Structured Light Tridimensional Assessment of Lower Eyelid Ectropion: A New Technique

INTRODUCTION

Collection of accurate image documentation is essential in reconstructive surgery (1). This makes it possible to compare preoperative and postoperative parameters, and it facilitates communication with the patients (1, 2, 5). A valid image archive is a useful tool for surgeons in order to improve current techniques, develop new techniques, guide surgery choices (1, 2, 5), focus on and understand procedure errors, and perform a careful follow-up (1). The realization of a two-dimensional image has always supported the practice of surgery, especially reconstructive surgery. However, the human body is a three-dimensional object, and many studies (2, 3, 4) have shown that two-dimensional image documentation is often inadequate. Two-dimensional image documentation allows a limited number of measurements that are more variable and difficult to calibrate than manual measurements made on the patient directly (3).

In the past, many methods have been investigated for the realization and the recording of three-dimensional facial measurements (6) including laser scanning (2, 7), electromagnetic 3D digitizers (8), stereophotogrammetry (3, 9) developed according to the advances in computer technology, and others. A new application for the structured light three-dimensional scanner measurement system is presented in this paper: the lower eyelid ectropion.

This measurement device projects a known grid of structured light on an object, and a CCD Camera records the deformations undergone by the incident light. Then a dedicated piece of software processes the lines and warps back to the exact shape of the object struck by light. This three-dimensional skin measuring system has an accuracy of 6 mm, a data acquisition time of 68 msec and is contact-free (data provided by manufacturer).

Lower lid ectropion is a very common condition in older persons. Frequency increases steadily with age. Defined as an eversion of the eyelid away from the globe, the condition is classified according to its anatomic features as involutional, cicatricial, tarsal, congenital, or neurogenic/paralytic. One of the main causal factors is horizontal lid laxity, usually due to age related weakness of the canthal ligaments and the pretarsal orbicularis. It has been suggested that patients with involutional ectropion have an age-normal (or larger than normal) tarsal plate, and a normal or decreased orbicularis tone, in conjunction with canthal tendon laxity.

Ectropion can be bilateral or unilateral, congenital or acquired (10). When contact between the lower eyelid and the scleral surface is lost, the ocular globe is unprotected below, and there is insufficient lachrymal fluid. The most common sequelae of the ectropion are: dryness of the anterior ocular surface, epiphora, conjunctivitis and corneal erosions.

In this study we evaluated the accuracy of structured light three-dimensional imaging of the lower eyelids, the reproducibility of this new method and the correlation of the three-dimensional morphology with ocular discomfort signs.

MATERIALS AND METHODS

This exploratory evaluation of a new diagnostic technology was performed in accordance with the ethical standards stated in the Declaration of Helsinki and was approved by the institutional review board. No exclusion criteria were considered.

The investigation was conducted on 10 patients (all males) who were enrolled in the study after signing informed consent. 6 patients suffered from unilateral involutional ectropion, and 4 patients suffered from bilateral involutional ectropion. Their age ranged between 70 and 82 years with a mean age of 76.1 years.

The three-dimensional eyelid images were captured by the optical measuring device “PRIMOS-Pico” (Phaseshift Rapid In vivo Measurement Of Skin), made by GFMesstechnik GmbH. This device is a calibrated structured light three-dimensional scanner supported by the “PRIMOS 5.6” software. It uses a digital stripe projection technique, based on a digital micro mirror projector (DMD: Digital Micro mirror Devices) of Texas Instruments USA, with 800 X 600 micro mirrors. The stripes are projected onto the surface of the measuring object, and their projections are recorded with a defined triangulation angle by a CCD camera. The topography of the measuring object is calculated from the stripes position and the value of all registered individual image points.

The following were obtained after processing with the dedicated software: a two-dimensional image, a three-dimensional image (Fig. 1), a grey scale code image for topographical evaluation (Fig. 2), and 12 profilometries plotted along previously selected lines (Fig. 3, Fig. 4).

