AUTHOR=Chen Zhi , Huang Yu-Hua , Kong Feng-Ming , Ho Wai Yin , Ren Ge , Cai Jing 

TITLE=A super-voxel-based method for generating surrogate lung ventilation images from CT

JOURNAL=Frontiers in Physiology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2023.1085158

DOI=10.3389/fphys.2023.1085158

ISSN=1664-042X

ABSTRACT=<p><bold>Purpose:</bold> This study aimed to develop and evaluate <inline-formula id="inf1"><mml:math id="m1" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula><bold>,</bold> a super-voxel-based method for surrogate computed tomography ventilation imaging (CTVI).</p><p><bold>Methods and Materials:</bold> 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 (<italic>D</italic><sub><italic>mean</italic></sub>) and mean ventilation values (<italic>Vent</italic><sub><italic>mean</italic></sub>), respectively. The final CT-derived ventilation images were generated by interpolation from the <italic>D</italic><sub><italic>mean</italic></sub> values to yield <inline-formula id="inf2"><mml:math id="m2" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula>. For the performance evaluation, the voxel- and region-wise differences between <inline-formula id="inf3"><mml:math id="m3" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> 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, <inline-formula id="inf4"><mml:math id="m4" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">H</mml:mi><mml:mi mathvariant="normal">U</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> and <inline-formula id="inf5"><mml:math id="m5" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">J</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula>, and compared with the SPECT images.</p><p><bold>Results:</bold> The correlation between the <italic>D</italic><sub><italic>mean</italic></sub> and <italic>Vent</italic><sub><italic>mean</italic></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 <inline-formula id="inf6"><mml:math id="m6" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> method achieved a stronger average correlation (0.62 ± 0.10) with SPECT, which was significantly better than the correlations achieved with the <inline-formula id="inf7"><mml:math id="m7" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">H</mml:mi><mml:mi mathvariant="normal">U</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> (0.33 ± 0.14, <italic>p</italic> &lt; 0.05) and <inline-formula id="inf8"><mml:math id="m8" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">J</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> (0.23 ± 0.11, <italic>p</italic> &lt; 0.05) methods. For the region-wise evaluation, the Dice similarity coefficient of the high functional region for <inline-formula id="inf9"><mml:math id="m9" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> (0.63 ± 0.07) was significantly higher than the corresponding values for the <inline-formula id="inf10"><mml:math id="m10" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">H</mml:mi><mml:mi mathvariant="normal">U</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> (0.43 ± 0.08, <italic>p</italic> &lt; 0.05) and <inline-formula id="inf11"><mml:math id="m11" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">J</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> (0.42 ± 0.05, <italic>p</italic> &lt; 0.05) methods.</p><p><bold>Conclusion:</bold> The strong correlation between <inline-formula id="inf12"><mml:math id="m12" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">I</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi><mml:mi mathvariant="normal">V</mml:mi><mml:mi mathvariant="normal">D</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> and SPECT demonstrates the potential usefulness of this novel method of ventilation estimation for surrogate ventilation imaging.</p>