JP2009291517A - Method of analyzing sectional image of anterior ocular segment and apparatus of photographing anterior ocular segment, recording medium thereof and program thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an analysis method for evaluating simulatively the visibility of an objective site observed by a gonioscope using the sectional image of an anterior ocular segment. <P>SOLUTION: The analysis method of analyzing the sectional image of an anterior ocular segment including a corneal vertex and a chamber angle area by optometry comprises: a step for selecting a sectional image to analyze; a step for setting information about the sectional position of the selected sectional image; a step for setting a standard straight line within the sectional image; a step for setting an analysis target at a characteristic site within the chamber angle area; a step for setting a gonioscopic information for the standard straight line; a step for setting a straight line which passes through the point to be analyzed as an analytical straight line based on the standard straight line and the gonioscopic information; and a step for analyzing on whether a certain tissue is present on the analytical straight line or not based on the information about the position or brightness of a picture element which composes the sectional image present on the analytical straight line. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention simulates the appearance of a corner area of an eye to be observed by a corner mirror based on a plurality of cross-sectional images acquired by an apparatus capable of photographing a cross section of an anterior segment used in the ophthalmic field. In particular, the present invention relates to a method for acquiring information for reference in the evaluation of angle-closure closely related to glaucoma, and an apparatus / storage medium and program using the method.

  One of the diseases handled in the ophthalmic field is glaucoma. Glaucoma is a disease in which the discharge of fluid circulating in the eyeball, called aqueous humor, worsens and causes an increase in pressure in the eyeball, thereby causing damage to the optic nerve that has been compressed. There are various factors that worsen the discharge of aqueous humor, but one of the reasons is the narrow circulation path.

  FIG. 1 shows an outline of an eyeball 10. The aqueous humor circulated in the eyeball 10 is discharged through the anterior eye portion shown as the region A. FIG. 2 is an enlarged view of the anterior segment, and is composed of tissues represented by the cornea 11, the sclera 12, the iris 13, the crystalline lens 14, and the like. In particular, a region sandwiched between the cornea 11 and the inner surface of the sclera 12 and the iris 13 indicated by a region B is called a corner angle. FIG. 3 is an enlarged view of the corner area (area B in FIG. 2). Most of the aqueous humor is discharged from a site called the trabecular meshwork TM that exists between the scleral cape SS (the boundary between the sclera 12 and the iris 13) and the Schwalbe line SL (the boundary between the cornea 11 and the sclera 12). Therefore, observation of the corner area has been given as an important examination item for glaucoma. Non-Patent Document 1 defines many criteria for classifying in detail for each symptom in order to reflect the factors of onset in a wide range of treatments for glaucoma, among which are based on observational observations at the corners. Some are classified.

For the observation of corners, inspection contact lenses called corner mirrors are widely used. FIG. 4 shows a basic structure of a corner mirror classified as an indirect observation type, and the inside of the housing 21 is composed of a main body 23 and one or a plurality of inclined mirrors 22. The inclination of the mirror 22 is arranged in a positional relationship of an angle θ _M with respect to the observation plane Pv so that an image in the vicinity of the trabecular meshwork TM is observed when the central axis of the angle mirror is aligned with the center of the eyeball 10. Has been. The main body 23 is an optical member having high light transmittance, and _ng in parentheses is the refractive index of the member. The contact surface Pc that is in contact with the cornea 11 of the eyeball 10 when observing the corner is a curved surface along the eyeball surface having a radius indicated by Rg in FIG. Here, Dg shown in FIG. 4 is the diameter of the eyepiece of the angle mirror. The mirror angle θ _M , the curved surface radius Rg of the contact surface Pc, and the diameter Dg of the eyepiece are prepared in various specifications depending on the manufacturer or the number of mirrors to be incorporated.

従って、観察対象となる部位の位置とθ_Mが特定されることにより、ＶＢとして観察される光線の観察部位からミラー２２までの経路に相当する直線ＶＢ’が導き出されることになる。しかしながら、角膜と隅角鏡の境界等における屈折により、眼球１０の内外の観察光線の経路を同一直線によって表すことはできない。図５（ｂ）は、角膜と隅角鏡の接眼面における屈折の影響を示したものである。図５（ａ）に示すＶＢのように、観察面Ｐvに対して垂直方向に進行する光線は、ミラー２２に入射する角度がθ_αのものに限定される。しかしながら、角膜１１と隅角鏡２１の境界における屈折の影響を受け直線ＶＢ０に沿って進行する光線は、観察部位ＣＰを通過しない直線ＶＢ０’に沿って点Ｑ０に到達したものであるため、ミラー２２に入射する角度がθ_αであるが観察面Ｐvに観察部位ＣＰの像を導くものではない。観察面Ｐvに向けて観察部位ＣＰの像を導く光線は、観察部位を通過した光線が屈折によりミラー２２に角度θ_αで入射するようになったもので、図５（ｂ）では観察部位ＣＰならびに点Ｑを通る直線が該当する。この直線は、前述のミラー角度θＭの他に隅角鏡の形状や角膜１１の曲率等の情報に基づいて特定することが可能であり、手順については実施例にて述べる。 FIG. 5 shows the relationship between the angle mirror and the position of the eyeball 10 and the observation light beam at the time of observation. The image in the corner area travels along the straight line VB ′ shown in FIG. 5A, is reflected by the mirror 22 in a direction parallel to the central axis of the corner mirror, and is guided to the observer as an observation light beam VB. In this case, an observation image from a direction forming an angle of Pv 'and theta _alpha is incident on the mirror 22. Since Pv ′ is parallel to the observation plane Pv, it is geometrically derived that θ _α and θ _M have the following relationship.

