Cargando…

Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan

For stage-I lung adenocarcinoma, the 5-years disease-free survival (DFS) rates of non-invasive adenocarcinoma (non-IA) is different with invasive adenocarcinoma (IA). This study aims to develop CT image based artificial intelligence (AI) schemes to classify between non-IA and IA nodules, and incorpo...

Descripción completa

Detalles Bibliográficos
Autores principales: Xia, Xianwu, Gong, Jing, Hao, Wen, Yang, Ting, Lin, Yeqing, Wang, Shengping, Peng, Weijun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7136522/
https://www.ncbi.nlm.nih.gov/pubmed/32296645
http://dx.doi.org/10.3389/fonc.2020.00418
_version_ 1783518267963342848
author Xia, Xianwu
Gong, Jing
Hao, Wen
Yang, Ting
Lin, Yeqing
Wang, Shengping
Peng, Weijun
author_facet Xia, Xianwu
Gong, Jing
Hao, Wen
Yang, Ting
Lin, Yeqing
Wang, Shengping
Peng, Weijun
author_sort Xia, Xianwu
collection PubMed
description For stage-I lung adenocarcinoma, the 5-years disease-free survival (DFS) rates of non-invasive adenocarcinoma (non-IA) is different with invasive adenocarcinoma (IA). This study aims to develop CT image based artificial intelligence (AI) schemes to classify between non-IA and IA nodules, and incorporate deep learning (DL) and radiomics features to improve the classification performance. We collect 373 surgical pathological confirmed ground-glass nodules (GGNs) from 323 patients in two centers. It involves 205 non-IA (including 107 adenocarcinoma in situ and 98 minimally invasive adenocarcinoma), and 168 IA. We first propose a recurrent residual convolutional neural network based on U-Net to segment the GGNs. Then, we build two schemes to classify between non-IA and IA namely, DL scheme and radiomics scheme, respectively. Third, to improve the classification performance, we fuse the prediction scores of two schemes by applying an information fusion method. Finally, we conduct an observer study to compare our scheme performance with two radiologists by testing on an independent dataset. Comparing with DL scheme and radiomics scheme (the area under a receiver operating characteristic curve (AUC): 0.83 ± 0.05, 0.87 ± 0.04), our new fusion scheme (AUC: 0.90 ± 0.03) significant improves the risk classification performance (p < 0.05). In a comparison with two radiologists, our new model yields higher accuracy of 80.3%. The kappa value for inter-radiologist agreement is 0.6. It demonstrates that applying AI method is an effective way to improve the invasiveness risk prediction performance of GGNs. In future, fusion of DL and radiomics features may have a potential to handle the classification task with limited dataset in medical imaging.
format Online
Article
Text
id pubmed-7136522
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-71365222020-04-15 Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan Xia, Xianwu Gong, Jing Hao, Wen Yang, Ting Lin, Yeqing Wang, Shengping Peng, Weijun Front Oncol Oncology For stage-I lung adenocarcinoma, the 5-years disease-free survival (DFS) rates of non-invasive adenocarcinoma (non-IA) is different with invasive adenocarcinoma (IA). This study aims to develop CT image based artificial intelligence (AI) schemes to classify between non-IA and IA nodules, and incorporate deep learning (DL) and radiomics features to improve the classification performance. We collect 373 surgical pathological confirmed ground-glass nodules (GGNs) from 323 patients in two centers. It involves 205 non-IA (including 107 adenocarcinoma in situ and 98 minimally invasive adenocarcinoma), and 168 IA. We first propose a recurrent residual convolutional neural network based on U-Net to segment the GGNs. Then, we build two schemes to classify between non-IA and IA namely, DL scheme and radiomics scheme, respectively. Third, to improve the classification performance, we fuse the prediction scores of two schemes by applying an information fusion method. Finally, we conduct an observer study to compare our scheme performance with two radiologists by testing on an independent dataset. Comparing with DL scheme and radiomics scheme (the area under a receiver operating characteristic curve (AUC): 0.83 ± 0.05, 0.87 ± 0.04), our new fusion scheme (AUC: 0.90 ± 0.03) significant improves the risk classification performance (p < 0.05). In a comparison with two radiologists, our new model yields higher accuracy of 80.3%. The kappa value for inter-radiologist agreement is 0.6. It demonstrates that applying AI method is an effective way to improve the invasiveness risk prediction performance of GGNs. In future, fusion of DL and radiomics features may have a potential to handle the classification task with limited dataset in medical imaging. Frontiers Media S.A. 2020-03-31 /pmc/articles/PMC7136522/ /pubmed/32296645 http://dx.doi.org/10.3389/fonc.2020.00418 Text en Copyright © 2020 Xia, Gong, Hao, Yang, Lin, Wang and Peng. http://creativecommons.org/licenses/by/4.0/ 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) and the copyright owner(s) 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.
spellingShingle Oncology
Xia, Xianwu
Gong, Jing
Hao, Wen
Yang, Ting
Lin, Yeqing
Wang, Shengping
Peng, Weijun
Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title_full Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title_fullStr Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title_full_unstemmed Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title_short Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan
title_sort comparison and fusion of deep learning and radiomics features of ground-glass nodules to predict the invasiveness risk of stage-i lung adenocarcinomas in ct scan
topic Oncology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7136522/
https://www.ncbi.nlm.nih.gov/pubmed/32296645
http://dx.doi.org/10.3389/fonc.2020.00418
work_keys_str_mv AT xiaxianwu comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT gongjing comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT haowen comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT yangting comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT linyeqing comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT wangshengping comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan
AT pengweijun comparisonandfusionofdeeplearningandradiomicsfeaturesofgroundglassnodulestopredicttheinvasivenessriskofstageilungadenocarcinomasinctscan