Translator Disclaimer
1 September 2009 Simultaneous three-dimensional optical coherence tomography and intravital microscopy for imaging subpleural pulmonary alveoli in isolated rabbit lungs
Author Affiliations +
Abstract
There is a growing interest in analyzing lung mechanics at the level of the alveoli in order to understand stress-related pathogenesis and possibly avoid ventilator associated lung injury. Emerging quantitative models to simulate fluid mechanics and the associated stresses and strains on delicate alveolar walls require realistic quantitative input on alveolar geometry and its dynamics during ventilation. Here, three-dimensional optical coherence tomography (OCT) and conventional intravital microscopy are joined in one setup to investigate the geometric changes of subpleural alveoli during stepwise pressure increase and release in an isolated and perfused rabbit lung model. We describe good qualitative agreement and quantitative correlation between the OCT data and video micrographs. Our main finding is the inflation and deflation of individual alveoli with noticeable hysteresis. Importantly, this three-dimensional geometry data can be extracted and converted into input data for numerical simulations.

1.

Introduction

1.1.

Medical Background

During intensive care, mechanical ventilation of the patient often becomes necessary to avoid hypoxia and hypercapnia. Ventilator settings, which mainly control breathing rate, tidal volume, and airway pressure, are commonly chosen from standard operating procedures that take body weight and oxygenation into account but do not give much consideration to lung mechanics. In a clinical study of patients suffering from adult respiratory distress syndrome,1 the group who was ventilated with a lower tidal volume per body weight showed better survival and outcome than the conventionally ventilated group, suggesting that the normally chosen tidal volume is harmful to the preinjured lung.2 The potential for ventilator-associated lung injury (VALI) and ventilator-induced lung injury (VILI) has propelled intensive research into the underlying mechanisms.3, 4 Mechanical stress on the lung parenchyma has been identified as a major contributor to development of VILI/VALI. This stress not only produces barotrauma and volume trauma (i.e., the rupture of alveolar walls), but also so-called biotrauma from proinflammatory signaling by the mechanically stressed alveolar epithelial and endothelium cells or from the stressed alveolar-capillary barrier. Because shear stress and stretch acting on the alveolar walls cannot be directly measured, it is essential to simply observe the geometric deformations of alveolar structures during ventilation and relate them to the ventilator settings.

1.2.

Methods for Imaging Pulmonary Alveoli

Computer tomography (CT), the routine diagnostic tool for lung injury,5 lacks the resolution for imaging single alveoli and micro-CT need fixed and stained tissue to make measurements on an alveolar scale. Magnet resonance imaging with hyperpolarized helium only indirectly allows evaluating alveolar size.6 The resolution necessary to image single alveoli (diameter=50100μm) can presently only be reached by optical methods such as in vivo video microscopy (IVM) of subpleural alveoli in small animal models.7, 8 Two-dimensional IVM images provide information in the form of bright reflections determining the projection of the outer diameter of the alveoli. These reflections can be caused by alveolar walls but also by the capillary network in the pleura or by a pleura-enforcing collagen network. Thus, interpretation of these images is not unambiguous. As a result, controversy and open questions still exist about alveolar mechanics, particularly regarding alveolar deformation during volume uptake. At least four mechanisms—all consistent with the pressure volume curve of the whole lung—have been hypothesized: (i) opening/recruitment of additional collapsed alveoli, (ii) balloonlike stretching of the alveolar wall, (iii) unfolding of the alveolar wall, and (iv) expansion of the alveolar duct connecting the alveoli.4, 9 Direct experimental evidence of the existence of all or some of these deformation mechanisms is required before clarifying how they influence the alveolar stress level and the development of biotrauma.

To avoid the problems of IVM, particularly the image artifacts and general uncertainty about whether the optical image corresponds to actual alveolar geometry, we propose three-dimensional (3-D) alveolar imaging via optical coherence tomography (OCT). OCT provides 3-D imaging of air-filled subpleural lung tissue up to a depth of 500μm with a resolution of <10μm . Therefore, alveolar areas can be quantified in several accurately defined spatial planes beneath the pleura. We previously acquired cross-sectional images of subpleural alveoli in the ventilated and perfused rabbit lung and found a qualitative difference in alveolar mechanics between high tidal volume ventilation with zero end-expiratory pressure (ZEEP) and low tidal volume ventilation with high positive end-expiratory pressure (PEEP).10 Although ZEEP led to repetitive alveolar collapse and recruitment, PEEP maintained stable alveolar opening. Quantitative analysis of the alveolar cross sections, however, was hindered by shifting of the alveoli in and out of the plane during tidal ventilation that produced errors in calculating the air-filled area. Therefore, here we present 3-D OCT11, 12 of the perfused isolated rabbit lung with stepwise increased constant positive airway pressure (CPAP). IVM images of the subpleural alveoli were also simultaneously acquired in order to make direct qualitative and quantitative comparisons between the two methods.

2.

Material and Methods

2.1.

Isolated Rabbit Lungs

Use of isolated, perfused, and ventilated rabbit lung as a model has been previously described.10 All experiments were performed in accordance with Ref. 13. The lungs were initially ventilated by volume-controlled ventilation for 20min using a KTR 4 small-animal ventilator (Hugo-Sachs, Freiburg, Germany) with a 3-cm H2O PEEP. Respiratory rate was set at 30min and tidal volume at 8mlkg body weight (bw). Spirometry was carried out using a Fleisch pneumotachograph (Hugo-Sachs, Freiburg, Germany). CPAP was applied and stepwise increased to 25cm H2O and then stepwise decreased to 5cm H2O . Pressurized air and a PEEP valve was used to adjust CPAP levels. At each CPAP level (5, 10, 15, 20, 25, 20, 15, 10, 5cm H2O ), a 3-D-OCT (200×200×512pixels800×800×2048μm3) scan was performed and an IVM micrograph (1600×1200pixels1500×1150μm2) was taken.

2.2.

Combined OCT and IVM

The combined OCT-IVM system setup is shown in Fig. 1 . This system is also suitable for in vivo investigation, as previously reported.14 Compared to our formerly used OCT system, a microscope imaging system has been added and the system has been improved in terms of adding a galvanometer y -scanner axis, higher acquisition speed, and doubled measurement depth. The OCT light source is a fiber-coupled superluminescence diode with 1-mW output power and a 50-nm (FWHM) broad spectrum centered at 840nm (Superlum, Moscow, Russia). The near-infrared light is guided to the scanner head with an integrated 50:50 (sample arm: reference arm) nonpolarizing beamsplitter, a reference arm adjustable in length and attenuation, two galvanometer mounted scanner mirrors (Cambridge Technology Inc., Lexington, Kentucky), a dichroic mirror separating the visible microscope light from the near-infrared light used for OCT, and a focusing objective. The backscattered near-infrared sample light is collected into the interferometer, superimposed with the reference light, and guided to the grating spectrometer. The interference spectrum is focused onto a 2048-pixel -wide charge-coupled device line camera with a 25-MHz pixel clock, 14-bit analog-to-digital conversion, and a FireWire connection to the computer (Micro-Epsilon Optronic, Langebrueck, Germany). In addition, an IVM is integrated into the system. The sample is illuminated by a ring of white light-emitting diodes. The diffuse visible reflections from the sample are imaged through the objective, the dichroic mirror, and two additional lenses onto a color complementary oxide semiconductor camera (Sumix, Oceanside, California), which is mounted to the scanner head unit and connected to the computer via a universal serial bus. The integrated OCT and IVM imaging unit and the mechanical ventilator system are all connected to the same computer. The OCT-IVM acquisition unit is mounted on a ball-and-socket mount combined with three motorized translation stages for positioning. Synchronized computer programs implemented in LabVIEW 8.2 (National Instruments, Austin, Texas) are used to acquire respiratory data (airway pressure, tidal volume, arterial pressure), microscope images, and interference spectra for OCT, and to control the translation stages and galvanometer scanners. Rescaling of the interference spectra to wave number, spectral shaping, fast Fourier transformation, and display as gray value images can either be done online for a cross-sectional preview B-mode image or offline after acquiring the spectra and saving them on hard disk. Additional image processing (despeckle, resclicing, thresholding, etc.) and quantification was performed with ImageJ (ImageJ 1.37v, National Institutes of Health, Bethesda, Maryland) and IMAQ Vision 8.2.1 (National Instruments, Austin, Texas). The IVM images were analyzed with a tablet monitor system, allowing manual tracing of alveolar boundaries with a pen. AMIRA (AMIRA 4.1.0) image analyzing software was used for 3-D reconstruction of alveolar parenchyma.

Fig. 1

Experimental setup for 3-D imaging of subpleural alveoli in isolated and perfused rabbit lungs. A fiber coupler and single-mode fibers (NIR SMF) connect the light source (SLD) centered at 840nm (FWHM 50nm ), the interferometer integrated in the scanner head (SH) and the spectrometer. The near-infrared light (NIR) is coupled in the interferometer by a collimator and divided into reference and sample beam by a beamsplitter cube. The sample beam is deflected by a galvanometer scanner unit (GU) for the x and y direction and focused by an objective (L2) to the sample. The backscattered light is superimposed with the reference light, focused (L1) and reflected by the reference mirror (RM), and coupled in the SMF and transmitted to the spectrometer, where it is spectrally resolved by a grating (1100linesmm) . Interference spectrum is detected by a CCD line (2048pixels) with an acquisition rate of 12kHz (A-scan rate). The system provides a lateral resolution of 7μm and an axial one of 8μm in air ( 6μm in tissue, n=0 1.33). A dichroic mirror (DM) separates the light for IVM, which is then focused with lenses L3 and L4. The IVM image is acquired by a CMOS camera, which is also integrated in the SH. A ring light, consisting of 12 superhigh light-emitting diodes, is positioned at the front of the SH above the sample for illumination. The SH is mounted on a translation stage to carry out optimal positioning in front of the isolated lung. A conventional personal computer was used for data acquisition, position control, and image processing.

054020_1_027905jbo1.jpg

2.3.

Image Processing

OCT image stacks were tilted until the plane of view was parallel to the pleura. Five slices of the stack at 40, 44, 48, 52, and 56μm , beneath the strongest pleura reflection were taken for en face sectional analysis (Fig. 2 ). The image was despeckled by taking the median of the five planes along the z -axis and a 3×3 kernel median filter in the x-y plane. A histogram stretch to the full gray value dynamic range was performed. The resulting image [Fig. 2b] was inverted and filtered by a threshold filter based on a 32×32 kernel gray value histogram analysis yielding a binary image. Small areas were omitted and small holes (one pixel) were filled. The resulting image was used as a segmentation mask for separating tissue from air. Quantification of the areas of the air-filled alveolar structures was performed. The IVM images were processed with a pixel-based graphics program (Adobe Photoshop 7.0) and a graphics tablet monitor. Bright boundaries were identified as alveolar walls, manually retraced, and flood filled with the electronic pen. The resulting air-filled alveolar structures were quantified in the same way as the OCT images. IVM images and OCT sectional planes were registered by applying a fixed rescaling to the OCT images and a rigid manual registration with the IVM image. Air-filled structures present in both image types at all pressure levels were selected for quantitative comparison. Structures in contact with the image boundaries at any pressure level were not analyzed.

Fig. 2

Individual steps in the analysis of OCT images. (a) Volume rendered OCT image stack of subpleural alveolar lung parenchyma of an isolated rabbit lung with an applied CPAP of 25cm H2O . The dashed white line denotes the cutting plane of the extracted OCT en face image shown in (b). (b) Median of the five separated OCT en face images. The air-filled regions are black and the alveolar wall tissue is gray. (c) Automatic segmentation of the air-filled regions.

054020_1_027905jbo2.jpg

3.

Results

3.1.

Alveolar Expansion in IVM and OCT Images

To quantify alveolar expansion using images gathered via OCT, we performed a series of steps (Fig. 2). First, the volume of a 3-D OCT image series was rendered [Fig. 2a]. OCT en face images were then extracted from a plane beneath the pleura visceralis [Fig. 2a, dashed white line], which was clearly visible as bright reflection. Here we show a representative OCT en face image from 45μm beneath the pleura [Fig. 2b]. Automatic segmentation was then used to highlight the air-filled areas before quantification [Fig. 2c].

To compare the effectiveness of OCT versus IVM, we simultaneously obtained OCT and IVM images from identical alveolar tissue at CPAPs of 5cm H2O and 25cm H2O (Fig. 3 ). In the IVM images, circular and convex areas were surrounded by bright boundaries. At 5cm H2O CPAP, the single circular structures had diameters of 50μm and the convex structures with several indentations were a few hundred micrometers. We thus identified these bordered areas as alveolar clusters in which the individual alveoli create the convex bulges. Qualitatively, as pressure increased the diameter of the cluster, the bulges increased and the indentations become more pronounced. Expansion caused by higher CPAP levels was noticeable using both OCT and IVM; however, the OCT technique overestimated alveolar septum thickness.

Fig. 3

Identical alveolar tissue acquired by IVM (a) and (b) or OCT (c) and (d) for an applied CPAP of 5 (a) and (c) and 25cm H2O (b) and (d). OCT en face images (d) representing alveolar structures of 45μm beneath the pleura visceralis. Identical alveolar clusters designated as (a) to (d).

054020_1_027905jbo3.jpg

The boundaries of the alveolar clusters were used to quantify alveolar area. The increase in alveolar area as CPAP levels rose was on the same order of magnitude when detected by either OCT or IVM. Alveolar expansion was found to be about 40% for IVM and 50% for OCT. When changes in alveolar area during inflation and deflation were quantified and plotted, both OCT and IVM images also revealed hysteresis behavior (Fig. 4 ). Plots of the alveolar area versus CPAP level for OCT and IVM were normalized to the area measured at 5cm H2O CPAP in the inspiratory phase. A typical p-v curve shape was obtained, where inflation has a constant slope between two inflection points and deflation stays on a moderately tilted plateau until it drops steeply, except at the 10-cm H2O pressure level. In all cases, different slopes were found for pressure increase and decrease. We found curve progression to correlate between OCT and IVM. Both curves showed hysteresis behavior during increasing and decreasing CPAP levels, indicating that the parenchyma has nonlinear (nonelastic) behavior that has a possible memory effect in the tissue. We recognized that the deflation curve does not return to baseline area CPAP of 5cm H2O . We also observed no recruitment or derecruitment activities in healthy lungs because the number of alveoli in the investigation area was constant at all CPAP levels.

Fig. 4

Alveolar area change for increasing and decreasing CPAP measured by (a) OCT and (b) IVM. All areas are normalized to the area measured at an applied CPAP of 5cm H2O in the inspiratory phase. Both methods show similar results, alveolar areas enlarge with increasing CPAP and contract by decreasing CPAP. The alveolar area change is non-linear and obeys hysteresis. ( n=4 lungs)

054020_1_027905jbo4.jpg

Because OCT is a 3-D imaging technique, as opposed to 2-D IVM imaging, the acquired data sets enable 3-D reconstruction of subpleural alveolar structures. Figure 5 shows a 3-D reconstruction of an individual alveolus at an applied CPAP of 5 and 25cm H2O . This technique reveals the alveolar expansion in all spatial directions. Additionally, we calculated 3-D triangular meshes to describe the alveolar structures (Fig. 5, overlay). Production of these meshes is an essential precondition to developing numerical models for simulating ventilation at the level of the alveoli.

Fig. 5

Volume uptake of a single alveolar structure applying CPAP of (a) 5cm H2O (left) and (b) 25cm H2O . The volume increase is evident for higher CPAP levels. 3-D OCT data sets enable calculation of 3-D triangular meshes of the alveolar geometry, as shown here.

054020_1_027905jbo5.jpg

3.2.

Comparison of Both OCT and IVM

To directly compare OCT and IVM techniques, we used a Bland Altman plot (Fig. 6 ).15 We performed linear regression and obtained a regression line with a slope of 0.17 (Fig. 6, black line). Thus, absolute alveolar area as measured via OCT is 17% (p<0.001) smaller than IVM.

Fig. 6

Comparison of both imaging techniques using a Bland-Altman plot. The regression line is black and has a slope of 0.17. Measurements taken from OCT are 17% smaller than from IVM. The detected bias is 615.5μm2 .

054020_1_027905jbo6.jpg

4.

Discussion

Here, we successfully combine OCT and IVM in one setup. This offers a number of advantages for analyzing alveolar mechanics, particularly a cross-sectional view at a freely selectable depth beneath the pleura and confirmation that circular structures observed via IVM are real alveoli. On the other hand, the isolated lung model does not reflect an in vivo situation because both pulmonary and thoracic components constituted the respiratory pressure volume curve. Furthermore, the fact that the deflation curve does not return to baseline area of 5cm H2O could imply a progressive alveolar volume uptake when applying CPAP for longer times. We used our setup to observe alveolar expansion in a cross-sectional area in response to rising pressure and revealed the nonelastic mechanics of this process. Our results agree with the alveolar expansion observed in a previous studies using confocal laser scanning microscopy to image alveolar tissue in isolated lungs16, 17and microscopic imaging of strained lung tissue blocs.18 The detected hysteresis between inflation and deflation matches well with the physiological quasi-respiratory pressure volume curves in vivo.19 However, there are differences compared to in vivo IVM analysis of healthy and injured lungs,20 suggesting alveoli do not change in volume. One important factor in our study is the deficiency of coverslips to avoid artifacts, such as deformation of alveolar geometry by suction or pressure. Our combined setup also ensures that movements in or out the focal plane cannot be misinterpreted as the recruitment of previously collapsed alveoli.