Therefore, by specifying the position of the site to be observed and θ _M , a straight line VB ′ corresponding to the path from the observed site of the light beam observed as VB to the mirror 22 is derived. However, due to refraction at the boundary between the cornea and the angle mirror, the path of the observation light beam inside and outside the eyeball 10 cannot be represented by the same straight line. FIG. 5 (b) shows the influence of refraction on the eyepiece surface of the cornea and the angle mirror. Like VB shown in FIG. 5A, light rays traveling in a direction perpendicular to the observation plane Pv are limited to those having an incident angle of θ _{α on} the mirror 22. However, the ray traveling along the straight line VB0 under the influence of refraction at the boundary between the cornea 11 and the corner mirror 21 has reached the point Q0 along the straight line VB0 ′ that does not pass through the observation site CP. the angle of incidence 22 is a theta _alpha but does not lead to image the observed region CP in the observation plane Pv. Light directing an image of the observed region CP toward the viewing surface Pv is intended that light rays passing through the observation site was made incident at an angle theta _alpha to the mirror 22 by refraction, and FIG. 5 (b) the observed region CP In addition, a straight line passing through the point Q is applicable. This straight line can be specified based on information such as the shape of the angle mirror and the curvature of the cornea 11 in addition to the aforementioned mirror angle θM, and the procedure will be described in the embodiment.

  In recent years, various manufacturers have provided devices capable of taking images near the corners using optical or ultrasonic waves and observing the taken images, and have been used for acquiring useful information in diagnosis. The angle mirror is much cheaper than the above-described photographing apparatus and is still widely used worldwide. However, it is impossible to always hold the angle mirror at the same position with respect to the eyeball 10, and not only when the observers are different, but also by the same observer depending on the positional relationship between the eyeball 10 and the angle mirror. There is a problem that the observation results vary.

  The above-mentioned device that captures the vicinity of the corner angle improves the resolution of the acquired image by observing the vicinity of the corner angle by using a high-frequency ultrasonic wave that has been conventionally used for tomographic image capturing. A so-called ultrasonic microscope (UBM) has appeared.

  Further, in Patent Document 1, a slit projection optical axis passing through the corneal apex of an eye to be examined is used as a rotation axis, and an optical system configured based on the Shine-Pluke principle is rotated to simultaneously photograph an eyeball cross-section. In order to identify the position of the lens, a rotation angle information is obtained, and a plurality of photographed cross-sectional images are correlated based on the rotation angle information, so that a position such as turbidity in the crystalline lens can be grasped three-dimensionally. ing. FIG. 6 shows the relationship between the photographing position (cut plane) and the photographed image in photographing with the optical system rotated. FIGS. 6A and 6B show the cornea 11 and sclera 12 constituting the anterior segment. 6 (c) to 6 (e) are views of the crystalline lens 14 viewed from the front and side, and each of C1-C1 to C3-C3 having different angles passing through the apex P of the cornea 11 shown in FIG. 6 (a). The cross-sectional image image | photographed in the position is shown. As shown in FIGS. 6C to 6E, images obtained by photographing the cross section passing through the apex P of the cornea 11 have the same size of the anterior eye part included in the photographing range even if the photographing position is moved. For this reason, the acquired images are almost the same size. Here, when C1-C1 is set as the reference plane, the rotation angle information of the images photographed at the positions C2-C2 and C3-C3 is 45 ° and 90 °, respectively.

