A tomography system for imaging deep tissues
The number of people with major eye disease, such as glaucoma, age-related macular degeneration (ARMD), and diabetic retinopathy, is set to increase and to become a major public health problem.1 Several methods are used to diagnose and monitor the progression of such diseases. Optical coherence tomography (OCT) is a noninvasive method for obtaining 3D images of biological tissues. Such images contain information about the microstructures and microvasculature (small vessels and capillaries of the circulatory system) composition of the tissue that correlates with diseases.
However, until recently, the high light attenuation of biological tissues prevented OCT systems from imaging deep tissue structures. Images of the choroid have had application in diseases such as uveal inflammation2 and ARMD.3 But the reported maximum imaging speed of these systems has been 47kHz line scan rate, a very slow value that is sensitive to motion artifacts and limits the maximum value of blood flow measured.
We have developed a spectral domain OCT system that enables the acquisition of the microstructural and microvascular composition of both the retina and choroid using an ultrahigh sensitive optical microangiography (OMAG) method, a technique that enables visualization of small blood vessels and capillaries.4–6 The system uses a 1050nm wavelength that has higher penetration depth inside tissues compared to the 800nm wavelength that is commonly used. The system also uses a camera that has a 92kHz line scan rate, which is the fastest spectral domain system reported for this wavelength. This value reduces motion artifacts and increases the measured blood flow in comparison with 47kHz systems.
We compared images obtained from the retina of a volunteer using the new system, which has a central wavelength of 1050nm and a 92kHz line scan rate, with a system that has a light source with a central wavelength of 850nm and an 80kHz line scan rate. The data was processed using OMAG software. Both systems had ∼140 frames/s, and each frame had 512 (depth) by 360 (lateral) pixels to cover ∼3mm.
Figure 1 presents the microstructural and microvascular images. The 1050nm system details more structural information at greater depth compared to the 800nm system—white arrows in Figure 1(A) and (B)—including the retina, the choroid, and the choroid interface with the sclera. In the retinal layer, both systems show abundant capillary flows: red arrows in Figure 1(C) and (D). However, the signal from the choroidal layer was strongly attenuated for the 800nm light source. The 1050nm system provided a strong blood flow signal even from deep within the choroidal layer, demonstrating a significant advantage in imaging choroidal blood flows.
Due to the depth-resolved nature of our system, we segmented the flow images obtained by our 1050nm system into six layers: see Figure 1(D). The retinal layer was divided into the radial peripapillary capillaries, the inner capillaries that lay in the ganglion cell layer, and the outer capillaries from the inner to the outer plexiform layers. The choroidal layer was divided into three sublayers that represent the vessels just below the choriocapillary layer, medial arterioles and venules, and outer arteries and veins, respectively. Figure 2 shows the maximum amplitude projection of the vessels at each layer. The retinal capillary networks visualized by the 1050nm system are similar to those previously reported using the 800nm system.6 However, due to the higher penetration depth achieved with the 1050nm system, clearer visualization of the choroidal vessels is obtained. These vessels may supply nutrition to the choroidal capillary bed.
Our system obtains images with higher acquisition speed than conventional OCT systems, and our OMAG software can extract microvascular information from deep tissue layers such as the choroid. These images are anticipated to be used for tissue diagnostics. We also plan to work on increasing the acquisition speed of the system and to apply it for other diseases, such as cancer.
Roberto Reif received his PhD in biomedical engineering from Boston University in 2008. He is now a postdoctoral research fellow focusing on biomedical optics.
Lin An is a PhD candidate in the Department of Bioengineering. He is focusing on using optical microangiography to image ocular blood vessels.
Ruikang K. Wang obtained his PhD in optical information processing from Glasgow University, UK. He is currently a professor in bioengineering and leads the Biophotonics and Imaging Laboratory.