The isolated lung model and constant airway pressure procedure do have advantages over in vivo experiments, such as easy synchronization with imaging, lack of focusing problems, and control over physiological parameters without systemic effects. However, findings on alveolar mechanics in the stepwise inflated isolated lung cannot be generalized to in vivo dynamics. In principle, IVM and OCT methods, alone or in combination, are applicable in vivo using a transparent thorax window if the image synchronization and focusing problems can be solved. The best correlation between both techniques was observed using OCT en face images from 45μm beneath the pleura because the alveolar area measured at this depth is approximately the equatorial area of the alveoli, assuming the alveolar diameter is 100μm .

The difference between OCT- and IVM-derived measurements of air-filled areas might be due to several reasons. One reason might be the different transversal resolution that results in OCT overestimating alveolar wall thickness and, as a consequence, underestimating the air-filled regions. IVM images show the maximum expansion of the alveoli in the xy plane. This contrasts to OCT en face images, which represent the actual spatial plane and always result in smaller alveolar areas. The automatic segmentation generated for images also results in quantification differences of the alveolar clusters. The larger relative expansion measured by OCT is probably caused by overestimating alveolar wall thickness. This also underrates the air-filled areas of smaller alveolar clusters in comparison to larger clusters. In this case, the systematic mistake in measuring air-filled areas is more pronounced for alveoli during an applied CPAP of 5cm H2O than 25cm H2O . Our normalization of the air-filled area size to that at 5cm H2O leads to an apparently larger expansion measured in OCT images.

Here, we demonstrate that it is possible to carry out spatial segmentation of single alveolar clusters based on 3-D OCT data sets and use this data to 3-D reconstruct subpleural alveolar structures. It should be noted that speckle artifacts typical for OCT imaging hinder the exact segmentation of the air-filled spaces in the OCT data sets. OCT systems with higher resolution and faster acquisition times should improve OCT image quality and thus improve segmentation quality. Importantly, as we illustrate here, the simultaneous IVM and OCT setup has the potential to provide dynamic geometry data of alveolar structures as basis for numerical models of the lung at the level of the alveoli.

Acknowledgments

This project was supported by the German Research Foundation (DFG) “Protective Artificial Respiration” No. KO 1814/6-1.

References

1. 

K. F. Udobi, E. Childs, and K. Touijer, “Acute respiratory distress syndrome,” Am. Fam. Physician, 67 315 –322 (2003). 0002-838X Google Scholar

2. 

Acute Respiratory Distress Syndrome Network, “Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome,” N. Engl. J. Med., 342 1301 –1308 (2000). https://doi.org/10.1056/NEJM200005043421801 0028-4793 Google Scholar

3. 

J. A. Frank and M. A. Matthay, “Science review: mechanisms of ventilator-induced injury,” Crit. Care, 7 233 –241 (2003). https://doi.org/10.1186/cc1829 1364-8535 Google Scholar

4. 

D. Carney, J. DiRocco, and G. Nieman, “Dynamic alveolar mechanics and ventilator-induced lung injury,” Crit. Care Med., 33 S122 –S128 (2005). https://doi.org/10.1097/01.CCM.0000155928.95341.BC 0090-3493 Google Scholar

5. 

L. M. Malbouisson, J. C. Muller, J. M. Constantin, Q. Lu, L. Puybasset, and J. J. Rouby, “Computed tomography assessment of positive end-expiratory pressure-induced alveolar recruitment in patients with acute respiratory distress syndrome,” Am. J. Respir. Crit. Care Med., 163 1444 –1450 (2001). 1073-449X Google Scholar

6. 

E. van Beek, J. M. Wild, H. U. Kauczor, W. Schreiber, J. P. Mugler, and E. E. de Lange, “Functional MRI of the lung using hyperpolarized 3-helium gas,” J. Magn. Reson Imaging, 20 540 –554 (2004). https://doi.org/10.1002/jmri.20154 1053-1807 Google Scholar

7. 

B. D. Daly, G. E. Parks, C. H. Edmonds, C. W. Hibbs, and J. C. Norman, “Dynamic alveolar mechanics as studied by videomicroscopy,” Respir. Physiol., 24 217 –232 (1975). https://doi.org/10.1016/0034-5687(75)90115-2 0034-5687 Google Scholar

8. 

