Research

Clinical Confocal Microscopy

Overview

It is well established that confocal microscopy provides higher-resolution images with better rejection of out-of-focus information than conventional light microscopy. The optical sectioning ability of confocal microscopy allows images to be obtained from different depths within a thick tissue specimen, thereby eliminating the need for processing and sectioning. Because of its non-invasive optical sectioning capability, confocal microscopy is ideally suited to the study of corneal tissue in intact living animals and patients, and can generate quantitative 3-D data over time. Using a technique termed confocal microscopy through-focusing (CMTF, see details below), a stack of 2-D images spanning the full thickness of the cornea can be generated, and 3-D images of corneal structure can be reconstructed and followed temporally. In addition to information regarding the reflectivity, morphology and organization of cells obtained from the 2-D and 3-D confocal images, the thickness and reflectivity of corneal sublayers (e.g., epithelium, stroma) can also be obtained from the CMTF data.

In the clinic, confocal microscopy allows assessment of the cellular response of the cornea to infection, injury, and surgical procedures. Confocal microscopy following photorefractive keratectomy (PRK) has demonstrated that the development of clinical corneal haze is correlated with the activation of corneal keratocytes and transformation to a fibroblast or myofibroblast phenotype. These activated cells are more reflective than quiescent corneal keratocytes, and synthesize extracellular matrix components that also reduce corneal transparency. Keratocyte activation has also been identified by confocal microscopy following microkeratome-assisted LASIK.

Confocal microscopy has also been used to assess the corneal response to LASIK with flap creation using IntraLase. Previous studies have shown that the IntraLase® provides more consistent flap thickness with fewer complications than traditional microkeratomes, and results in better visual outcome in most patients. However, one important concern regarding IntraLase® is that the procedure induces more damage than a microkeratome and can therefore stimulate a more pronounced wound healing response. In recent studies, we have used confocal microscopy to compare the effects of different IntraLase® raster energies, pulse frequencies, and post-operative steroid treatments on the corneal response to intraLASIK, (Refs 1-3). Our data suggests that LASIK with Intralase® provides more reproducible flap thickness and fewer interface particles than previously observed using mechanical microkeratomes. However, Intralase® can also induce more significant keratocyte activation, which may underlie clinical observations of haze in some patients. Our results indicate that activation may be reduced by using lower raster energies and an extended steroid treatment regimen.

Technical Description


Current Microscope Design (From References 4-5)

A simplified schematic of the optical pathway of the current TSCM system used in our department shown in Fig. 1.

Figure 1

The TSCM uses a 0.25% transmittance disk (20 micron pinholes), with a patented hole pattern which has been optimized to eliminate scan lines often associated with TSCM imaging. A Dage VE1000 low light level video camera is used for image detection. The current TSCM system in our laboratory uses a specially designed surface contact objective (24X, 0.6NA, 0-1.5 mm variable working distance). Unlike other objective designs, the position of the focal plane relative to the objective tip is varied by moving the lenses within the objective casing (Fig. 2).

Figure 2

Objective lens movement is controlled using an Oriel 18011 Encoder Mike Controller with a digital readout interfaced to a personal computer via a serial port. A program has been written to convert the encoder mike reading to the corresponding z-axis position (depth) of the focal plane in microns using a third order polynomial. This number is continuously written into the user bit register of a Time Code Generator, which displays this number onto a video monitor during recording and/or playback.

In order to determine the z-axis resolution of the TSCM system, the axial response was determined by focusing through a perfect planar reflector (front surface mirror), and measuring the reflected intensity curve (Fig. 3).

Figure 3

This is a standard technique for characterizing confocal systems. Measurements were made at three different focal depths within the range normally used to study the cornea. All three curves show considerable spherical aberration, but there is very little change in the shape of the curves as a function of depth. The measured full width at half maximum (FWHM), an indicator of the optical slice thickness, was 9.02 at a depth of 400 microns, and increased to 9.27 microns at a depth of 800 microns.

