Ex Vivo Optical Coherence Tomography Analysis of Resected Human Bladder with a Forward-Looking Microelectromechanical Systems Mirror-Based Catheter

Abstract

A technique that enables real-time diagnosis of bladder cancer is needed. Optical coherence tomography (OCT) is a promising technique, but a forward-looking OCT catheter is necessary for OCT to enable bladder cancer diagnosis. This study aims to describe the design of a novel forward-looking microelectromechanical systems (MEMS)-based OCT catheter, assess the performance characteristics, and evaluate its ability to identify histopathological characteristics of bladder specimens. A description of the OCT catheter and systems used is provided. Performance characteristics were measured with a beam profiler and microscopy slide (mirror for dispersion and thickness for lateral calibration). Ex vivo measurements were performed on resected bladder tissue from patients undergoing a radical cystectomy. A forward-looking OCT probe with an outer diameter of 2.52 mm and a rigid length of 17 mm was designed and evaluated. The focus position was measured as 10.9 mm from the MEMS mirror, with a Rayleigh length of 2.55 mm. Several histopathological features could be correlated to OCT features of the ex vivo measurements. In conclusion, a forward-looking OCT probe that can be inserted in the working channel of a rigid cystoscope was designed and evaluated. Performance characteristics were overall in line with simulated expectations.

Keywords: bladder cancer; microelectromechanical systems; optical coherence tomography; optical imaging.

Figure 1

Schematic and real-life OCT system set-up. (A) Schematic overview of the different components within the OCT system. The HSL swept-source laser is fiber-coupled to the IVS-1000 interferometer. A portion of the light is directed towards the Wavelength Division Multiplexer (WDM), where the guide light (Thorlabs LP635) is injected into the fiber and directed towards the MEMS probe. The reflected light from the sample passes through the splitter and interferes with light from the reference arm of the set-up. (B) Real-life OCT system set-up consisting of a: motorized X-Y-Z platform; b: HSL 1 high-speed swept-source laser; c: IVS-1000 interferometer and data acquisition system; d: computer, and e: tip of the OCT catheter.

Figure 2

Schematic representation of the catheter design. (A) Schematic illustration of the components integrated into the rigid part of the catheter. (B) Schematic illustration of the light propagation from the prism to the MEMS mirror and out of the probe. The different light bundles illustrate the different angles at which the light is directed out of the probe.

Figure 3

False color plot of the optimization of the scan angle and mirror-to-0-delay distance. In (A), the results (in mm) of the differences (compared to 1 mm) between all microscopy slide combinations of different angles and distances to 0-delay are plotted. The 0-delay point is marked with an orange bar in (B,C). The bar next to (A) displays the definition of the colors in the plot ranging from a summed difference of zero to 0.2 mm. In (B), the B-scans of the microscopy slide at different distances are merged into one B-scan, and the outer sides of the microscopy slides are marked. The microscopy slide contains bevels on both sides, so the outer side of the bevel is marked. The resulting OCT image with the optimal angle and distance to 0-delay is shown in (C).

Figure 4

Illustrative depiction of the interrelation between individual B-scans. The independent B-scans were captured in the direction of Y with a noted spacing (Y). With the motorized X-Y-Z platform, a new row of B-scans started next to the first B-scan with a known spacing (X) in between the B-scans. The B-scans then were stitched together in the direction of X to form ‘slides’.

Figure 5

Measured spot size relative to distance from 0-delay.

Figure 6

Dimensions of OCT B-scan. The dimensions in the image are not to scale. The distance from the MEMS mirror to zero delay is 7.5 mm, and the axial range of the optical coherence tomography (OCT) image is 8.2 mm. The focus point is at 10.9 mm from the microelectromechanical systems (MEMS) mirror. The Rayleigh length is 2.55 mm. With a MEMS frequency of 562 Hz, the angle is 7.8 degrees.

Figure 7

OCT images of five patients with their corresponding hematoxylin and eosin-stained histopathology slides. (A,B) Corresponding optical coherence tomography (OCT) and histopathology showing a ureter crossing the sample at the red arrow; (C,D) corresponding OCT and histopathology showing corresponding contour of the sample with a large dent at the red arrow; (E,F) corresponding OCT and histopathology showing several conformities with red arrows. From left to right: a Von Brunn nest, dent in the contour, higher-intensity OCT image at the tumor location in histopathology; the rightmost arrow also shows a higher-intensity OCT image at the tumor location in histopathology; (G,H) corresponding OCT and histopathology, showing a round tumor with the red arrow; (I,J) corresponding OCT and histopathology, showing with the left arrow a tumor necrotic cavity, appearing as a black area with high-scattering structure on top in the OCT image, and at the right deformed contour of the sample; (K,L) corresponding OCT and histopathology showing the chemotherapy effect with the red arrow, which is seen as higher intensity on the OCT image. The scalebar is applicable to the histopathology images in all directions and to the OCT images in the vertical direction.