File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A super-voxel-based method for generating surrogate lung ventilation images from CT

TitleA super-voxel-based method for generating surrogate lung ventilation images from CT
Authors
Keywords4DCT
lung cancer
radiotherapy
super-voxel
ventilation
Issue Date26-Apr-2023
PublisherFrontiers Media
Citation
Frontiers in Physiology, 2023, v. 14 How to Cite?
Abstract

Purpose: This study aimed to develop and evaluate CTVISVD, a super-voxel-based method for surrogate computed tomography ventilation imaging (CTVI).

Methods and Materials: The study used four-dimensional CT (4DCT) and single-photon emission computed tomography (SPECT) images and corresponding lung masks from 21 patients with lung cancer obtained from the Ventilation And Medical Pulmonary Image Registration Evaluation dataset. The lung volume of the exhale CT for each patient was segmented into hundreds of super-voxels using the Simple Linear Iterative Clustering (SLIC) method. These super-voxel segments were applied to the CT and SPECT images to calculate the mean density values (Dmean) and mean ventilation values (Ventmean), respectively. The final CT-derived ventilation images were generated by interpolation from the Dmean values to yield CTVISVD. For the performance evaluation, the voxel- and region-wise differences between CTVISVD and SPECT were compared using Spearman’s correlation and the Dice similarity coefficient index. Additionally, images were generated using two deformable image registration (DIR)-based methods, CTVIHU and CTVIJac, and compared with the SPECT images.

Results: The correlation between the Dmean and Ventmean of the super-voxel was 0.59 ± 0.09, representing a moderate-to-high correlation at the super-voxel level. In the voxel-wise evaluation, the CTVISVD method achieved a stronger average correlation (0.62 ± 0.10) with SPECT, which was significantly better than the correlations achieved with the CTVIHU (0.33 ± 0.14, p < 0.05) and CTVIJac (0.23 ± 0.11, p < 0.05) methods. For the region-wise evaluation, the Dice similarity coefficient of the high functional region for CTVISVD (0.63 ± 0.07) was significantly higher than the corresponding values for the CTVIHU (0.43 ± 0.08, p < 0.05) and CTVIJac (0.42 ± 0.05, p < 0.05) methods.

Conclusion: The strong correlation between CTVISVD and SPECT demonstrates the potential usefulness of this novel method of ventilation estimation for surrogate ventilation imaging.


Persistent Identifierhttp://hdl.handle.net/10722/343803
ISSN
2023 Impact Factor: 3.2
2023 SCImago Journal Rankings: 1.006

 

DC FieldValueLanguage
dc.contributor.authorChen, Zhi-
dc.contributor.authorHuang, Yu-Hua-
dc.contributor.authorKong, Feng-Ming-
dc.contributor.authorHo, Wai Yin-
dc.contributor.authorRen, Ge-
dc.contributor.authorCai, Jing-
dc.date.accessioned2024-06-11T07:51:44Z-
dc.date.available2024-06-11T07:51:44Z-
dc.date.issued2023-04-26-
dc.identifier.citationFrontiers in Physiology, 2023, v. 14-
dc.identifier.issn1664-042X-
dc.identifier.urihttp://hdl.handle.net/10722/343803-
dc.description.abstract<p><strong>Purpose:</strong> This study aimed to develop and evaluate CTVISVD<strong>,</strong> a super-voxel-based method for surrogate computed tomography ventilation imaging (CTVI).</p><p><strong>Methods and Materials:</strong> The study used four-dimensional CT (4DCT) and single-photon emission computed tomography (SPECT) images and corresponding lung masks from 21 patients with lung cancer obtained from the Ventilation And Medical Pulmonary Image Registration Evaluation dataset. The lung volume of the exhale CT for each patient was segmented into hundreds of super-voxels using the Simple Linear Iterative Clustering (SLIC) method. These super-voxel segments were applied to the CT and SPECT images to calculate the mean density values (<em>D</em><sub><em>mean</em></sub>) and mean ventilation values (<em>Vent</em><sub><em>mean</em></sub>), respectively. The final CT-derived ventilation images were generated by interpolation from the <em>D</em><sub><em>mean</em></sub> values to yield CTVISVD. For the performance evaluation, the voxel- and region-wise differences between CTVISVD and SPECT were compared using Spearman’s correlation and the Dice similarity coefficient index. Additionally, images were generated using two deformable image registration (DIR)-based methods, CTVIHU and CTVIJac, and compared with the SPECT images.</p><p><strong>Results:</strong> The correlation between the <em>D</em><sub><em>mean</em></sub> and <em>Vent</em><sub><em>mean</em></sub> of the super-voxel was 0.59 ± 0.09, representing a moderate-to-high correlation at the super-voxel level. In the voxel-wise evaluation, the CTVISVD method achieved a stronger average correlation (0.62 ± 0.10) with SPECT, which was significantly better than the correlations achieved with the CTVIHU (0.33 ± 0.14, <em>p</em> < 0.05) and CTVIJac (0.23 ± 0.11, <em>p</em> < 0.05) methods. For the region-wise evaluation, the Dice similarity coefficient of the high functional region for CTVISVD (0.63 ± 0.07) was significantly higher than the corresponding values for the CTVIHU (0.43 ± 0.08, <em>p</em> < 0.05) and CTVIJac (0.42 ± 0.05, <em>p</em> < 0.05) methods.</p><p><strong>Conclusion:</strong> The strong correlation between CTVISVD and SPECT demonstrates the potential usefulness of this novel method of ventilation estimation for surrogate ventilation imaging.</p>-
dc.languageeng-
dc.publisherFrontiers Media-
dc.relation.ispartofFrontiers in Physiology-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.subject4DCT-
dc.subjectlung cancer-
dc.subjectradiotherapy-
dc.subjectsuper-voxel-
dc.subjectventilation-
dc.titleA super-voxel-based method for generating surrogate lung ventilation images from CT-
dc.typeArticle-
dc.identifier.doi10.3389/fphys.2023.1085158-
dc.identifier.scopuseid_2-s2.0-85159064119-
dc.identifier.volume14-
dc.identifier.eissn1664-042X-
dc.identifier.issnl1664-042X-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats