Multi-resolution convolutional neural networks for fully automated segmentation of acutely injured lungs in multiple species

Gerard, Sarah E.; Herrmann, Jacob; Kaczka, David W.; Musch, Guido; Fernandez-Bustamante, Ana; Reinhardt, Joseph M.

doi:10.1016/j.media.2019.101592

Cited by 70 publications

(58 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…No recruitment maneuver was performed. Lung parenchyma and vessel segmentations were obtained using multi-resolution convolutional neural networks [ 18 ], followed by manual refinement if necessary. Also, three regions of interests (ROIs) of equal lung tissue weight [ 19 , 20 ] were obtained along the ventral-dorsal and craniocaudal axes.…”

Section: Methodsmentioning

confidence: 99%

Computed tomography assessment of PEEP-induced alveolar recruitment in patients with severe COVID-19 pneumonia

et al. 2021

Self Cite

View full text Add to dashboard Cite

Background There is a paucity of data concerning the optimal ventilator management in patients with COVID-19 pneumonia; particularly, the optimal levels of positive-end expiratory pressure (PEEP) are unknown. We aimed to investigate the effects of two levels of PEEP on alveolar recruitment in critically ill patients with severe COVID-19 pneumonia. Methods A single-center cohort study was conducted in a 39-bed intensive care unit at a university-affiliated hospital in Genoa, Italy. Chest computed tomography (CT) was performed to quantify aeration at 8 and 16 cmH2O PEEP. The primary endpoint was the amount of alveolar recruitment, defined as the change in the non-aerated compartment at the two PEEP levels on CT scan. Results Forty-two patients were included in this analysis. Alveolar recruitment was median [interquartile range] 2.7 [0.7–4.5] % of lung weight and was not associated with excess lung weight, PaO2/FiO2 ratio, respiratory system compliance, inflammatory and thrombophilia markers. Patients in the upper quartile of recruitment (recruiters), compared to non-recruiters, had comparable clinical characteristics, lung weight and gas volume. Alveolar recruitment was not different in patients with lower versus higher respiratory system compliance. In a subgroup of 20 patients with available gas exchange data, increasing PEEP decreased respiratory system compliance (median difference, MD − 9 ml/cmH2O, 95% CI from − 12 to − 6 ml/cmH2O, p < 0.001) and the ventilatory ratio (MD − 0.1, 95% CI from − 0.3 to − 0.1, p = 0.003), increased PaO2 with FiO2 = 0.5 (MD 24 mmHg, 95% CI from 12 to 51 mmHg, p < 0.001), but did not change PaO2 with FiO2 = 1.0 (MD 7 mmHg, 95% CI from − 12 to 49 mmHg, p = 0.313). Moreover, alveolar recruitment was not correlated with improvement of oxygenation or venous admixture. Conclusions In patients with severe COVID-19 pneumonia, higher PEEP resulted in limited alveolar recruitment. These findings suggest limiting PEEP strictly to the values necessary to maintain oxygenation, thus avoiding the use of higher PEEP levels.

show abstract

Section: Methodsmentioning

confidence: 99%

Computed tomography assessment of PEEP-induced alveolar recruitment in patients with severe COVID-19 pneumonia

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…In order to standardise the acquisition of CT-based parameters and even more to ensure a fast determination of these parameters in clinical use, ARDS-specific lung analysis software would be desirable. In this regard, recent developments in the field of neuronal networks seem very promising [ 24 ].…”

Section: Discussionmentioning

confidence: 99%

A quantitative CT parameter for the assessment of pulmonary oedema in patients with acute respiratory distress syndrome

et al. 2020

View full text Add to dashboard Cite

Objectives The aim of this study was to establish quantitative CT (qCT) parameters for pathophysiological understanding and clinical use in patients with acute respiratory distress syndrome (ARDS). The most promising parameter is introduced. Materials and methods 28 intubated patients with ARDS obtained a conventional CT scan in end-expiratory breathhold within the first 48 hours after admission to intensive care unit (ICU). Following manual segmentation, 137 volume- and lung weight-associated qCT parameters were correlated with 71 clinical parameters such as blood gases, applied ventilation pressures, pulse contour cardiac output measurements and established status and prognosis scores (SOFA, SAPS II). Results Of all examined qCT parameters, excess lung weight (ELW), i.e. the difference between a patient’s current lung weight and the virtual lung weight of a healthy person at the same height, displayed the most significant results. ELW correlated significantly with the amount of inflated lung tissue [%] (p<0.0001; r = -0.66) and was closely associated with the amount of extravascular lung water (EVLW) (p<0.0001; r = 0.72). More substantially than the oxygenation index (PaO2/FiO2) or any other clinical parameter it correlated with the patients’ mean SOFA- (p<0.0001, r = 0.69) and SAPS II-Score (p = 0.0005, r = 0.62). Patients who did not survive intensive care treatment displayed higher values of ELW in the initial CT scans. Conclusions ELW could serve as a non-invasive method to quantify the amount of pulmonary oedema. It might serve as an early radiological marker of severity in patients with ARDS.

show abstract

“…Each volumetric image in the temporal sequence was labeled I n (x n ), where n indexes the number of images in the sequence 0 through N − 1, and x n is a vector representing 3-dimensional spatial position. Voxels corresponding to spatial positions within the lungs were identified by a fully automated segmentation algorithm using a deep convolutional neural network (Gerard et al, 2018(Gerard et al, , 2020, generating a distinct lung mask M n (x n ) for each image phase. The neural network was trained using manually segmented lungs in CT images obtained from multiple datasets of experimental lung injury models, including a subset of images from the current study.…”

Section: Image Acquisition and Processingmentioning

confidence: 99%

Quantifying Regional Lung Deformation Using Four-Dimensional Computed Tomography: A Comparison of Conventional and Oscillatory Ventilation

et al. 2020

Self Cite

View full text Add to dashboard Cite

Mechanical ventilation strategies that reduce the heterogeneity of regional lung stress and strain may reduce the risk of ventilator-induced lung injury (VILI). In this study, we used registration of four-dimensional computed tomographic (4DCT) images to assess regional lung aeration and deformation in 10 pigs under baseline conditions and following acute lung injury induced with oleic acid. CT images were obtained via dynamic axial imaging (Siemens SOMATOM Force) during conventional pressure-controlled mechanical ventilation (CMV), as well as high-frequency and multi-frequency oscillatory ventilation modalities (HFOV and MFOV, respectively). Our results demonstrate that oscillatory modalities reduce intratidal strain throughout the lung in comparison to conventional ventilation, as well as the spatial gradients of dynamic strain along the dorsal-ventral axis. Harmonic distortion of parenchymal deformation was observed during HFOV with a single discrete sinusoid delivered at the airway opening, suggesting inherent mechanical nonlinearity of the lung tissues. MFOV may therefore provide improved lung-protective ventilation by reducing strain magnitudes and spatial gradients of strain compared to either CMV or HFOV.

show abstract

Multi-resolution convolutional neural networks for fully automated segmentation of acutely injured lungs in multiple species

Cited by 70 publications

References 37 publications

Computed tomography assessment of PEEP-induced alveolar recruitment in patients with severe COVID-19 pneumonia

Computed tomography assessment of PEEP-induced alveolar recruitment in patients with severe COVID-19 pneumonia

A quantitative CT parameter for the assessment of pulmonary oedema in patients with acute respiratory distress syndrome

Quantifying Regional Lung Deformation Using Four-Dimensional Computed Tomography: A Comparison of Conventional and Oscillatory Ventilation

Contact Info

Product

Resources

About