  Furthermore, in Patent Document 2, a three-dimensional image is constructed from a cross-sectional image of the fundus region of the eye to be inspected by an optical coherence tomography apparatus (OCT) that can capture a cross section of an object with high resolution using optical interference. Techniques to do this are disclosed. The OCT shown in Patent Document 2 is an apparatus for imaging the fundus region, but OCT for imaging the anterior eye part including the corner area has also appeared. The apparatus disclosed in Patent Document 1 A cross-sectional image of a corner area with higher resolution can be acquired. FIG. 7 shows the relationship between the photographing position (cut plane) and the photographed image in photographing adopted in OCT, and FIGS. 7A and 7B show the cornea 11 and sclera 12 constituting the anterior segment. FIGS. 7C to 7E show cross-sectional images taken at the respective positions C1′-C1′-C3′-C3 ′ when the lens 14 is viewed from the front and side. As shown in FIGS. 7C to 7E, the photographed image is C2′-C2 ′ obtained by linearly moving the photographing position parallel to C1′-C1 ′ passing through the apex P of the cornea 11. And the image captured at the position of C3′-C3 ′, as the eyeball moves to the peripheral part of the eyeball, the anterior eye part included in the imaging range becomes smaller, and the acquired image does not include the corner area. As a result, the image that can be used for observing the corner area is only the image that was taken in the very vicinity of C1′-C1 ′. However, the image taken while moving up and down in the range of C0′-C0 ′ to C4′-C4 ′ shown in FIG. 7A corresponds to the rotation angle information described in Patent Document 1. By associating position information related to the shooting position in the vertical direction of the shooting position, it is possible to construct three-dimensional image data, and from the constructed three-dimensional image data, the same as in FIGS. 6C to 6E. It is possible to extract a simple cross-sectional image in an arbitrary direction.

  Note that the apparatus disclosed in Patent Document 1 or Patent Document 2 is provided by many manufacturers that can perform imaging without contact with the eyeball, and includes positioning and imaging of the eyeball and the apparatus. Many apparatuses are provided with means for detecting the positional relationship between the eyeball and the apparatus for automatic operation. Such an apparatus reduces the burden on the eyeball (subject) and achieves stable image acquisition that is not affected by the skill level of the operator of the apparatus.

However, in the standard related to the observation of the corner angle described in Non-Patent Document 1 described above, it is determined that the classification is performed based on the observation performed using the angle mirror. Therefore, in order to comply with Non-Patent Document 1, although the above-described imaging apparatus can reduce the burden on the eyeball and can observe a stable corner region without being affected by the skill level of the operator. The current situation is that the observation with a corner mirror is performed separately.
JP 2003-111731 A JP 2007-117714 A Japan Glaucoma Society, “Glaucoma Diagnosis Guidelines (2nd Edition)”, Journal of the Japan Eye, 2006, 110, 10, 777-814

  The present invention has been made in view of the above circumstances, and its object is to use a cross-sectional image of the anterior segment of the eye, specifically, an indirect observation type of angle spectroscopic inspection. It is an object to provide an anterior ocular segment cross-sectional image analysis method, an anterior ocular segment imaging device, a recording medium thereof, and a processing program thereof that enable the same evaluation as described above.

  In order to solve the above problems, the present invention is characterized by having the following configuration.

  The method for analyzing an anterior segment cross-sectional image according to claim 1 is a method of selecting a cross-sectional image to be analyzed in the anterior segment cross-sectional image analysis method including the corneal apex and the corner region of the eye to be examined; Setting information on the cross-sectional position of the selected cross-sectional image, setting a reference straight line in the cross-sectional image, setting an analysis target point in a characteristic part of the corner region, A step of setting corner mirror information with respect to a reference line, a step of setting a straight line passing through the analysis target point based on the reference line and the corner mirror information as an analysis line, and existing on the analysis line The method includes a step of analyzing whether or not a predetermined tissue exists on the analysis straight line based on the position of the pixels constituting the cross-sectional image or luminance information.

  The method for analyzing an anterior segmental cross-sectional image according to claim 2, wherein the information on the cross-sectional position is associated with the analysis result acquired in the step of performing the analysis, and the analysis result is classified according to the region where the analysis target point is set. It has the step which performs a classification | category, It is characterized by the above-mentioned.

  The anterior ocular segment image analysis device according to claim 3 is an ophthalmologic imaging apparatus having an imaging unit capable of acquiring a cross-sectional image of the anterior ocular segment including a corneal apex and a corner area of the eye to be examined. A cross-sectional image selecting means for selecting an image; a display means for displaying the cross-sectional image; a cross-sectional position information setting means for setting information on a cross-sectional position of the cross-sectional image displayed on the display means; and the display means. A reference straight line setting means for setting a reference straight line on the cross-sectional image, an analysis target point setting means for setting an analysis target point at a characteristic part of the corner angle region, and angle mirror information for the reference straight line. Corner mirror information setting means, analysis straight line setting means for setting a straight line passing through the analysis target point whose angle intersecting the reference straight line is determined based on the corner mirror information, and the analysis for And having an analysis means for analyzing whether a given tissue is present in the analysis straight line based on the position or brightness information of the pixels constituting the cross-sectional image that exists on the line.

