Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning.
<h4>Purpose</h4>To conduct a volumetric and movement analysis of lung parenchyma in prone positioning using deep neural networks (DNNs).<h4>Method</h4>We included patients with suspected interstitial lung abnormalities or disease who underwent full-inspiratory supine and pron...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2024-01-01
|
Series: | PLoS ONE |
Online Access: | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0299366&type=printable |
_version_ | 1797266396219965440 |
---|---|
author | Hyungin Park Soon Ho Yoon |
author_facet | Hyungin Park Soon Ho Yoon |
author_sort | Hyungin Park |
collection | DOAJ |
description | <h4>Purpose</h4>To conduct a volumetric and movement analysis of lung parenchyma in prone positioning using deep neural networks (DNNs).<h4>Method</h4>We included patients with suspected interstitial lung abnormalities or disease who underwent full-inspiratory supine and prone chest CT at a single institution between June 2021 and March 2022. A thoracic radiologist visually assessed the fibrosis extent in the total lung (using units of 10%) on supine CT. After preprocessing the images into 192×192×192 resolution, a DNN automatically segmented the whole lung and pulmonary lobes in prone and supine CT images. Affine registration matched the patient's center and location, and the DNN deformably registered prone and supine CT images to calculate the x-, y-, z-axis, and 3D pixel movements.<h4>Results</h4>In total, 108 CT pairs had successful registration. Prone positioning significantly increased the left lower (90.2±69.5 mL, P = 0.000) and right lower lobar volumes (52.5±74.2 mL, P = 0.000). During deformable registration, the average maximum whole-lung pixel movements between the two positions were 1.5, 1.9, 1.6, and 2.8 cm in each axis and 3D plane. Compared to patients with <30% fibrosis, those with ≥30% fibrosis had smaller volume changes (P<0.001) and smaller pixel movements in all axes between the positions (P = 0.000-0.007). Forced vital capacity (FVC) correlated with the left lower lobar volume increase (Spearman correlation coefficient, 0.238) and the maximum whole-lung pixel movements in all axes (coefficients, 0.311 to 0.357).<h4>Conclusions</h4>Prone positioning led to the preferential expansion of the lower lobes, correlated with FVC, and lung fibrosis limited lung expansion during prone positioning. |
first_indexed | 2024-03-07T16:29:04Z |
format | Article |
id | doaj.art-f3ef1510b272419981d8818cd0be0cad |
institution | Directory Open Access Journal |
issn | 1932-6203 |
language | English |
last_indexed | 2024-04-25T01:00:01Z |
publishDate | 2024-01-01 |
publisher | Public Library of Science (PLoS) |
record_format | Article |
series | PLoS ONE |
spelling | doaj.art-f3ef1510b272419981d8818cd0be0cad2024-03-11T05:32:13ZengPublic Library of Science (PLoS)PLoS ONE1932-62032024-01-01192e029936610.1371/journal.pone.0299366Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning.Hyungin ParkSoon Ho Yoon<h4>Purpose</h4>To conduct a volumetric and movement analysis of lung parenchyma in prone positioning using deep neural networks (DNNs).<h4>Method</h4>We included patients with suspected interstitial lung abnormalities or disease who underwent full-inspiratory supine and prone chest CT at a single institution between June 2021 and March 2022. A thoracic radiologist visually assessed the fibrosis extent in the total lung (using units of 10%) on supine CT. After preprocessing the images into 192×192×192 resolution, a DNN automatically segmented the whole lung and pulmonary lobes in prone and supine CT images. Affine registration matched the patient's center and location, and the DNN deformably registered prone and supine CT images to calculate the x-, y-, z-axis, and 3D pixel movements.<h4>Results</h4>In total, 108 CT pairs had successful registration. Prone positioning significantly increased the left lower (90.2±69.5 mL, P = 0.000) and right lower lobar volumes (52.5±74.2 mL, P = 0.000). During deformable registration, the average maximum whole-lung pixel movements between the two positions were 1.5, 1.9, 1.6, and 2.8 cm in each axis and 3D plane. Compared to patients with <30% fibrosis, those with ≥30% fibrosis had smaller volume changes (P<0.001) and smaller pixel movements in all axes between the positions (P = 0.000-0.007). Forced vital capacity (FVC) correlated with the left lower lobar volume increase (Spearman correlation coefficient, 0.238) and the maximum whole-lung pixel movements in all axes (coefficients, 0.311 to 0.357).<h4>Conclusions</h4>Prone positioning led to the preferential expansion of the lower lobes, correlated with FVC, and lung fibrosis limited lung expansion during prone positioning.https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0299366&type=printable |
spellingShingle | Hyungin Park Soon Ho Yoon Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. PLoS ONE |
title | Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. |
title_full | Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. |
title_fullStr | Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. |
title_full_unstemmed | Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. |
title_short | Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning. |
title_sort | deep learning segmentation and registration driven lung parenchymal volume and movement ct analysis in prone positioning |
url | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0299366&type=printable |
work_keys_str_mv | AT hyunginpark deeplearningsegmentationandregistrationdrivenlungparenchymalvolumeandmovementctanalysisinpronepositioning AT soonhoyoon deeplearningsegmentationandregistrationdrivenlungparenchymalvolumeandmovementctanalysisinpronepositioning |