TY - JOUR
T1 - Ground Glass Lesions on Chest Imaging
T2 - Evaluation of Reported Incidence in Cancer Patients Using Natural Language Processing
AU - Van Haren, Robert M.
AU - Correa, Arlene M.
AU - Sepesi, Boris
AU - Rice, David C.
AU - Hofstetter, Wayne L.
AU - Mehran, Reza J.
AU - Vaporciyan, Ara A.
AU - Walsh, Garrett L.
AU - Roth, Jack A.
AU - Swisher, Stephen G.
AU - Antonoff, Mara B.
N1 - Publisher Copyright:
© 2019 The Society of Thoracic Surgeons
PY - 2019/3
Y1 - 2019/3
N2 - Background: Ground glass opacities (GGOs) on computed tomography (CT) have gained significant recent attention, with unclear incidence and epidemiologic patterns. Natural language processing (NLP) is a powerful computing tool that collects variables from unstructured data fields. Our objective was to characterize trends of GGO detection using NLP. Methods: Patients were identified at a large quaternary referral center who underwent chest CT from 2000 to 2016 via query of institutional databases. NLP was used to identify imaging reports with GGOs and to obtain additional demographic data. Incidence of reported GGOs was tracked over time. Multivariate regression was used to identify predictors of GGOs identified on chest CT. Results: A total of 244,391 chest CTs were included, with 35,386 (14.5%) revealing GGOs. There was a significant relationship between advancing year of chest CT and likelihood of reported GGOs (p < 0.001). GGOs were more likely to occur in older subjects (60.5 vs 58.5 years, p < 0.001), males (54.6% vs 51.5%, p < 0.001), and nonwhite races (21.2% Asian, 15.6% Hispanic, 14.4% black, 14.0% white; p < 0.001). Certain occupational histories predicted more frequent GGOs (p < 0.001), including transportation labor (47.4%), metal workers (42.3%), iron workers (33.3%), cabinetry (32.6%), and foremen (29.6%). Multivariate regression revealed age, sex, nonsmokers, increasing year of chest CT, and race as significant independent predictors of identifying GGOs. Conclusions: NLP explored a large cohort of patients who underwent chest CT over the study period. Demographic features predicting reported GGOs include age, sex, race, and occupation. GGO recognition continues to increase with time, and further studies investigating etiology and prognostic implications are necessary.
AB - Background: Ground glass opacities (GGOs) on computed tomography (CT) have gained significant recent attention, with unclear incidence and epidemiologic patterns. Natural language processing (NLP) is a powerful computing tool that collects variables from unstructured data fields. Our objective was to characterize trends of GGO detection using NLP. Methods: Patients were identified at a large quaternary referral center who underwent chest CT from 2000 to 2016 via query of institutional databases. NLP was used to identify imaging reports with GGOs and to obtain additional demographic data. Incidence of reported GGOs was tracked over time. Multivariate regression was used to identify predictors of GGOs identified on chest CT. Results: A total of 244,391 chest CTs were included, with 35,386 (14.5%) revealing GGOs. There was a significant relationship between advancing year of chest CT and likelihood of reported GGOs (p < 0.001). GGOs were more likely to occur in older subjects (60.5 vs 58.5 years, p < 0.001), males (54.6% vs 51.5%, p < 0.001), and nonwhite races (21.2% Asian, 15.6% Hispanic, 14.4% black, 14.0% white; p < 0.001). Certain occupational histories predicted more frequent GGOs (p < 0.001), including transportation labor (47.4%), metal workers (42.3%), iron workers (33.3%), cabinetry (32.6%), and foremen (29.6%). Multivariate regression revealed age, sex, nonsmokers, increasing year of chest CT, and race as significant independent predictors of identifying GGOs. Conclusions: NLP explored a large cohort of patients who underwent chest CT over the study period. Demographic features predicting reported GGOs include age, sex, race, and occupation. GGO recognition continues to increase with time, and further studies investigating etiology and prognostic implications are necessary.
UR - http://www.scopus.com/inward/record.url?scp=85060768393&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060768393&partnerID=8YFLogxK
U2 - 10.1016/j.athoracsur.2018.09.016
DO - 10.1016/j.athoracsur.2018.09.016
M3 - Article
C2 - 30612991
AN - SCOPUS:85060768393
SN - 0003-4975
VL - 107
SP - 936
EP - 940
JO - Annals of Thoracic Surgery
JF - Annals of Thoracic Surgery
IS - 3
ER -