Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19
Abstract Chest radiography (CXR) is the most widely-used thoracic clinical imaging modality and is crucial for guiding the management of cardiothoracic conditions. The detection of specific CXR findings has been the main focus of several artificial intelligence (AI) systems. However, the wide range...
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/31c65be9be1043d9a386af4cf67b5c5c |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:31c65be9be1043d9a386af4cf67b5c5c |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:31c65be9be1043d9a386af4cf67b5c5c2021-12-02T19:04:19ZDeep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-1910.1038/s41598-021-93967-22045-2322https://doaj.org/article/31c65be9be1043d9a386af4cf67b5c5c2021-09-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-93967-2https://doaj.org/toc/2045-2322Abstract Chest radiography (CXR) is the most widely-used thoracic clinical imaging modality and is crucial for guiding the management of cardiothoracic conditions. The detection of specific CXR findings has been the main focus of several artificial intelligence (AI) systems. However, the wide range of possible CXR abnormalities makes it impractical to detect every possible condition by building multiple separate systems, each of which detects one or more pre-specified conditions. In this work, we developed and evaluated an AI system to classify CXRs as normal or abnormal. For training and tuning the system, we used a de-identified dataset of 248,445 patients from a multi-city hospital network in India. To assess generalizability, we evaluated our system using 6 international datasets from India, China, and the United States. Of these datasets, 4 focused on diseases that the AI was not trained to detect: 2 datasets with tuberculosis and 2 datasets with coronavirus disease 2019. Our results suggest that the AI system trained using a large dataset containing a diverse array of CXR abnormalities generalizes to new patient populations and unseen diseases. In a simulated workflow where the AI system prioritized abnormal cases, the turnaround time for abnormal cases reduced by 7–28%. These results represent an important step towards evaluating whether AI can be safely used to flag cases in a general setting where previously unseen abnormalities exist. Lastly, to facilitate the continued development of AI models for CXR, we release our collected labels for the publicly available dataset.Zaid NabulsiAndrew SellergrenShahar JamshyCharles LauEdward SantosAtilla P. KiralyWenxing YeJie YangRory PilgrimSahar KazemzadehJin YuSreenivasa Raju KalidindiMozziyar EtemadiFlorencia Garcia-VicenteDavid MelnickGreg S. CorradoLily PengKrish EswaranDaniel TseNeeral BeladiaYun LiuPo-Hsuan Cameron ChenShravya ShettyNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-15 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Zaid Nabulsi Andrew Sellergren Shahar Jamshy Charles Lau Edward Santos Atilla P. Kiraly Wenxing Ye Jie Yang Rory Pilgrim Sahar Kazemzadeh Jin Yu Sreenivasa Raju Kalidindi Mozziyar Etemadi Florencia Garcia-Vicente David Melnick Greg S. Corrado Lily Peng Krish Eswaran Daniel Tse Neeral Beladia Yun Liu Po-Hsuan Cameron Chen Shravya Shetty Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| description |
Abstract Chest radiography (CXR) is the most widely-used thoracic clinical imaging modality and is crucial for guiding the management of cardiothoracic conditions. The detection of specific CXR findings has been the main focus of several artificial intelligence (AI) systems. However, the wide range of possible CXR abnormalities makes it impractical to detect every possible condition by building multiple separate systems, each of which detects one or more pre-specified conditions. In this work, we developed and evaluated an AI system to classify CXRs as normal or abnormal. For training and tuning the system, we used a de-identified dataset of 248,445 patients from a multi-city hospital network in India. To assess generalizability, we evaluated our system using 6 international datasets from India, China, and the United States. Of these datasets, 4 focused on diseases that the AI was not trained to detect: 2 datasets with tuberculosis and 2 datasets with coronavirus disease 2019. Our results suggest that the AI system trained using a large dataset containing a diverse array of CXR abnormalities generalizes to new patient populations and unseen diseases. In a simulated workflow where the AI system prioritized abnormal cases, the turnaround time for abnormal cases reduced by 7–28%. These results represent an important step towards evaluating whether AI can be safely used to flag cases in a general setting where previously unseen abnormalities exist. Lastly, to facilitate the continued development of AI models for CXR, we release our collected labels for the publicly available dataset. |
| format |
article |
| author |
Zaid Nabulsi Andrew Sellergren Shahar Jamshy Charles Lau Edward Santos Atilla P. Kiraly Wenxing Ye Jie Yang Rory Pilgrim Sahar Kazemzadeh Jin Yu Sreenivasa Raju Kalidindi Mozziyar Etemadi Florencia Garcia-Vicente David Melnick Greg S. Corrado Lily Peng Krish Eswaran Daniel Tse Neeral Beladia Yun Liu Po-Hsuan Cameron Chen Shravya Shetty |
| author_facet |
Zaid Nabulsi Andrew Sellergren Shahar Jamshy Charles Lau Edward Santos Atilla P. Kiraly Wenxing Ye Jie Yang Rory Pilgrim Sahar Kazemzadeh Jin Yu Sreenivasa Raju Kalidindi Mozziyar Etemadi Florencia Garcia-Vicente David Melnick Greg S. Corrado Lily Peng Krish Eswaran Daniel Tse Neeral Beladia Yun Liu Po-Hsuan Cameron Chen Shravya Shetty |
| author_sort |
Zaid Nabulsi |
| title |
Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| title_short |
Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| title_full |
Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| title_fullStr |
Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| title_full_unstemmed |
Deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and COVID-19 |
| title_sort |
deep learning for distinguishing normal versus abnormal chest radiographs and generalization to two unseen diseases tuberculosis and covid-19 |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/31c65be9be1043d9a386af4cf67b5c5c |
| work_keys_str_mv |
AT zaidnabulsi deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT andrewsellergren deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT shaharjamshy deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT charleslau deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT edwardsantos deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT atillapkiraly deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT wenxingye deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT jieyang deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT rorypilgrim deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT saharkazemzadeh deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT jinyu deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT sreenivasarajukalidindi deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT mozziyaretemadi deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT florenciagarciavicente deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT davidmelnick deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT gregscorrado deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT lilypeng deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT krisheswaran deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT danieltse deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT neeralbeladia deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT yunliu deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT pohsuancameronchen deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 AT shravyashetty deeplearningfordistinguishingnormalversusabnormalchestradiographsandgeneralizationtotwounseendiseasestuberculosisandcovid19 |
| _version_ |
1718377200355377152 |