Transfer learning with chest X-rays for ER patient classification

Authors

Jonathan Stubblefield, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Mitchell Hervert, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Jason L. Causey, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Jake A. Qualls, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Wei Dong, Ann Arbor Algorithms, Michigan
Lingrui Cai, Ann Arbor Algorithms, Ann Arbor, MI, 48103, USA.
Jennifer Fowler, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Emily Bellis, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-State
Karl Walker, University of Arkansas Pine Bluff
Jason H. Moore, University of Pennsylvania
Sara Nehring, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-StateFollow
Xiuzhen Huang, Center for No-Boundary Thinking (CNBT) at Arkansas State University - Jonesboro, The Joint Translational Research Lab of A-StateFollow

Document Type

Article

Publication Title

Scientific reports

PubMed ID

33262425

MeSH Headings (Medical Subject Headings)

Deep Learning; Disease (classification); Emergency Service, Hospital; Humans; Patients (classification); Radiography, Thoracic; Respiratory Distress Syndrome (diagnostic imaging, etiology); Retrospective Studies

Abstract

One of the challenges with urgent evaluation of patients with acute respiratory distress syndrome (ARDS) in the emergency room (ER) is distinguishing between cardiac vs infectious etiologies for their pulmonary findings. We conducted a retrospective study with the collected data of 171 ER patients. ER patient classification for cardiac and infection causes was evaluated with clinical data and chest X-ray image data. We show that a deep-learning model trained with an external image data set can be used to extract image features and improve the classification accuracy of a data set that does not contain enough image data to train a deep-learning model. An analysis of clinical feature importance was performed to identify the most important clinical features for ER patient classification. The current model is publicly available with an interface at the web link: http://nbttranslationalresearch.org/ .

First Page

20900

DOI

10.1038/s41598-020-78060-4

Publication Date

12-1-2020

Recommended Citation

Stubblefield, Jonathan; Hervert, Mitchell; Causey, Jason L.; Qualls, Jake A.; Dong, Wei; Cai, Lingrui; Fowler, Jennifer; Bellis, Emily; Walker, Karl; Moore, Jason H.; Nehring, Sara; and Huang, Xiuzhen, "Transfer learning with chest X-rays for ER patient classification" (2020). Arkansas Biosciences Institute. 57.
https://arch.astate.edu/abi/57