  The anterior segment cross-sectional image analysis apparatus according to claim 4 associates the information on the cross-sectional position with the analysis result acquired by the analysis unit, and classifies the analysis result according to a region where an analysis target point is set. It has an analysis result classification means.

  The anterior segment cross-sectional image analysis apparatus according to claim 5 is characterized in that the analysis result classified by the analysis result classification means is displayed on the display means by numerical value or graphic for each analysis target point.

  The recording medium according to claim 6 performs the analysis in a computer-processable recording medium in which a program for analyzing a cross-sectional image of the anterior ocular segment including the corneal apex and the corner area of the eye to be examined is recorded. A step of selecting a cross-sectional image, a step of setting information of a cross-sectional position of the selected cross-sectional image, a step of setting a reference straight line in the cross-sectional image, and an object to be analyzed in a characteristic part of the corner area A step of setting a point, a step of setting angle mirror information with respect to the reference line, and a line passing through the analysis target point at which an angle intersecting the reference line is determined based on the angle mirror information A predetermined tissue on the analysis line based on the step of setting as a straight line and the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis line And executes a step of performing analysis on whether standing.

  The program according to claim 7 is a computer-processable program for analyzing a cross-sectional image of an anterior ocular segment including a corneal apex and a corner area of an eye to be examined, and selecting a cross-sectional image to be analyzed; Setting information on the cross-sectional position of the selected cross-sectional image, setting a reference straight line in the cross-sectional image, setting an analysis target point in a characteristic part of the corner region, Setting a corner mirror information for a reference straight line, setting a straight line passing through the analysis target point at which an angle intersecting the reference straight line is determined based on the corner mirror information as the analysis straight line; Whether or not a predetermined tissue exists on the analysis straight line based on the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis straight line And executes a step of performing analysis.

  With the configuration described above, the visibility of the part to be observed with the angle mirror can be evaluated in a pseudo manner using the image of the cross section of the anterior segment of the eyeball. It is possible to preliminarily provide the doctor with information that can be used as a reference when determining the region where the focus is placed on the subject and determining whether or not the observation with the angle mirror is necessary.

  Hereinafter, in order to clarify the present invention more specifically, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 8 shows an external appearance of an eyeball photographing apparatus that acquires an eyeball cross-sectional image by ultrasonic waves according to the first embodiment of the present invention. In general, the acquisition of an ultrasonic image is not expressed as “imaging”, but the intensity of an ultrasonic echo signal reflected and received by an object is reflected in luminance. The image data includes information equivalent to image data acquired by an optical imaging device using a light receiving means such as a CCD. Therefore, in this specification, the acquisition of an image using ultrasound is also referred to as “imaging”. Describe.

  The eyeball photographing device 100 includes a circuit (not shown) necessary for controlling the device, a liquid crystal display 108 for displaying a cross-sectional image and an analysis result, and a rotary knob for inputting information necessary for device operation and analysis. 106, a touch panel 107, a printer 110 that prints analysis results and the like, and a probe 104 that transmits and receives ultrasonic waves to photograph, and the main body and the probe 104 are connected via a connection connector 105.

  A probe for transmitting and receiving ultrasonic waves and a drive mechanism (not shown) are provided inside the probe 104, and ultrasonic signals can be transmitted two-dimensionally by driving the probe linearly or slidingly. ing. In addition, since the transmission / reception range of the ultrasonic signal is changed by changing the transmission frequency and driving range of the probe, by replacing another probe 104 ′ having a different driving mechanism from the probe in the connection connector 105, FIG. ~ Images in the range shown in Fig. 3 can be taken. For example, since the ultrasonic wave propagates farther as the transmitted frequency is lower, the entire eyeball 10 shown in FIG. 1 can be photographed when the frequency is about 10 MHz. However, when the frequency is increased to 50 MHz, the surface of the lens 14 Since it propagates only to the vicinity, the photographing range becomes as narrow as shown in FIG. However, since the resolution of an image captured by increasing the frequency is improved, it is more effective to set the frequency higher in order to perform a precise inspection. Here, the frequency switching and drive control of the probe are realized by using a number of techniques disclosed heretofore. Further, the touch panel 107 is attached to the surface of the liquid crystal display 108, and various operations of the apparatus can be performed by touching a display area of input items displayed on the screen of the liquid crystal display 108.

