Text Report Analysis to Identify Opportunities for Optimizing Target Selection for Chest Radiograph Artificial Intelligence Models.

Authors

Carl Sabottke
Jason Lee
Alan Chiang
Bradley Spieler
Raza Mushtaq, Department of Medical Imaging, College of Medicine, Banner - University Medical Center, University of Arizona, 1501 N. Campbell Ave., P.O. Box 245067, Tucson, AZ 85724, USA.Follow

Document Type

Article

Abstract

Our goal was to analyze radiology report text for chest radiographs (CXRs) to identify imaging findings that have the most impact on report length and complexity. Identifying these imaging findings can highlight opportunities for designing CXR AI systems which increase radiologist efficiency. We retrospectively analyzed text from 210,025 MIMIC-CXR reports and 168,949 reports from our local institution collected from 2019 to 2022. Fifty-nine categories of imaging finding keywords were extracted from reports using natural language processing (NLP), and their impact on report length was assessed using linear regression with and without LASSO regularization. Regression was also used to assess the impact of additional factors contributing to report length, such as the signing radiologist and use of terms of perception. For modeling CXR report word counts with regression, mean coefficient of determination, R

Publication Date

2-1-2024

Publication Title

J Imaging Inform Med

ISSN

2948-2933

Volume

37

Issue

1

First Page

402

Last Page

411

PubMed ID

38343239

Digital Object Identifier (DOI)

10.1007/s10278-023-00927-5

Recommended Citation

Sabottke, Carl; Lee, Jason; Chiang, Alan; Spieler, Bradley; and Mushtaq, Raza, "Text Report Analysis to Identify Opportunities for Optimizing Target Selection for Chest Radiograph Artificial Intelligence Models." (2024). Neuroradiology. 105.
https://scholar.barrowneuro.org/neuroradiology/105