CT-derived Biomechanical Metrics Improve Agreement Between Spirometry and Emphysema

Bhatt, Surya P.; Bodduluri, Sandeep; Newell, John D.; Hoffman, Eric A.; Sieren, Jessica C.; Han, MeiLan K.; Reinhardt, Joseph M.

doi:10.1016/j.acra.2016.02.002

Cited by 28 publications

(25 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in a given patient, it is usual to see multiple subtypes of emphysema such as centrilobular and panlobular emphysema with differential impact on lung function 22–24. The degree of CT emphysema does not translate linearly into airflow obstruction,3 10 and it is likely that different types and distribution patterns of emphysema contribute differentially to airflow obstruction 22–24. Although we did not study emphysema subtypes, the mean Jacobian determinant, by offering information of regional volume change with respiration, likely reflects a more direct physiological link between structural lung disease and the lung mechanics reflected by spirometry.…”

Section: Discussionmentioning

confidence: 99%

“…A sum of squared tissue volume difference method was used as a similarity metric. This method has been shown previously to be effective in lung image registration protocols 10 11. The transformation matrix from the registration process was used to derive regional tissue expansion and contraction measures between TLC and FRC volumes.…”

Section: Methodsmentioning

confidence: 99%

“…A substantial number of patients with COPD, however, have significant discordance between spirometry and CT findings. We recently demonstrated that a biomechanical measure of structural lung disease (the Jacobian determinant, a measure of local lung volume change assessed through matching images acquired at multiple lung volumes) substantially explains the differences between static single-volume CT images and airflow obstruction on spirometry 10 11. However, it remains important to determine whether the Jacobian determinant is associated with standard clinical measures of patient outcomes.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Biomechanical CT metrics are associated with patient outcomes in COPD

Bodduluri

Bhatt

Hoffman

et al. 2017

Thorax

Self Cite

View full text Add to dashboard Cite

Background Traditional metrics of lung disease such as those derived from spirometry and static single-volume CT images are used to explain respiratory morbidity in patients with chronic obstructive pulmonary disease (COPD), but are insufficient. We hypothesized that the mean Jacobian determinant, a measure of local lung expansion and contraction with respiration, would contribute independently to clinically relevant functional outcomes. Methods We applied image registration techniques to paired inspiratory-expiratory CT scans and derived the Jacobian determinant of the deformation field between the two lung volumes to map local volume change with respiration. We analyzed 490 participants with COPD with multivariable regression models to assess strengths of association between traditional CT metrics of disease and the Jacobian determinant with respiratory morbidity including dyspnea (mMRC), St Georges Respiratory Questionnaire (SGRQ) score, six-minute walk distance (6MWD), and the BODE index, as well as all-cause mortality. Results The Jacobian determinant was significantly associated with SGRQ (adjusted regression co-efficient β = −11.75,95%CI −21.6 to −1.7;p=0.020), and with 6MWD (β=321.15, 95%CI 134.1 to 508.1;p<0.001), independent of age, sex, race, body-mass-index, FEV1, smoking pack-years, CT emphysema, CT gas trapping, airway wall thickness, and CT scanner protocol. The mean Jacobian determinant was also independently associated with the BODE index (β= −0.41, 95%CI −0.80 to −0.02; p = 0.039), and mortality on follow-up (adjusted hazards ratio = 4.26, 95%CI = 0.93 to 19.23; p = 0.064). Conclusion Biomechanical metrics representing local lung expansion and contraction improve prediction of respiratory morbidity and mortality and offer additional prognostic information beyond traditional measures of lung function and static single-volume CT metrics.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Biomechanical CT metrics are associated with patient outcomes in COPD

Bodduluri

Bhatt

Hoffman

et al. 2017

Thorax

Self Cite

View full text Add to dashboard Cite

show abstract

“…This method has been shown previously to be effective in lung image registration protocols. [92,93] The transformation matrix from the registration process was used to derive regional tissue expansion and contraction measures between TLC and FRC volumes. In this study, we used the Jacobian determinant metric to represent regional deformation patterns in COPD patients.…”