The medial canthus – lateral canthus distance was assessed on the patients directly with a ruler. This distance was taken as a reference to verify the calibration and reliability of the digital measuring system. By means of the same ruler, on the 6 patients with unilateral ectropion the investigators tried to measure the distance between the scleral surface and the lower eyelid external free margin in the middle point, and these measurements were compared to the digital ones.

On the grey scale coded topography image (convex surfaces are light grey, concave ones dark grey) 12 vertical selection lines were plotted with the dedicated software. These lines were spaced at intervals of 2 mm starting from the medial canthus (see Fig. 3). These lines were grouped, 4 lines in every group, in 3 regions referred to the lower lid: medial, intermediate and lateral. For every line a two-dimensional profilometry has been extracted. To build a profilometry is a prerogative of the three-dimensional reconstructions software. That is not possible on two-dimensional images. The profilometry allows the investigators to do Z-axis measurements and provides a three-dimensional understanding of the sample. In order to diagnose the right degree of ectropion, the distance between the scleral surface and external free margin of the lower eyelid was assessed along each line on the profilometry, by means of the manufacturer’s software (see Fig. 4), and an arithmetic mean for each region was calculated.

Two different investigators took each measurement twice on two different days. All subjects received complete ophthalmological evaluation with particular regard to the presence of ocular surface changes and discomfort.

Statistical analysis

Arithmetic means and 95% confidence interval (95% CI) were used to describe continuous variables. A t-test was applied to compare by arithmetic means the manual and computer measurements of the two investigators.

Bland and Altman (11) plots were constructed to compare and assess the inter-observer and intra-observer reproducibility of this new method with the MedCalc vers. 7.4.3.0 software.

The repeatability coefficient was calculated to assess the variation in taken by on the same measurement and under the same conditions. The repeatability coefficient is a precision measure that represents the value below which the absolute difference between two repeated test results may be expected to lie with a probability of 95%.

RESULTS

In the unilateral ectropion group (six patients) the arithmetic mean of the distance between the scleral surface and the lower eyelid external free margin measured on the profilometry in the medial region is 2.3 mm (min. 1.9 mm – max. 3.4 mm). The arithmetic mean in the central region is 3.8 mm (min. 2.9 mm – max. 5.1 mm). The arithmetic mean in the lateral region is 2.8 mm (min. 2.0 mm – max. 4.2 mm).

In the bilateral ectropion group (4 patients) the arithmetic mean in the medial region is 2.4 mm (min. 2.0 mm – max. 3.5 mm). In the central region is 4.0 mm (min. 3.2 mm – max. 5.2 mm). In the lateral region is 3.3 mm (min. 2.4 mm – max. 4.5 mm).

Four patients (2 from the bilateral group and 2 from the monolateral group) had some ocular surface changes and discomfort (all the patients reported the sensation of sand in the eye, and 3 patients had corneal surface alterations). In this subgroup the arithmetic mean of the distance between scleral surface and the lower eyelid external free margin measured on the profilometry in the medial region is 2.0 mm (min. 1.9 mm – max. 2.8 mm). The arithmetic mean in the central region is 4.5 mm (min. 3.4 mm – max. 5.2 mm). The arithmetic mean in the lateral region is 3.1 mm (min. 2.8 mm – max. 4.0 mm).

The t-test performed between the means of the two manual measures series and the two computer measures series of interchantal distance showed no significant difference (p>0.05). Moreover, in the manual measures, which obviously show less spatial resolution than computer ones, the mean of the differences between the two series (Investigator 1, Investigator 2) is – 0.35 mm instead of 0.02 mm for the computer ones.

The t-test performed between the means of the two manual measures series and the two computer measures series of the distance between the scleral surface and the lower eyelid external free margin measured in the 6 patients with unilateral ectropion showed no significant difference (p>0.05). The investigators reported that it was very difficult and time-consuming to take the manual scleral surface lower eyelid external free margin measurement with a ruler.

