lu, Lin
Li, Yajun
Schwartz, Lawrence H.
Zhao, Binsheng
author_sort Yoon, Jin H.
collection PubMed
description Achieving high feature reproducibility while preserving biological information is one of the main challenges for the generalizability of current radiomics studies. Non-clinical imaging variables, such as reconstruction kernels, have shown to significantly impact radiomics features. In this study, we retrain an open-source convolutional neural network (CNN) to harmonize computerized tomography (CT) images with various reconstruction kernels to improve feature reproducibility and radiomic model performance using epidermal growth factor receptor (EGFR) mutation prediction in lung cancer as a paradigm. In the training phase, the CNN was retrained and tested on 32 lung cancer patients’ CT images between two different groups of reconstruction kernels (smooth and sharp). In the validation phase, the retrained CNN was validated on an external cohort of 223 lung cancer patients’ CT images acquired using different CT scanners and kernels. The results showed that the retrained CNN could be successfully applied to external datasets with different CT scanner parameters, and harmonization of reconstruction kernels from sharp to smooth could significantly improve the performance of radiomics model in predicting EGFR mutation status in lung cancer. In conclusion, the CNN based method showed great potential in improving feature reproducibility and generalizability by harmonizing medical images with heterogeneous reconstruction kernels.
format Online
Article
Text
id pubmed-8707549
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-87075492021-12-25 Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies Yoon, Jin H. Sun, Shawn H. Xiao, Manjun Yang, Hao Lu, Lin Li, Yajun Schwartz, Lawrence H. Zhao, Binsheng Tomography Article Achieving high feature reproducibility while preserving biological information is one of the main challenges for the generalizability of current radiomics studies. Non-clinical imaging variables, such as reconstruction kernels, have shown to significantly impact radiomics features. In this study, we retrain an open-source convolutional neural network (CNN) to harmonize computerized tomography (CT) images with various reconstruction kernels to improve feature reproducibility and radiomic model performance using epidermal growth factor receptor (EGFR) mutation prediction in lung cancer as a paradigm. In the training phase, the CNN was retrained and tested on 32 lung cancer patients’ CT images between two different groups of reconstruction kernels (smooth and sharp). In the validation phase, the retrained CNN was validated on an external cohort of 223 lung cancer patients’ CT images acquired using different CT scanners and kernels. The results showed that the retrained CNN could be successfully applied to external datasets with different CT scanner parameters, and harmonization of reconstruction kernels from sharp to smooth could significantly improve the performance of radiomics model in predicting EGFR mutation status in lung cancer. In conclusion, the CNN based method showed great potential in improving feature reproducibility and generalizability by harmonizing medical images with heterogeneous reconstruction kernels. MDPI 2021-12-03 /pmc/articles/PMC8707549/ /pubmed/34941646 http://dx.doi.org/10.3390/tomography7040074 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Yoon, Jin H.
Sun, Shawn H.
Xiao, Manjun
Yang, Hao
Lu, Lin
Li, Yajun
Schwartz, Lawrence H.
Zhao, Binsheng
Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title_full Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title_fullStr Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title_full_unstemmed Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title_short Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies
title_sort convolutional neural network addresses the confounding impact of ct reconstruction kernels on radiomics studies
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8707549/
https://www.ncbi.nlm.nih.gov/pubmed/34941646
http://dx.doi.org/10.3390/tomography7040074
work_keys_str_mv AT yoonjinh convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT sunshawnh convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT xiaomanjun convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT yanghao convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT lulin convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT liyajun convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT schwartzlawrenceh convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies
AT zhaobinsheng convolutionalneuralnetworkaddressestheconfoundingimpactofctreconstructionkernelsonradiomicsstudies