  FIG. 9 shows a schematic configuration of the eyeball photographing apparatus 100 of FIG. A control circuit 101 for driving and controlling the entire apparatus includes an arithmetic circuit 102 for performing analysis and the like, a memory 103 for storing images or analysis results, and the input devices 106 and 107 shown in FIG. The printer 110 and the probe 104 are connected, and various controls such as photographing, image display, and analysis are performed by transmitting a control signal in the direction of the arrow shown in the drawing. The operation of the apparatus having the above configuration will be described. At the time of photographing, the examiner inputs setting of photographing conditions or photographing information from the touch panel 107 and then inputs a driving signal for the probe 104. At this time, information such as an imaging condition input from the touch panel 107 is stored in the memory 103 via the control circuit 101, a drive signal is transmitted to the control circuit 101, and the control circuit 101 drives and probes in the probe 104. The control which transmits an ultrasonic signal from is performed. The examiner irradiates an ultrasonic signal in the eyeball of the subject by bringing the probe 104 into direct or indirect contact with the subject's eye. The irradiated ultrasonic signal is reflected at various tissue boundaries in the eyeball, and the reflected signal is received by the probe. The reflected signal received by the probe is processed by the arithmetic circuit 102 via the control circuit 101 and displayed on the liquid crystal display 108 as a two-dimensional image. At this time, the image data processed in the arithmetic circuit 102 is also transferred to the memory 103, and the image data can be recalled and displayed on the liquid crystal display 108 even after the driving of the probe 104 is stopped. Further, by inputting information such as the shooting position of the display image from the touch panel 107, the information is stored in the memory 103 in association with the image being displayed. Here, the memory 103 can store a plurality of image data. If the shooting position or the like is determined in advance, it may be performed in accordance with the input of shooting conditions before driving the probe 104 to perform shooting.

  FIG. 10 shows whether or not the angle mirror observation of the characteristic part of the corner area such as SS / TM / SL shown in FIG. 3 is performed using the image obtained by photographing the range shown in FIG. It is the flow which showed the procedure which performs evaluation about.

  In S1, a cross-sectional image to be analyzed is selected. As described above, in the eyeball photographing apparatus 100 of the present embodiment, the ultrasonic signal received by the probe 104 is sent to the memory 103 in addition to the liquid crystal display 108 via the arithmetic circuit 102. Therefore, it is possible to select an image acquired in the past as an object of analysis by calling an image stored in the memory 103, and the operator can select an image displayed on the liquid crystal display 108 and the memory 103. An image to be analyzed is selected from the stored images.

  Subsequently, in S2, the influence of the inclination of the eyeball image in the photographed image is eliminated with respect to the observation light beam in the eyeball indicated by VB ′ described above for the eyeball image whose direction is different for each image due to the influence of the eye movement at the time of photographing. A reference straight line BL is set as a reference for setting under different conditions. FIG. 11 shows an example of a method for setting the reference straight line BL, in which the normal line of the apex P of the cornea 11 that is generally used as a reference for alignment of the apparatus is set as the reference straight line BL. First, as shown in FIG. 11A, the operator is allowed to designate the position of the vertex P in the display image. The designation is performed by touching the touch panel surface at the corresponding position of the display image with a stick having a thin tip, or by moving a cursor movable to an arbitrary position displayed superimposed on the image to the corresponding position. Subsequently, using the luminance distribution in the vicinity of the vertex P indicated by the region D in FIG. 11B, two points on the surface of the cornea 11 located at a predetermined distance from the vertex P (in the region D of FIG. 11B). By automatically setting both ends: CFL, CFR), the surface curved surface of the cornea 11 can be approximated by a circle DC. The CFL and CFR may be specified by the operator in the same manner as the apex P of the cornea 11. Here, the straight line TL in the figure is a tangent at the apex P of the cornea 11 of the approximate circle DC. Therefore, as shown in FIG. 11C, the straight line passing through the center CC of the approximate circle and the vertex P of the cornea 11 is nothing but the normal of the vertex P of the cornea 11 that is orthogonal to the tangent TL at the vertex P of the cornea 11. This straight line can be set as the reference straight line BL. FIG. 11D shows the relationship between the set reference straight line BL and the vertical line VL of the display screen, and the angle Δθ represents the inclination of the cross-sectional image with respect to the display screen. By rotating the image in the direction that cancels this Δθ, the image can always be displayed in the same direction, and it is possible to easily perform the process of eliminating the influence of inclination when using positional information on the screen. As a reference for rotation, a point existing on the reference straight line BL such as the apex P of the cornea 11 may be set.

  Next, in S3, an analysis target point is set. Here, characteristic parts of the corner area (sclera cape SS, trabecular mesh band TM, Schwalbe line SL, etc.) are set as analysis target points. In the setting, the operator designates the corresponding part of the image displayed on the liquid crystal display 108 in the same manner as the vertex P of the cornea 11 is set in S2 described above. FIG. 12 shows a state in which the scleral cape SS is designated as an analysis target point in an image obtained by enlarging and displaying the region B ′ of FIG. 11C in which the reference straight line BL is set in S2 on the liquid crystal display 108. is there. The reason why the enlarged display is performed is that the corner area is small in the image in which the ranges shown in FIGS. 11A to 11D are displayed. Therefore, the enlarged image is more suitable for specifying the analysis target point in detail. It is because it is preferable.

