Estimating the Volume of Nodules and Masses on Serial Chest Radiography Using a Deep-Learning-Based Automatic Detection Algorithm: A Preliminary Study

Estimating the Volume of Nodules and Masses on Serial Chest Radiography Using a Deep-Learning-Based Automatic Detection Algorithm: A Preliminary Study

The purpose of this study was to assess the volume of the pulmonary nodules and masses on serial chest X-rays (CXRs) from deep-learning-based automatic detection algorithm (DLAD)-based parameters. In a retrospective single-institutional study, 72 patients, who obtained serial CXRs ( = 147) for pulmo...

Full description

Saved in:
Bibliographic Details
Published in:Diagnostics (Basel) 2023-06, Vol.13 (12), p.2060
Main Authors: Lim, Chae Young, Cha, Yoon Ki, Chung, Myung Jin, Park, Subin, Park, Soyoung, Woo, Jung Han, Kim, Jong Hee
Format: Article
Language:eng
Subjects:
Algorithms
chest X-rays
convolutional neural networks
deep-learning
Hypotheses
Medical research
Metastasis
Neural networks
Pilot projects
Pneumonia
Probability
pulmonary nodules
Radiography
Regression analysis
Scanners
Software
Tuberculosis
volume estimation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The purpose of this study was to assess the volume of the pulmonary nodules and masses on serial chest X-rays (CXRs) from deep-learning-based automatic detection algorithm (DLAD)-based parameters. In a retrospective single-institutional study, 72 patients, who obtained serial CXRs ( = 147) for pulmonary nodules or masses with corresponding chest CT images as the reference standards, were included. A pre-trained DLAD based on a convolutional neural network was developed to detect and localize nodules using 13,710 radiographs and to calculate a localization map and the derived parameters (e.g., the area and mean probability value of pulmonary nodules) for each CXR, including serial follow-ups. For validation, reference 3D CT volumes were measured semi-automatically. Volume prediction models for pulmonary nodules were established through univariable or multivariable, and linear or non-linear regression analyses with the parameters. A polynomial regression analysis was performed as a method of a non-linear regression model. Of the 147 CXRs and 208 nodules of 72 patients, the mean volume of nodules or masses was measured as 9.37 ± 11.69 cm (mean ± standard deviation). The area and CT volume demonstrated a linear correlation of moderate strength (i.e., R = 0.58, RMSE: 9449.9 mm m in a linear regression analysis). The area and mean probability values exhibited a strong linear correlation (R = 0.73). The volume prediction performance based on a multivariable regression model was best with a mean probability and unit-adjusted area (i.e. 7975.6 mm , the smallest among the other variable parameters). The prediction model with the area and the mean probability based on the DLAD showed a rather accurate quantitative estimation of pulmonary nodule or mass volume and the change in serial CXRs.
ISSN:2075-4418
2075-4418