What Is Scanning Laser Ophthalmoscopy?

A variety of eye diseases can be accompanied by thinning of the retinal nerve fiber layer (RNFL), some of which occur before obvious clinical symptoms and signs appear. However, in most cases, the examination and photography of RNFL is a difficult technique. These Technology is not only subjective, non-quantitative, unstable, time-consuming, and dependent on the skill level of the operator, it also has limited sensitivity and specificity. Polarized laser scanner (SLP) is a new technology that appears after retinal thickness analyzer (RTA), confocal laser scanner (HRT), and coherent tomography (OCT). It uses the principle of birefringence produced by retinal ganglion cell axon to realize the measurement of RNFL thickness. It can provide automatic, objective and quantifiable RNFL measurement methods with high repeatability, test results and clinical evaluation of optic disc structure. And visual function check.

A variety of eye diseases can be accompanied by thinning of the retinal nerve fiber layer (RNFL), some of which occur before obvious clinical symptoms and signs appear. However, in most cases, the examination and photography of RNFL is a difficult technique. These Technology is not only subjective, non-quantitative, unstable, time-consuming, and dependent on the skill level of the operator, it also has limited sensitivity and specificity. Polarized laser scanner (SLP) is a new technology that appears after retinal thickness analyzer (RTA), confocal laser scanner (HRT), and coherent tomography (OCT). It uses the principle of birefringence produced by retinal ganglion cell axon to realize the measurement of RNFL thickness. It can provide automatic, objective and quantifiable RNFL measurement methods with high repeatability, test results and clinical evaluation of optic disc structure. And visual function check.

Chinese name

Polarized laser scanner

Foreign name

Polarized laserscanners

A variety of eye diseases can be accompanied by a thinning of the retinal nerve fiber layer (RNFL), some of which occurred before obvious clinical symptoms and signs appeared more than 20 years ago. Quigley et al. 40% of optic axons disappeared. In most cases, the inspection and shooting of RNFL is a difficult technique. These techniques are not only subjective, non-quantitative, and unstable, but also time-consuming. Depending on the skill of the operator, their sensitivity and specificity are also limited. . Scanning laser polarimetry (SLP) is a new technology that appears after retinal thickness analyzer (RTA), confocal laser scanner (HRT), and coherent tomography (OCT). It uses the principle of birefringence produced by retinal ganglion cell axon to realize the measurement of RNFL thickness. It can provide automatic, objective and quantifiable RNFL measurement methods with high repeatability. The test results and clinical evaluation of optic disc structure and Visual function check is consistent ^[1] .

Polarized Laser Scanner Technology Principles

1 . Birefringence

The parallel arrangement of axonal microtubules in retinal ganglion cells makes RNFL a birefringent medium. SLP uses confocal laser scanning ophthalmoscope and integrated polarimeter to quantitatively evaluate the retardation of RNFL by the polarized light formed by the birefringence of auxiliary light passing through the RNFL twice. In birefringence, the slow optical axis and the RNFL beam are aligned in the same direction, and the retardation is proportional to its thickness. The SLP system uses a near-infrared diode-polarized laser (wavelength of 780 am) as the light source. Through the refractive matrix of the eye, it focuses on a certain point of the retina around the optic disc and passes through the RNFL with birefringence. The change causes a delay in polarized light, and the polarized light reflection is detected and analyzed by a polarization modulator, and stored in an electronic computer. Repeat the measurement by moving to a nearby retinal site with a scanning device. Each final delayed image consists of 256 × 256 pixels, each pixel corresponds to the delay value of its corresponding part, the scan threshold is 200 × 200, and the acquisition time is 0.7 s. Because the diameter of the microtubules contained in the parallel RNFL axons is smaller than the wavelength of polarized light, the rate of advancement of some of the two parts of the polarized light can be changed to generate a phase delay of the polarized light. It is proportional, so the thickness of RNFL measured by SLP is a relative thickness, but it has a high correlation with the thickness measured by tissue section method. The correlation coefficient r = 0.7, P <0.001, which is a retardation value of polarized laser It is equivalent to 7.4 m RNFL thickness ^[1] .