Confocal microscopy through-focusing (CMTF) (From Reference 6)

CMTF is based on the observation that different corneal sub-layers produce a different amount of backscattering when imaged using confocal microscopy; thus an intensity profile can provide information about the depth and thickness of corneal cell layers. CMTF of the cornea not only provides a 3-D display of corneal structure, but more importantly provides quantitative data regarding the depth and thickness of different tissue layers and structures. A CMTF curve is obtained by scanning through the cornea from the epithelium to endothelium at a constant lens speed. CMTF images are detected using the Dage camera set at a constant gain, KV and black level, and recorded onto Super VHS videotape. Since the video images are captured and digitized at 30 frames/second, during a continuous scan, consecutive images are separated in the z-axis by 1-2 μ, (depending on lens speed). In our current system, CMTF scans are digitized in real-time into system memory, and the CMTF intensity data is obtained by calculating the average pixel intensity in a central region of each image.

In order to identify the corneal sub-layers (e.g. epithelium, basal lamina, and endothelium) corresponding to the CMTF intensity peaks, the image intensity curve is displayed below an image viewing window and corresponding images are displayed as the cursor is moved along the curve. In this way, representative images from each peak can be selected and saved interactively. The software also allows the user to mark and record the positions of the peaks providing a record of the z-axis depth of any structure.

Using this system, accurate and reproducible measurements of corneal, epithelial, and stromal thickness can be obtained in both animals and humans. Two major peaks corresponding to the superficial epithelium anteriorly (Fig 4A) and the corneal endothelium posteriorly (Fig 4E) are present in normal CMTF curves. Corneal thickness is measured by calculating the z-axis distance between these CMTF peaks. CMTF intensity profiles also show smaller peaks corresponding to the basal corneal epithelial nerve plexus (Fig 4B) and the anterior layer of corneal keratocytes (Fig 4C). The distance between the basal epithelial nerve plexus and the superficial epithelium can be used to measure epithelial thickness. CMTF images can also be used for 3-D reconstruction of the cornea (Fig. 4F).

Figure 4A-G

Online CMTF Software (From References 7-8)

The major functional parts of our system include a TSCM with a 24X, 0.6NA applanating objective with a variable working distance, a lens motion controller (Oriel 18011), a low light level video camera (Dage VE1000) and a high performance Personal Computer running Windows NT with a PCI image acquisition board (Data Translation DT3152). The layout of the hardware interfaces is shown in Figure 5.

To allow on-line confocal imaging, software needed to perform some or all of the following tasks simultaneously: 1) interact with the Oriel objective lens motion controller (via serial port 1) in order to control focusing and read back the current focal position, 2) output the focal position to the time code generator (via serial port 2), 3) acquire images via the DT3152 board, 4) perform image analysis, and 5) display the results. To meet these design constraints, we developed a user-friendly Windows program using Microsoft Visual C++ (version 6.0). The program features multitasking, overlapped I/O and the creation of special classes to simplify device interface.

Figure 5



For on-line CMTF acquisition, the program first positions the focal plane to the zero position (at the tip of the objective). A signal is then sent to the Oriel controller through serial port 1 to start lens movement. Image acquisition is then immediately started by sending an acquire command to the DT3152. We are currently using a "sychronous acquire" mode which insures that no frames are missed during the acquisition. After the user-selected number of frames has been digitized (typically 400 for a normal CMTF), the acquire process is terminated and a signal is sent to the Oriel to halt lens movement. Immediately after an on-line CMTF scan ceases, the intensity curve is generated instantly and displayed in a "curve pane" on the right half of the PC screen (Fig 6).

Figure 6

The user can then mark the positions in the curve corresponding to the layers of interest in order to obtain thickness information on-line. Intensity curve data, along with the CMTF settings and measurement information, can be saved to a text file. This file can be exported to other software such as Microsoft Excel when necessary.

