Comprehensive volumetric optical microscopy in vivo

Abstract

Comprehensive volumetric microscopy of epithelial, mucosal and endothelial tissues in living human patients would have a profound impact in medicine by enabling diagnostic imaging at the cellular level over large surface areas. Considering the vast area of these tissues with respect to the desired sampling interval, achieving this goal requires rapid sampling. Although noninvasive diagnostic technologies are preferred, many applications could be served by minimally invasive instruments capable of accessing remote locations within the body. We have developed a fiber-optic imaging technique termed optical frequency-domain imaging (OFDI) that satisfies these requirements by rapidly acquiring high-resolution, cross-sectional images through flexible, narrow-diameter catheters. Using a prototype system, we show comprehensive microscopy of esophageal mucosa and of coronary arteries in vivo. Our pilot study results suggest that this technology may be a useful clinical tool for comprehensive diagnostic imaging for epithelial disease and for evaluating coronary pathology and iatrogenic effects.

Conflict of interest statement

COMPETING INTERESTS STATEMENT

The authors declare that they have competing financial interests (see the Nature Medicine website for details).

Figures

Figure 1

Principles of comprehensive optical frequency-domain imaging (OFDI). Minimally invasive catheters or endoscopes provide for access of the optical fiber to the organ or system of interest. An optical beam is focused into the tissue, and the echo-time delay and amplitude of light reflected from the tissue microstructure at different depths are determined by detecting spectrally resolved interference between the tissue sample and a reference, as the source laser wavelength is rapidly varied from 1,264 nm to 1,376 nm. A Fourier transform of this signal forms image data along the axial line (A-line), which is determined by the optical beam emitted from the probe. A-lines are continuously acquired as the probe is actuated to provide spatial scanning of the beam in two directions that are orthogonal to the axial line (rotational and pull-back motion by a rotary junction). The resulting three-dimensional data sets can be rendered and viewed in arbitrary orientations for gross screening, and individual high-resolution cross-sections can be displayed at specific locations of interest.

Figure 2

Comprehensive microscopy of a porcine esophagus in vivo. (a) The 14-GB volumetric data set was rendered and downsampled for presentation in arbitrary orientations and perspectives. The vascular network within the submucosa is readily apparent without image enhancement or exogenous contrast agents. Cross-sectional images can be located on the volume image for higher-resolution viewing. (b) Longitudinal cross-section through esophageal wall at location denoted in a (inverted with epithelium at the top; dimensions: 45 mm horizontal, 2.6 mm vertical). In the raw data, we observed a periodic vertical offset corresponding to the motion of the beating heart. We used a simple surface-aligning algorithm to reduce this artifact, but a residual vertical banding can still be observed with a period of 300 μm corresponding to a heart rate of 90 beats/min. The longitudinal spacing between adjacent A-lines was 32 μm. (c) Unwrapped transverse section (cylindrical coordinates r and θ are mapped to vertical and horizontal) at location denoted in a (dimensions: 57 mm horizontal, 2.6 mm vertical). Both sections show imaging through the entire esophageal wall. (d) Magnification of boxed area in c, showing the squamous epithelium (e), lamina propria (lp), muscularis mucosa (mm), submucosa (sm) and muscularis propria (mp). (e) Representative histology section (H&E stain) obtained from the anatomical region corresponding to that depicted in d. Scale bars, 500 μm.

Figure 3

Comprehensive microscopy of a porcine coronary artery in vivo. (a) Three-dimensional cut-away rendering of the volumetric data set acquired with an intravascular catheter in the right coronary artery of a living swine. The volume comprises 400 circular (r-θ) sections at a spacing of 50 μm acquired in 3.7 s during the injection of saline at a rate of 3 ml/s. The ostia of three side branches can be clearly seen. (b) A single circular cross-section at the location denoted in a. The opaque guidewire and the sector that it obscures are denoted by the asterisk. Orange arrow indicates image of the transparent catheter sheath. Red arrows designate intima (i), media (m) and adventitia (a). (c) A longitudinal section rendered from the volumetric data, using a spatial low-pass filtering of the surface and subsequent surface alignment to reduce spatial distortion arising from pulsation. (d) Magnification of boxed area in c, showing trilaminar structure of vessel wall in longitudinal perspective. Image data was acquired at a rate of 108 frames/s. Scale bars, 1 mm.

Figure 4

Volumetric imaging of a stented porcine coronary artery in vivo. (a) Three-dimensional cut-away rendering of the volumetric data set acquired with an intravascular catheter in the circumflex coronary artery of a living swine after balloon angioplasty and stent implantation. Blue, stent; red, intima and media; gray, adventia and surrounding tissue. The volume comprises 500 circular (r-θ) sections at a spacing of 50 μm acquired in 6 s during the injection of saline at 3 ml/s. (b–f) Individual OFDI cross-sectional images at the five locations marked in a. The metal-based stent produces strongly reflected signals and leaves radial shadow patterns in c and d. The dissected intima and media layers are shown (e; orange asterisk). Tissue prolapse between the stent struts is visualized in c and d (yellow asterisk). Scale bars, 1 mm.

Figure 5

Cross-sectional images of dissected coronary artery and histology comparison. (a,b) Circular cross-sectional OFDI images acquired 15 min after stent deployment at the stent border (a), and 1.0 mm distal to the stent (b). Disruption of the intima and media is evident, as well as thrombus (arrows). (c,d) Histology sections (elastin stain) obtained at locations corresponding to those depicted in a and b, respectively, showing similar features of disruption and thrombus. Scale bars, 500 μm.