The Bland and Altman plot for intra-observer reproducibility (Graph1) shows an arithmetic mean of the difference of 0.03 mm and a standard deviation of 0.07 mm. The same plot made for inter-observer reproducibility (Graph 2) shows an arithmetic mean of the difference of 0.02 mm and a standard deviation of 0.07 mm. In the two plots there is a homogenous distribution of the differences between the two data set of measurements and an agreement inside the 95% of the confidence interval.

The repeatability coefficient of the first Investigator was 4.3%. The repeatability coefficient of the second Investigator was 3.5%.

DISCUSSION

As Nunu et al. said in their study (6), the human face is a three-dimensional structure, and an adequate technique of taking images and measures should be able to record it. Two-dimensional photographs are not able to fix the depth dimension, even if frontal and lateral profiles are viewed together, are highly ambient light dependent, and are very difficult to calibrate (6). The development of optical 3D shape measurement methods is rapidly gaining importance as the industry increases its demands for high-tech performance of final products, short production times, low manufacturing costs and overall product quality. In this paper we present a new method of lower eyelids ectropion evaluation with a structured light three-dimensional scanner, as an example for the use of this modern technology in periocular area evaluation. A structured-light three-dimensional scanner is a device for measuring the three-dimensional shape of an object using projected light patterns and a camera system. It produces a three-dimensional image that contains accurate information of the three space coordinates.

The results of the statistical tests in this study emphasize that this new method is accurate, repeatable and reproducible in assessing lower eyelid ectropion as well as other eyelid malpositions or orbital morphological alterations. The structured light three-dimensional images evaluation provides better spatial resolution than manual measures with a more rapid acquisition time for a large data set.

The dedicated software performs a large number of measurements from a single frontal three-dimensional scan that are not possible in two-dimensional images or directly on the patient. Accurate and objective diagnosis is possible due to the very high level of accuracy of this portable measurement system and the possibility of overlapping subsequent images. The high-resolution profilometry allows depth measurements along the z-axis, which is only feasible on three-dimensional reconstructions. Furthermore, the very short image acquisition time (68 msec) minimizes movement artefacts. This new method of evaluation transforms the diagnosis of ectropion from purely clinical and supported by a small number of measurements, to objective, measurable, accurate and reproducible in all the three dimensions of the space. It is now possible to examine the lower lid with a high-resolution imaging technique and to distinguish objectively the lower lid ectropion in predominantly lateral, intermediate or medial planes. The results of this study showed a correlation between central and lateral ectropion affected patients with ocular surface discomfort. In the near future further studies will analyze the different clinical presentations and the best surgical choice in every different case. One of the most important benefits of this three-dimensional measuring device is the opportunity to work with an interactive archive, on which new measurements of any kind are always possible. Moreover overlapping images allow the control of the evolution of the eyelids malposition or the achievement and maintenance of the therapeutic outcome (12).

This study establishes a protocol for using the structured light three-dimensional scanner measurement system for the assessment of the lower eyelid ectropion, and similar studies with larger number of patients are recommended.

It is clear that the application of this method to the other eyelids malpositions and orbital area morphological alterations can be fruitful, both in the diagnosis and for the pre- and post-treatment evaluation.

CONCLUSION

The aim of this study was to outline a protocol to diagnose objectively the inferior eyelid ectropion with a three-dimensional preoperative reconstruction made by a structured light 3D scanner and to assess the accuracy and reproducibility of this new method.

The investigation was conducted on 10 patients: 6 with unilateral ectropion, 4 with blateral ectropion. The measurements were taken on two successive days by two different investigators on the patients, on frontal three-dimensional scan and on profilometry.

The measurements taken on the 3D reconstruction were more accurate than in vivo measurements. The structured light 3D scanner makes it possible to distinguish objectively a diagnosis of ectropion in lateral, intermediate and medial planes.

The structured light scanner is a portable and very accurate measuring device for good three-dimensional reconstruction of the eyelids and periocular area and allows transforming the ectropion and other eyelids malpositions from a purely clinical diagnosis to an objective diagnosis.

Address for correspondence:

Dr. P. Mezzana

Via Merulana 61/A

00185 Rome, Italy

E-mail: pmezzana@gmail.com