In S4, corner mirror information for virtual observation using a corner mirror is set. The information set here includes the mirror angle θ _M shown in FIG. 4, the radius Rg of the eyepiece surface Pc, the diameter Dg of the eyepiece, and the refractive index of the member of the main body 23. FIG. 13 is an example of an input screen for corner mirror information, and the operator inputs each item.

従って、角膜の頂点Ｐの位置ならびに角膜の表面の曲率半径Ｒcを特定することで、基準０（Ｏg）を正確に設定することが可能になる。本実施形態においては、Ｓ２の基準直線の設定の過程において、角膜の頂点Ｐの位置及び角膜の曲率半径が特定されているため、そのデータを利用することにより、新規に入力を行なうことが不要となる。Ｒc＝７．７ｍｍ、Ｒg＝７．４ｍｍ、Ｄg＝１２ｍｍの場合、ΔＬは約０．２ｍｍとなり、角膜の頂点Ｐから２次元座標の基準までの距離はおよそ７．２ｍｍとなる。
点Ｑは、解析対象点である点ＶＰを通過する観察光線が図５の直線ＶＢ’と同じθαの傾きとなる屈折を生じる隅角鏡の接眼面ＣＳの位置であり、括弧付で記載されているｎ_g、ｎ_cは隅角鏡の本体及び角膜の屈折率を表している。角膜の屈折率ｎ_cの値は１．３７６が一般的に使用されている。また、涙液、防水等を含む眼球を構成する組織の屈折率はおよそ１．３３〜１．３８と水の屈折率に近いため、本発明においては眼球内の組織の境界における屈折は生じないものとして取扱っている。本ステップにおいて行なう解析用直線の設定はＶＰ・Ｑ間の線分ＡＬを特定することであり、これは点Ｑの位置が決定されることで達成される。基準０と点Ｑを結ぶ直線ＯＱがＸ軸に対して角度θの関係である場合、点Ｑの座標は（Ｒg×cosθ，Ｒg×sinθ）と表わされる。従って、解析対象点ＶＰの座標を（ｘ，ｙ）とすると線分ＡＬのＸ軸に対する傾斜角θβは、

また、点Ｑでは点Ｐからの光線の入射角θ_cと（図示しない）隅角鏡のミラーに向かう光線の出射角θ_gの間にはスネルの法則により、

の関係が成立する。ここで、入射角θｃ、出射角θｇを前述のθ、θ_α、θ_βにより表すと、

さらに、θ_αは図５の説明において前述したように隅角鏡のミラー角度θ_Mにより示すことができるため、前述のスネルの法則の式は、

と表すことができる。上式において、直線ＯＱの角度θ以外の値は、解析対象点ならびに隅角鏡情報としてＳ３ならびにＳ４において設定済のため、θを特定することにより解析用直線である線分ＡＬが導き出される。解析対象点ＶＰの座標（−５．８４,３．８）、想定する隅角鏡のミラー角度θ_M＝６２°、接眼面の曲率半径Ｒg＝７．４ｍｍ、本体屈折率＝１．５２である場合、θ＝７８．５°と特定されるため、点Ｑの座標（１．４８，７．２５）が導かれる。 In S5, an analysis straight line corresponding to the optical path in the eyeball of the observation light beam is set based on the information set in S2 to S4. FIG. 14 shows the path of the observation light beam at the analysis target point set in the eyeball geometrically with reference to the eyepiece surface of the angle mirror in the observation with the angle mirror shown in FIG. The reason why the reference is the eyepiece of the angle mirror is that the observation light beam is refracted at this position. In FIG. 14, the eyepiece of the corner mirror corresponds to a semicircle GS, and two-dimensional coordinates with the center (Og) as the reference 0 are set. The two-dimensional coordinate Y-axis corresponds to the reference straight line BL in the cross-sectional image. The CS indicated by the dotted arc is the surface of the cornea, and the intersection (P) between the arc CS and the Y axis is the apex of the cornea. Here, if it is assumed that the apex P of the cornea is in contact with the eyepiece GS of the angle mirror, the center of the above-described two-dimensional coordinates is the vertex of the eyepiece GS of the angle mirror from the apex P of the cornea on the reference straight line BL. This corresponds to a position separated by a radius of curvature. Strictly speaking, when the radius of curvature Rc of the cornea surface is smaller than the radius of curvature Rg of the eyepiece surface GS of the angle mirror, a gap is generated in the vicinity of the eyepiece surface GS of the angle mirror and the apex P of the cornea. Since it is slight, the eyepiece surface GS of the corner mirror and the apex P of the cornea can be handled as being in contact with each other. For reference, a method of accurately setting the contact state of the angle mirror with the eyeball that changes depending on the relationship between CS and GS and the above-described reference 0 of the two-dimensional coordinates will be described with reference to FIG. FIG. 15 shows the contact relationship between the eyepiece surface of the cornea and the angle mirror according to the size of the radius GS of the GS and the radius Rc of CS, and the distance relationship between the center of curvature Og of the eyepiece surface GS of the angle mirror and the apex P of the cornea. It is shown. In addition, Oc described in each figure of FIG. 15 is the center of curvature of the surface of the cornea. FIG. 15A shows a case where Rg ≧ Rc. In this case, the angle mirror comes into contact with the eyeball at the apex P of the cornea. Therefore, the distance between the center of curvature Og of the eyepiece GS of the corner mirror and the apex P of the cornea is the radius Rg of the eyepiece GS of the corner mirror. FIG. 15B shows a case where Rg <Rc. In this case, the angle mirror contacts the eyeball not at the apex P of the cornea but at the annular ridgelines (PI1 and PI2 in this figure) of the eyepiece, and the apex P of the cornea and the eyepiece of the angle mirror A gap of ΔL is generated between the surfaces GS. Accordingly, the distance between the center of curvature Og of the eyepiece surface GS of the angle mirror and the apex P of the cornea is Rg−ΔL. FIG. 15C shows the relationship between points existing on the left side of the center line CL listed in FIG. Rg−ΔL is expressed by the following equation from the geometric relationship of each of a plurality of triangles related to the eyepiece surface GS of the angle mirror formed by these points and the surface CS of the cornea.