Immediately following the completion of an online CMTF scan, the intensity curve, the side views, and the 3-D reconstructed volume projection with a default orientation and size are displayed. Due to the high computing power of current PCs and the programming techniques used, updating these views occurs in real-time. Thus, the user can interactively pick a region of interest by dragging a the corners of a highlighted "region of interest" box in the x-y plane, while the side views and the 3-D projection are continuously updated. This gives the user great power and flexibility for visualizing and localizing a structure inside the cornea interactively. The side views of the cornea have been particularly useful, because they provide a view similar to that of an ophthalmic slit-lamp (although at much higher resolution and contrast) which is familiar to most clinicians.

The 3-D reconstructions have been used for identifying and localizing the source of corneal haze after PRK (Fig. 7), and the flap interface following LASIK.

Figure 7

The previous CMTF program produced a numerical area value after the user typed in the start and end points of the haze peak using the keyboard. In this new software environment, area is calculated and displayed graphically by clicking on the peak start and end points with the cursor. Together with the interactive 3-D reconstruction and display, obtaining quantitative data on the amount and location of corneal haze becomes convenient and rapid.

Using the standard frame rate (30 frames per second) and a typical lens speed of 160 μm per second, the average CMTF focal plane speed is about 64 μm per second. Thus, a 400-image sequence will scan 853 μm, which more than covers the average thickness of human cornea. Images are sub-sampled (320 x 240 pixels), so that a stack of 400 images uses 30 MB of RAM. The image stack can be saved to disk, and loaded back into the TSCM2.0 software environment whenever it is desirable to re-review, display, or further analyze the scan. We have even developed a version of the program that runs without the digitizing board, and can be used on a laptop computer. Of course, by disabling the lens control commands and accepting input from videotape, the program can also perform off-line CMTF acquisition and data analysis if necessary.

Overall this program provides convenient, online, 3-D image acquisition and analysis in a user-friendly environment.

References

  1. Petroll WM, Goldberg D, Lindsey SS, Kelley PS, Cavanagh HD, Bowman RW, Parmar DN, Verity SM, McCulley JP. Confocal assessment of the Corneal Response to intracorneal lens insertion and LASIK with flap creation using IntraLase®. J Cataract Refract Surg 32:1119-1128, 2006.
  2. Hu MY, McCulley JP, Cavanagh HD, Bowman RW, Verity SM, Mootha VV, Petroll WM. Comparison of the corneal response to LASIK with flap creation using the IntraLase FS15 and FS30 lasers: Clinical and confocal microscopy findings. J Cataract Refract Surg 33:673-681, 2007.
  3. Petroll WM, Bowman RW, Cavanagh HD, Verity SM, Mootha VV, McCulley JP. Assessment of Keratocyte Activation Following LASIK with Flap Creation using the IntraLase® FS60 Laser. J Refract Surg (In Press).
  4. Cavanagh HD, Petroll WM, Alizadeh H, He Y-G, McCulley JP, and Jester JV. Clinical and diagnostic use of in vivo confocal microscopy in patients with corneal disease. Ophthalmology 100(10):1444-1454, 1993.
  5. Petroll WM, Jester JV, Cavanagh HD. 3-Dimensional imaging of corneal cells using in vivo confocal microscopy. J Microsc 170:213-219, 1993.
  6. Li H, Petroll WM, Moller-Pederson T, Maurer JK, Cavanagh HD, Jester JV. Epithelial and corneal thickness measurements by in vivo confocal microscopy through focusing (CMTF). Curr Eye Res 16:214-221, 1997.
  7. Li J, Jester JV, Cavanagh HD, Black TD, Petroll WM. On-line 3-dimensional confocal imaging in vivo. Invest Ophthalmol Vis Sci 41:2954-2953, 2000.
  8. Moller-Pederson T, Vogel MD, Li H, Petroll WM, Cavanagh HD, Jester JV. Quantification of stromal thinning, epithelial thickness, and corneal haze following photorefractive keratectomy using in vivo confocal microscopy. Ophthalmology 104:360-368, 1997.