2 Calculation and compensation of artificial corneal birefringence

All birefringent structures will change the polarization of the beam. The total retardation depends on the anterior segment (cornea, lens) and RNFL thickness. The accuracy of measuring RNFL thickness using SLP depends on extracting the amount of delay caused by RNFL from the total amount of delay. In order to eliminate the interference of corneal birefringence on the thickness measurement of RNFL, a polarized light scanner incorporates a variable corneal compensator. It uses the macula as a polarimeter in the eye to measure and exclude special corneal polarization axis (CPA) and corneal polarization measure (CPM) of the eye. The macular Henle fiber layer can produce radial birefringence because of the uniform arrangement of the photoreceptor axis. It is thought that the macular polarization image without corneal birefringence compensation has a unique bow tie shape and bimodal shape. A bow tie pattern is created due to the interaction of the corneal birefringence and the Henle fiber layer of the macula, which is used to determine the corneal birefringence. When the short axis of the cornea and the short axis of the Henle fibrous layer are parallel, the bright part of the "bow tie" is generated, so that the total amount of retardation is obtained. When the short axis of the cornea is perpendicular to the short axis of the Henle fibrous layer, the dark part of the "bow tie" is generated, which cancels out the amount of delay. In this way the birefringent part of the front section can be read directly from the direction of the bright part of the "bow tie". SLP can determine whether corneal compensation is appropriate by examining macular delayed images. However, macular lesions can interfere with the integrity of the Henle fibrous layer and produce an indefinite "bow tie" pattern, making the "bow tie" rule no longer applicable to corneal compensation ^[1] .

Polarized Laser Scanner Image Data Analysis

The polarized laser scanner GDx VCC uses a fixed scanning annular area with a diameter of 3.2 mm and a center disc. It is a high-quality, precision-focused scan that maintains good focus when the eyeball moves slightly. The software automatically gives a quantitative score of 1 to 10 according to the positioning, refraction, and eye adjustments. The acceptable image quality is more than 8 points ^[1] .

Images of RNFLs of normal thickness appear bright yellow and red (thicker) above and below, and green and blue (thinner) on the nasal and temporal sides. The temporal, superior, nasal, inferior, and temporal (TSNIT) plots show the thickness of the RNFL of the calculated loop, indicating the range of normal values. Basic parameters include the ellipse average (TSNIT average): the average of all pixels in the elliptical measurement ring; the upper average (superi-or average): the average of all pixels in the upper 1200 area within the ellipse measurement ring; ): 120 inside and below the ellipse measurement ring. The average value of all pixels in the area; TSNIT standard deviation: The probability of the RNFL thickness value of a part appearing in the normal person database, usually in 4 × 4 pixels; Inter-Eye symmetry; Nerve fiber indicator (NFI). The color difference is used to show the statistical difference from normal eye data ^[1] .

There is currently no generally accepted standard for judging RNFL images or delay parameters, and no accepted definition of anomalous scans has been established. If a GDx VCC scan has a TSNIT average, upper average, lower average, TSNIT standard deviation, binocular symmetry, and nerve fiber index (NFI) at P <0.01, it is usually considered abnormal. GDx VCC scans are now considered to be normal and abnormal when P <0.05 for TSNIT average, upper average, lower average, TSNIT standard deviation, binocular symmetry, and nerve fiber index (NFI). It has been suggested that> 47 (P <0.01) or> 30 (P <0.05) be used as the upper limit of NFI ^[1] .

SLP found that the possibility of RNFL changes includes: changes in the absolute value of the delay value, RNFL four-quadrant thickness changes, RNFL thickness profile changes, and color coded RNFL thickness changes relative to the base value. However, like OCT, the possibility of change lacks statistical significance, which limits the ability to distinguish between changes in measurement variability, and this algorithm has not been confirmed ^[1] .

The development of polarized laser scanners

Since the technology was introduced ten years ago, SLP has undergone several generations of hardware and software updates ^[1] .

( A) hardware update

1. Nerve fiber analyzer

The first device (NFA I) was launched in 1992. It was a separate device and was later replaced by a dual device (NFA II). The high repeatability of applying NFA has been reported, and the measurement of delay has been shown to correlate well with visual function J ^[1] .