Section: Ct Image-based Measuresmentioning

confidence: 99%

“…We recently demonstrated that a biomechanical measure of structural lung disease (the Jacobian determinant, an elasticity measure of local lung volume change assessed through matching images acquired at multiple lung volumes) substantially explains the differences between static CT images and airflow obstruction on spirometry. [92,93] However, it remains important to determine if the Jacobian determinant is associated with standard clinical measures of patient outcomes. Such determination will enhance its use in better phenotyping patients.…”

Section: Introductionmentioning

confidence: 99%

CT image registration-based lung mechanics In COPD

Bodduluri¹

Self Cite

View full text Add to dashboard Cite

Direct estimation of regional lung volume change from paired and single CT images using residual regression neural network

et al. 2023

Self Cite

View full text Add to dashboard Cite

BackgroundChest computed tomography (CT) enables characterization of pulmonary diseases by producing high‐resolution and high‐contrast images of the intricate lung structures. Deformable image registration is used to align chest CT scans at different lung volumes, yielding estimates of local tissue expansion and contraction.PurposeWe investigated the utility of deep generative models for directly predicting local tissue volume change from lung CT images, bypassing computationally expensive iterative image registration and providing a method that can be utilized in scenarios where either one or two CT scans are available.MethodsA residual regression convolutional neural network, called Reg3DNet+, is proposed for directly regressing high‐resolution images of local tissue volume change (i.e., Jacobian) from CT images. Image registration was performed between lung volumes at total lung capacity (TLC) and functional residual capacity (FRC) using a tissue mass‐ and structure‐preserving registration algorithm. The Jacobian image was calculated from the registration‐derived displacement field and used as the ground truth for local tissue volume change. Four separate Reg3DNet+ models were trained to predict Jacobian images using a multifactorial study design to compare the effects of network input (i.e., single image vs. paired images) and output space (i.e., FRC vs. TLC). The models were trained and evaluated on image datasets from the COPDGene study. Models were evaluated against the registration‐derived Jacobian images using local, regional, and global evaluation metrics.ResultsStatistical analysis revealed that both factors – network input and output space – were significant determinants for change in evaluation metrics. Paired‐input models performed better than single‐input models, and model performance was better in the output space of FRC rather than TLC. Mean structural similarity index for paired‐input models was 0.959 and 0.956 for FRC and TLC output spaces, respectively, and for single‐input models was 0.951 and 0.937. Global evaluation metrics demonstrated correlation between registration‐derived Jacobian mean and predicted Jacobian mean: coefficient of determination (r2) for paired‐input models was 0.974 and 0.938 for FRC and TLC output spaces, respectively, and for single‐input models was 0.598 and 0.346. After correcting for effort, registration‐derived lobar volume change was strongly correlated with the predicted lobar volume change: for paired‐input models r2 was 0.899 for both FRC and TLC output spaces, and for single‐input models r2 was 0.803 and 0.862, respectively.ConclusionsConvolutional neural networks can be used to directly predict local tissue mechanics, eliminating the need for computationally expensive image registration. Networks that use paired CT images acquired at TLC and FRC allow for more accurate prediction of local tissue expansion compared to networks that use a single image. Networks that only require a single input image still show promising results, particularly after correcting for effort, and allow for local tissue expansion estimation in cases where multiple CT scans are not available. For single‐input networks, the FRC image is more predictive of local tissue volume change compared to the TLC image.

show abstract

CT-derived Biomechanical Metrics Improve Agreement Between Spirometry and Emphysema

Cited by 28 publications

References 37 publications

Biomechanical CT metrics are associated with patient outcomes in COPD

Biomechanical CT metrics are associated with patient outcomes in COPD

CT image registration-based lung mechanics In COPD

Direct estimation of regional lung volume change from paired and single CT images using residual regression neural network

Contact Info

Product

Resources

About