Therefore, by specifying the position of the apex P of the cornea and the curvature radius Rc of the surface of the cornea, the reference 0 (Og) can be accurately set. In the present embodiment, since the position of the apex P of the cornea and the radius of curvature of the cornea are specified in the process of setting the reference straight line in S2, it is not necessary to newly input by using the data. It becomes. When Rc = 7.7 mm, Rg = 7.4 mm, and Dg = 12 mm, ΔL is about 0.2 mm, and the distance from the apex P of the cornea to the reference of the two-dimensional coordinate is about 7.2 mm.
Point Q is the position of the eyepiece CS of the angle mirror that causes refraction in which the observation light beam passing through the point VP, which is the analysis target point, has the same inclination of θα as the straight line VB ′ in FIG. and has n _g, n _c denotes the main body and the refractive index of the cornea of the corner mirror. The value of the refractive index n _c of the cornea 1.376 are commonly used. In addition, since the refractive index of the tissue constituting the eyeball including tears, waterproofing, etc. is approximately 1.33 to 1.38, which is close to the refractive index of water, refraction at the tissue boundary in the eyeball does not occur in the present invention. It is handled as a thing. The setting of the straight line for analysis performed in this step is to specify the line segment AL between VP and Q, and this is achieved by determining the position of the point Q. When the straight line OQ connecting the reference 0 and the point Q has an angle θ relationship with respect to the X axis, the coordinates of the point Q are represented as (Rg × cos θ, Rg × sin θ). Therefore, if the coordinates of the analysis target point VP are (x, y), the inclination angle θβ of the line segment AL with respect to the X axis is

Further, at the point Q, between the incident angle θ _c of the light beam from the point P and the outgoing angle θ _g of the light beam toward the mirror of the angle mirror (not shown), Snell's law

The relationship is established. Here, when the incident angle θc and the outgoing angle θg are expressed by the aforementioned θ, θ _α , θ _β ,

Furthermore, since θ _α can be expressed by the mirror angle θ _M of the angle mirror as described above in the description of FIG. 5, the above Snell's law equation is

It can be expressed as. In the above equation, values other than the angle θ of the straight line OQ have already been set in S3 and S4 as analysis target points and corner mirror information, and therefore, a line segment AL that is an analysis straight line is derived by specifying θ. The coordinates (−5.84, 3.8) of the analysis target point VP, the assumed mirror angle θ _M = 62 ° of the angle mirror, the curvature radius Rg = 7.4 mm of the eyepiece surface, and the main body refractive index = 1.52. In some cases, θ = 78.5 ° is specified, so the coordinates (1.48, 7.25) of the point Q are derived.

