Computed tomography lung volume estimation and its relation to lung capacities and spine deformation

Daghighi, Abtin; Tropp, Hans

doi:10.21037/jss.2018.12.14

Cited by 9 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an additional check, we investigated the relationship between CT-derived and PFT-derived volumes in 137 cases where both were available. This provides results in line with previous studies on CTderived lung volumes 31,32 : Although CT-derived lung volume and TLV are well correlated (r = 0.78), there are considerable differences in some patients. To overcome the issues with the CT-derived labels on the RUMC-real dataset, we further fine-tuned the best networks from that experiment with PFT-derived labels.…”

Section: Discussionsupporting

confidence: 92%

“…As an additional check, we investigated the relationship between CT‐derived and PFT‐derived volumes in 137 cases where both were available. This provides results in line with previous studies on CT‐derived lung volumes 31,32 : Although CT‐derived lung volume and TLV are well correlated ( r = 0.78), there are considerable differences in some patients.…”

Section: Discussionsupporting

confidence: 91%

See 1 more Smart Citation

Automated estimation of total lung volume using chest radiographs and deep learning

et al. 2022

View full text Add to dashboard Cite

Background Total lung volume is an important quantitative biomarker and is used for the assessment of restrictive lung diseases. Purpose In this study, we investigate the performance of several deep‐learning approaches for automated measurement of total lung volume from chest radiographs. Methods About 7621 posteroanterior and lateral view chest radiographs (CXR) were collected from patients with chest CT available. Similarly, 928 CXR studies were chosen from patients with pulmonary function test (PFT) results. The reference total lung volume was calculated from lung segmentation on CT or PFT data, respectively. This dataset was used to train deep‐learning architectures to predict total lung volume from chest radiographs. The experiments were constructed in a stepwise fashion with increasing complexity to demonstrate the effect of training with CT‐derived labels only and the sources of error. The optimal models were tested on 291 CXR studies with reference lung volume obtained from PFT. Mean absolute error (MAE), mean absolute percentage error (MAPE), and Pearson correlation coefficient (Pearson's r) were computed. Results The optimal deep‐learning regression model showed an MAE of 408 ml and an MAPE of 8.1% using both frontal and lateral chest radiographs as input. The predictions were highly correlated with the reference standard (Pearson's r = 0.92). CT‐derived labels were useful for pretraining but the optimal performance was obtained by fine‐tuning the network with PFT‐derived labels. Conclusion We demonstrate, for the first time, that state‐of‐the‐art deep‐learning solutions can accurately measure total lung volume from plain chest radiographs. The proposed model is made publicly available and can be used to obtain total lung volume from routinely acquired chest radiographs at no additional cost. This deep‐learning system can be a useful tool to identify trends over time in patients referred regularly for chest X‐ray.

show abstract

Section: Discussionsupporting

confidence: 92%

Section: Discussionsupporting

confidence: 91%

Automated estimation of total lung volume using chest radiographs and deep learning

et al. 2022

View full text Add to dashboard Cite

show abstract

“…As an additional check we investigated the relationship between CT-derived and PFT-derived volumes in 137 cases where both were available. This provides results in line with previous studies on CT-derived lung volumes 31,32 : although CT-derived lung volume and TLV are well correlated (r=0.78), there are considerable differences in some patients.…”

Section: Discussionsupporting

confidence: 92%

Automated Estimation of Total Lung Volume using Chest Radiographs and Deep Learning

Sogancioglu¹,

Scholten²,

Boulogne³

et al. 2021

Preprint

View full text Add to dashboard Cite

Total lung volume is an important quantitative biomarker and is used for the assessment of restrictive lung diseases. In this study, we investigate the performance of several deep-learning approaches for automated measurement of total lung volume from chest radiographs. 7621 posteroanterior and lateral view chest radiographs (CXR) were collected from patients with chest CT available. Similarly, 928 CXR studies were chosen from patients with pulmonary function test (PFT) results. The reference total lung volume was calculated from lung segmentation on CT or PFT data, respectively. This dataset was used to train deep-learning architectures to predict total lung volume from chest radiographs. The experiments were constructed in a step-wise fashion with increasing complexity to demonstrate the effect of training with CT-derived labels only and the sources of error. The optimal models were tested on 291 CXR studies with reference lung volume obtained from PFT. The optimal deep-learning regression model showed an MAE of 408 ml and an MAPE of 8.1% and Pearson's r = 0.92 using both frontal and lateral chest radiographs as input. CT-derived labels were useful for pre-training but optimal performance was obtained by fine-tuning the network with PFT-derived labels. We demonstrate, for the first time, that state-of-the-art deep learning solutions can accurately measure total lung volume from plain chest radiographs. The proposed model can be used to obtain total lung volume from routinely acquired chest radiographs at no additional cost and could be a useful tool to identify trends over time in patients referred regularly for chest x-rays.

show abstract

“…It is assumed that the restrictive pulmonary function is the result of both malformed costovertebral joints and chest wall deformities due to asymmetrical HO formation along the spine and thoracic cage ( Kaplan et al, 2010 ). One could argue that the severity of the chest wall deformity (kyphosis or scoliosis) should have an impact on the restriction of the thoracic cage and therefore pulmonary function ( Daghighi and Tropp, 2019 ; Farrell and Garrido, 2020 ). Two of the patients with relatively mild thoracic deformities did, however, show severe restricted pulmonary function.…”

Section: Discussionmentioning

confidence: 99%

Deterioration of pulmonary function: An early complication in Fibrodysplasia Ossificans Progressiva

et al. 2021

View full text Add to dashboard Cite

Fibrodysplasia Ossificans Progressiva (FOP) is a genetic disease characterized by the formation of heterotopic ossification (HO) in connective tissues. HO first develops in the thoracic region, before more peripheral sites are affected. Due to HO along the thoracic cage, its movements are restricted and pulmonary function deteriorates. Because development of HO is progressive, it is likely that pulmonary function deteriorates over time, but longitudinal data on pulmonary function in FOP are missing. Longitudinal pulmonary function tests (PFTs) from seven FOP patients were evaluated retrospectively to assess whether there were changes in pulmonary function during aging. Forced vital capacity (FVC), forced expiratory volume in one second (FEV1), total lung capacity (TLC), residual volume (RV) and diffusing lung capacity for carbon dioxide divided by alveolar volume (DLCO/VA) were included. In addition, HO volume along the thorax together with its progression as identified by whole body low dose CT scans were correlated to PFT data. Per patient, aged 7–57 years at the time of the first PFT, three to nine PFTs were available over a period of 6–18 years. Restrictive pulmonary function, identified by TLC or suspected by FVC, was found in all, but one, patients. In three patients, TLC, FVC or both decreased further during the follow-up period. All, but one, patients had an increased RV. The DLCO/VA ratio was normal in all FOP patients. Interestingly, FEV1 increased after a surgical intervention to unlock the jaw. In four out of five patients total HO volume in the thoracic region progressed beyond early adulthood, but no further decline in FVC was observed. In conclusion, restrictive pulmonary function was found in the majority of patients already at an early age. Our data suggest that the deterioration in pulmonary function is age dependent.

show abstract

Computed tomography lung volume estimation and its relation to lung capacities and spine deformation

Cited by 9 publications

References 17 publications

Automated estimation of total lung volume using chest radiographs and deep learning

Automated estimation of total lung volume using chest radiographs and deep learning

Automated Estimation of Total Lung Volume using Chest Radiographs and Deep Learning

Deterioration of pulmonary function: An early complication in Fibrodysplasia Ossificans Progressiva

Contact Info

Product

Resources

About