2. GDx nerve fiber analyzer

The GDx Nerve Fiber Analyzer (GDx NFA) was launched in 1996. The standardized database includes 400 eyes of different ages and races and a vascular exclusion algorithm to enhance the repeatability of the examination. The repeatability of detection in aphakia, intraocular lens, and glaucoma has been reported ^[1] .

( Two) software updates

1. Fixed corneal compensation mode

The GDx nerve fiber analyzer integrates a fixed corneal compensation algorithm (fLxed corneal compensation algorithm, FCC) to overcome the birefringence of the cornea. Some studies have confirmed that the magnitude and axial position of the corneal polarization are variable and related to the RNFL thickness detected by SLP. This product does not completely overcome the birefringence of the cornea and therefore deviates from the fixed corneal compensation. Moreover, the uncorrected corneal birefringence expands the standard data for RNFL thickness, thus significantly reducing the sensitivity and specificity of this technique ^[1] .

2. Variable corneal compensation mode

The GDx variable corneal compensation algorithm (VCC) uses the macular birefringence pattern to determine eye-specific corneal birefringence, and has been used in clinical applications to neutralize corneal birefringence with macular degeneration. The delay parameter shows the difference from standard data statistics by color block code. A two-dimensional spatial RNFL probability plot shows the statistical likelihood of glaucoma damage. Compared with fixed corneal compensation (FCC), the correlation between GDx VCC and visual function has increased, greatly improving the ability to discriminate optic nerve damage in glaucoma "J, and enhancing the ability of combined OCT to evaluate RNFL. Medeiros and others have compared GDx VCC, The sensitivity of HRT II and OCT to RNFL detection of suspicious glaucoma. Conclusion: The three are highly specific and similar in sensitivity (the previous reports suggest that the latter two are more sensitive than GDx). After using variable corneal compensation, GDxVCC Sensitivity has increased significantly. Magacho and Mermoud have also used HRT and GDx to measure the same group of patients, and found that the variation of HRT measurement values is greater than GDx ^[1] .

3 Enhanced corneal compensation mode

The previous literature reports that the VCC mode is better than the FCC mode, but in the detection of some myopia patients, atypical delayed images will be generated. These images are due to the low signal-to-noise ratio in the SLP measurement, which affects the measured RNFL delay value. Accuracy. Researchers have therefore designed an enhanced corneal compensation algorithm (ECC). A large birefringent oblique line is added to the measurement path in this mode to increase the total amount of retardation. In the direct measurement results of VCC mode, the amount of delay is low. The birefringence is obtained every time the macular area is measured, and then the actual value of the measurement is converted into the delay value of the RNFL by calculation. The ECC mode is more effective in excluding the interference of the anterior segment polarization and the sensitivity to detect neurofibrous layer lesions ^[1] .

Toth and Hollo compared RNFL measurements for patients undergoing LASIK and concluded that the ECC mode can better neutralize atypical polarized images and the corneal-induced retardation value than the VCC mode ^[1] .

Clinical Performance of Polarized Laser Scanner

1 . Effectiveness

Comparing the RNFL defect sites shown in SLP and non-red fundus photographs, it is found that the SLP image compensated by the anterior segment birefringence can better reflect the true structure of RNFL, but the histological confirmation of the human eye has yet to be established. The variable anterior segment birefringence compensation has been proven to make accurate estimates of the corneal polarization axis and polarization amount in normal eyes and most patients with macular lesions ^[1] .

2 Sensitivity and specificity

In the normal population, the thickness of RNFL has a wide range. Researchers have used polarization-sensitive OCT (polarization-sensitive OCT) to scan normal people's RNFL and found that although it is stable in the scanning radius, it is measuring the nerve fibers around the optic nipple The thickness of the layer thickness obtained is highly volatile, and there is a large overlap between the glaucoma population and the normal population. The sensitivity and specificity for the examination of advanced glaucoma are higher than those of mild to moderate glaucoma. In a study comparing OCT and SLP, it was found that the structural information from both technologies was significantly correlated with glaucoma visual function. However, the RNFL thickness results (average thickness and total measurement results) provided by the delay parameters are less correlated with the average defect of the visual field than the structured delay parameters (such as adjusted scores, rate parameters and numbers) ^[1] .