H. J. Schiller, U. G. McCann, D. E. Carney, L. A. Gatto, J. M. Steinberg, and G. F. Nieman, “Altered alveolar mechanics in the acutely injured lung,” Crit. Care Med., 29 1049 –1055 (2001). https://doi.org/10.1097/00003246-200105000-00036 0090-3493 Google Scholar

9. 

R. D. Hubmayr, “Perspective on lung injury and recruitment: a skeptical look at the opening and collapse story,” Am. J. Respir. Crit. Care Med., 165 1647 –1653 (2002). https://doi.org/10.1164/rccm.2001080-01CP 1073-449X Google Scholar

10. 

A. Popp, M. Wendel, L. Knels, T. Koch, and E. Koch, “Imaging of the three-dimensional alveolar structure and the alveolar mechanics of a ventilated and perfused isolated rabbit lung with Fourier transform optical coherence tomography,” J. Biomed. Opt., 11 (1), 14015 (2006). https://doi.org/10.1117/1.2162158 1083-3668 Google Scholar

11. 

A. F. Fercher, W. Drexler, C. K. Hitzenberger, and T. Lasser, “Optical coherence tomography—principles and applications,” Rep. Prog. Phys., 66 239 –303 (2003). https://doi.org/10.1088/0034-4885/66/2/204 0034-4885 Google Scholar

12. 

D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science, 254 1178 –1181 (1991). https://doi.org/10.1126/science.1957169 0036-8075 Google Scholar

13. 

Guide for the Care and Use of Laboratory Animals, 7th ed.Institute of Laboratory Animal Resources, National Academy Press, Washington, DC (1996). Google Scholar

14. 

S. Meissner, M. Mertens, A. Krüger, A. Tabuchi, W. Kuebler, and E. Koch, “In vivo three-dimensional Fourier domain optical coherence tomography of subpleural alveoli combined with intra vital microscopy in the mouse model,” Biomedical Optics, Optical Society of America, Paper No. BWF4, (2008). Google Scholar

15. 

J. M. Bland and D. G. Altman, “Measuring agreement in method comparison studies,” Stat. Methods Med. Res., 8 135 –160 (1999). https://doi.org/10.1191/096228099673819272 0962-2802 Google Scholar

16. 

C. E. Perlman and J. Bhattacharya, “Alveolar expansion imaged by optical sectioning microscopy,” J. Appl. Physiol., 103 1037 –1044 (2007). https://doi.org/10.1152/japplphysiol.00160.2007 8750-7587 Google Scholar

17. 

E. Namati, J. Thiesse, J. de Ryk, and G. McLennan, “Alveolar dynamics during respiration: are the pores of Kohn a pathway to recruitment?,” Am. J. Respir. Cell Mol. Biol., 38 (5), 572 –578 (2008). https://doi.org/10.1165/rcmb.2007-0120OC 1044-1549 Google Scholar

18. 

K. K. Brewer, H. Sakai, A. M. Alencar, A. Majumdar, S. P. Arold, K. R. Lutchen, E. P. Ingenito, and B. Suki, “Lung and alveolar wall elastic and hysteretic behavior in rats: effects of in vivo elastase treatment,” J. Appl. Physiol., 95 1926 –1936 (2003). 8750-7587 Google Scholar

19. 

B. Jonson and C. Svantesson, “Elastic pressure-volume curves: what information do they convey?,” Thorax, 54 82 –87 (1999). 0040-6376 Google Scholar

20. 

J. D. DiRocco, D. E. Carney, and G. F. Nieman, “Correlation between alveolar recruitment/derecruitment and inflection points on the pressure-volume curve,” Intensive Care Med., 33 1204 –1211 (2007). https://doi.org/10.1007/s00134-007-0629-8 0342-4642 Google Scholar
©(2009) Society of Photo-Optical Instrumentation Engineers (SPIE)
Sven Meissner, Lilla Knels, Alexander Krueger, Thea Koch, and Edmund Koch "Simultaneous three-dimensional optical coherence tomography and intravital microscopy for imaging subpleural pulmonary alveoli in isolated rabbit lungs," Journal of Biomedical Optics 14(5), 054020 (1 September 2009). https://doi.org/10.1117/1.3247149
Published: 1 September 2009
JOURNAL ARTICLE
6 PAGES


SHARE
Advertisement
Advertisement
Back to Top