Fig 1.
Principle of polarimetric interferometry of the corneal stroma. The incident light projected by the LED light source passes through the first polarized filter (B) and illuminates the cornea. The backscattered light after interaction with corneal tissue passes through the second cross-polarized filter (A) and is captured by the detector camera.
Fig 2.
Polarimetric interferometry images of the cornea obtained during acquisition and after software elaboration.
Frames of the acquired raw sequence of a single representative subject (A and B) showing the corneal cross-shaped pattern (A) and the hyperbolic-shaped pattern (B) of the interference figures. The entire sequence of at least 80 images was used to calculate a background illumination image (C). The aforementioned frames were reproduced after subtraction of background luminance and threshold filtering (D, E). A final summary static image (SUM image) was then calculated by further post-processing of the entire sequence and indicates areas where light does not change its polarization during the acquisition (F).
Fig 3.
Corneal polarimetric interferometry of type B pattern.
Frames of a representative raw sequence (A-E) in a case in which the isogyre gradually changes its corneal cross pattern (A) to form two distinct hyperbolic arcs in opposite quadrants (B-D). The maximal separation is at 45 degrees with respect to angle of cross-shaped pattern formation (C). At 90° rotation of the LCP with respect to frame A, the cross-shape pattern is again visible. (F) The final summary static image (SUM image) is characterized by a separation of the figure into two distinct zones localized in the corneal mid periphery along the horizontal meridians. Peripheral isochromes are visible on individual photograms.
Fig 4.
Corneal polarimetric interferometry of type A pattern.
Frames of representative raw sequence (A-E) in a case in which the isogyre maintains its corneal cross-shaped pattern with the rotating scan of the LCP. (F) The final summary static image (SUM image) is characterized by a single “pear shaped” dark area with the major axis along the horizontal meridians. Peripheral isochromes are visible on individual photograms.
Fig 5.
Processed images of four representative cases (right eyes).
The first row (case 1: a-d) shows a type A pattern while the last row represents a typical B pattern (case 4: m-o). Intermediate rows (case 2: e-h, and case 3: e-l) show two patients with intermediate appearance of interference figures, classified as B pattern because of two distinct optic axis (foci) in the SUM image (h, l). Starting from the left, the first column represents the maximum cross-shape pattern appearance (e, i, m), the second an intermediate phase (f, j, m) at 22°30' degree of rotation and the third the maximum hyperbolic arcs separation (g, k, o) at 45°. The last column shows the SUM images (d, h, l, p). Dashed red lines highlight the profiles of the hyperbolic arcs in clearly biaxial cases (Type B pattern) (case 2, 3 and 4). Asterisks mark possible location of the interference figure optic axis (foci).
Fig 6.
Enantiomorphism phenomenon in corneal polarimetric interferometry.
Summary images that illustrate corneal interference pattern enantiomorphism of the right eye (RE) and left eye (LE) of two representative patients. (A, B) “Pear shaped” uniaxial pattern (type B). (C, D) Double asymmetric peripheral pattern (type A): note that the interference patterns present nasal-inferior shift between + 15°and—15° (respectively for the right and left eye) with respect to the horizontal meridian (0°–180°).
Fig 7.
Corneal isochromatic interference figures in polarimetric interferometry enhanced after subtraction of background image.
Colored image showing the peripheral corneal isochromes, with a rounded square morphology. The matching changes of colour bands of the 1st through 3rd order of the Michel-Levy colour sequence are shown in the colour overlay.