Deep learning classification of lung cancer histology using CT images

Abstract Tumor histology is an important predictor of therapeutic response and outcomes in lung cancer. Tissue sampling for pathologist review is the most reliable method for histology classification, however, recent advances in deep learning for medical image analysis allude to the utility of radio...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Tafadzwa L. Chaunzwa, Ahmed Hosny, Yiwen Xu, Andrea Shafer, Nancy Diao, Michael Lanuti, David C. Christiani, Raymond H. Mak, Hugo J. W. L. Aerts
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/aee30c44ddfb434390b4c517b7926517
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:aee30c44ddfb434390b4c517b7926517
record_format dspace
spelling oai:doaj.org-article:aee30c44ddfb434390b4c517b79265172021-12-02T13:19:31ZDeep learning classification of lung cancer histology using CT images10.1038/s41598-021-84630-x2045-2322https://doaj.org/article/aee30c44ddfb434390b4c517b79265172021-03-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-84630-xhttps://doaj.org/toc/2045-2322Abstract Tumor histology is an important predictor of therapeutic response and outcomes in lung cancer. Tissue sampling for pathologist review is the most reliable method for histology classification, however, recent advances in deep learning for medical image analysis allude to the utility of radiologic data in further describing disease characteristics and for risk stratification. In this study, we propose a radiomics approach to predicting non-small cell lung cancer (NSCLC) tumor histology from non-invasive standard-of-care computed tomography (CT) data. We trained and validated convolutional neural networks (CNNs) on a dataset comprising 311 early-stage NSCLC patients receiving surgical treatment at Massachusetts General Hospital (MGH), with a focus on the two most common histological types: adenocarcinoma (ADC) and Squamous Cell Carcinoma (SCC). The CNNs were able to predict tumor histology with an AUC of 0.71(p = 0.018). We also found that using machine learning classifiers such as k-nearest neighbors (kNN) and support vector machine (SVM) on CNN-derived quantitative radiomics features yielded comparable discriminative performance, with AUC of up to 0.71 (p = 0.017). Our best performing CNN functioned as a robust probabilistic classifier in heterogeneous test sets, with qualitatively interpretable visual explanations to its predictions. Deep learning based radiomics can identify histological phenotypes in lung cancer. It has the potential to augment existing approaches and serve as a corrective aid for diagnosticians.Tafadzwa L. ChaunzwaAhmed HosnyYiwen XuAndrea ShaferNancy DiaoMichael LanutiDavid C. ChristianiRaymond H. MakHugo J. W. L. AertsNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Tafadzwa L. Chaunzwa
Ahmed Hosny
Yiwen Xu
Andrea Shafer
Nancy Diao
Michael Lanuti
David C. Christiani
Raymond H. Mak
Hugo J. W. L. Aerts
Deep learning classification of lung cancer histology using CT images
description Abstract Tumor histology is an important predictor of therapeutic response and outcomes in lung cancer. Tissue sampling for pathologist review is the most reliable method for histology classification, however, recent advances in deep learning for medical image analysis allude to the utility of radiologic data in further describing disease characteristics and for risk stratification. In this study, we propose a radiomics approach to predicting non-small cell lung cancer (NSCLC) tumor histology from non-invasive standard-of-care computed tomography (CT) data. We trained and validated convolutional neural networks (CNNs) on a dataset comprising 311 early-stage NSCLC patients receiving surgical treatment at Massachusetts General Hospital (MGH), with a focus on the two most common histological types: adenocarcinoma (ADC) and Squamous Cell Carcinoma (SCC). The CNNs were able to predict tumor histology with an AUC of 0.71(p = 0.018). We also found that using machine learning classifiers such as k-nearest neighbors (kNN) and support vector machine (SVM) on CNN-derived quantitative radiomics features yielded comparable discriminative performance, with AUC of up to 0.71 (p = 0.017). Our best performing CNN functioned as a robust probabilistic classifier in heterogeneous test sets, with qualitatively interpretable visual explanations to its predictions. Deep learning based radiomics can identify histological phenotypes in lung cancer. It has the potential to augment existing approaches and serve as a corrective aid for diagnosticians.
format article
author Tafadzwa L. Chaunzwa
Ahmed Hosny
Yiwen Xu
Andrea Shafer
Nancy Diao
Michael Lanuti
David C. Christiani
Raymond H. Mak
Hugo J. W. L. Aerts
author_facet Tafadzwa L. Chaunzwa
Ahmed Hosny
Yiwen Xu
Andrea Shafer
Nancy Diao
Michael Lanuti
David C. Christiani
Raymond H. Mak
Hugo J. W. L. Aerts
author_sort Tafadzwa L. Chaunzwa
title Deep learning classification of lung cancer histology using CT images
title_short Deep learning classification of lung cancer histology using CT images
title_full Deep learning classification of lung cancer histology using CT images
title_fullStr Deep learning classification of lung cancer histology using CT images
title_full_unstemmed Deep learning classification of lung cancer histology using CT images
title_sort deep learning classification of lung cancer histology using ct images
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/aee30c44ddfb434390b4c517b7926517
work_keys_str_mv AT tafadzwalchaunzwa deeplearningclassificationoflungcancerhistologyusingctimages
AT ahmedhosny deeplearningclassificationoflungcancerhistologyusingctimages
AT yiwenxu deeplearningclassificationoflungcancerhistologyusingctimages
AT andreashafer deeplearningclassificationoflungcancerhistologyusingctimages
AT nancydiao deeplearningclassificationoflungcancerhistologyusingctimages
AT michaellanuti deeplearningclassificationoflungcancerhistologyusingctimages
AT davidcchristiani deeplearningclassificationoflungcancerhistologyusingctimages
AT raymondhmak deeplearningclassificationoflungcancerhistologyusingctimages
AT hugojwlaerts deeplearningclassificationoflungcancerhistologyusingctimages
_version_ 1718393290770874368