  In S6, the analysis straight line AL set as described above is used to evaluate whether another tissue exists between the region where the analysis target point is set and the cornea 11 or not. FIG. 16 shows the image between the analysis target point of the analysis straight line AL set in S5 and the reference straight line, and FIG. 17 shows the luminance distribution of the pixels existing on the aforementioned line segment. It is. The set brightness in FIG. 17 is a threshold for identifying the boundary of the tissue in the captured image, and this value may be determined as appropriate based on the brightness distribution of the entire captured image. For example, a value that seems appropriate at 50 to 90% of the maximum luminance in the image may be selected, or a predetermined value may be determined in advance. 16 (a) and 17 (a) are examples of a state where the corner is normally opened, and FIGS. 16 (b) and 17 (b) are examples of a state where the opening of the corner is small. In FIG. 16A, the analysis straight line AL between the analysis target point SS and the reference straight line BL does not intersect the tissue boundary other than the back surface CP0 of the cornea 11, and therefore the analysis target point SS in FIG. There is no region exceeding the set luminance between the reference line BL and the vicinity of CP0. In FIG. 16B, since the iris 13 ′ exists between the analysis target point SS and the back surface CP0 of the cornea 11, the analysis straight line AL intersects at the front surfaces CP1 and CP2 of the iris 13 ′ in addition to the cornea 11, In 17 (b), it is extracted as an area exceeding the set luminance other than the vicinity of the reference straight line BL. Here, it is unlikely that a region other than the cornea exists in the region near the reference straight line BL in the analysis straight line AL, and the cornea 11 is optically almost transparent. From the luminance distribution excluding the analysis straight line AL (10% length) from the BL, the visibility of the analysis target point when the section from which the image is acquired is observed with the angle mirror is estimated. With the above procedure, whether or not an analysis target point can be observed with a corner mirror is analyzed for one cross-sectional image.

  FIG. 18 shows an example of displaying the result of the analysis of S6. In this example, the scleral cap SS is visually recognized with a reliability of 60% when the section from which the image to be analyzed is obtained is observed using the angle mirror of the specification shown in the second row. It is shown that it is possible. The reliability shown here evaluates the reliability of the analysis result using information such as the resolution (magnification) of the image used for the analysis. For example, when the resolution of the used image is high, the reliability is high, and when the resolution is low, the reliability is low. Alternatively, it may be calculated by comprehensively evaluating the set luminance value in the evaluation of the luminance distribution in S6, the length of a region that continuously exceeds the above-described set luminance value, and the like.

  FIG. 19 shows an anterior ocular segment image analysis apparatus according to the second embodiment of the present invention.

  An anterior ocular segment image analyzing apparatus 200 includes a control / analyzing unit 201 for performing analysis, an image data acquiring apparatus 205 for acquiring an image captured by another anterior segment cross-sectional image capturing apparatus 220, information necessary for analysis, and the like. A mouse 206 and a keyboard 207 for inputting, a display 208 for displaying cross-sectional images, analysis results, and the like, a printer 210, and the like are included. The control / analysis unit 201 includes a control circuit that controls the entire apparatus, a calculation circuit that performs analysis processing, and a storage unit that stores analysis programs, image data, calculation results, and the like. The control / analysis unit 201 can be configured using a commercially available personal computer. Commercially available products can be used for other components. Image data photographed by the anterior segment cross-sectional image photographing device 220 is provided by an external storage medium or the like, obtained from the image data obtaining device 205, and read into a storage unit in the control / analysis unit 201. If the anterior segment cross-sectional image capturing device 220 has a communication function, it is directly read into the storage unit in the control / analysis unit 201 without using the image data acquisition device 205 via a communication cable (not shown). It is also possible. The control / analysis unit 201 displays the image data read into the storage unit on the display 208. If a plurality of image data can be read into the storage unit, a list of image data may be displayed on the display 208, and the image data to be analyzed may be selected by the mouse 206 or the keyboard 207.

  Hereinafter, the present embodiment will be described based on the flow of FIG. 10 as in the first embodiment. The image data that can be used in the present invention is not limited to only image data obtained by photographing a cross section passing through the apex of the cornea shown in the first embodiment. For example, as shown in FIG. 7, if the position of the apex of the cornea is specified with respect to three-dimensional image data constructed from image data continuously taken by moving the imaging section in parallel, an arbitrary position can be obtained. A cross-sectional image can be acquired. In the present embodiment, it is assumed that a cross-sectional image capable of constructing three-dimensional image data, or already constructed three-dimensional image data is handled. The construction of three-dimensional image data based on a plurality of cross-sectional images photographed as shown in FIG. 7 is an example used in the image display of a fundus tomography apparatus using optical interference in Patent Document 2. Although detailed description of the procedure is omitted, it can be performed using a general image processing technique.

  FIG. 20 acquires a cross-sectional image in an arbitrary direction using three-dimensional image data constructed from a plurality of images acquired by an imaging mechanism generally employed in a tomographic apparatus using optical interference. An outline of the procedure is shown. FIG. 20A shows the positional relationship of the imaging plane of the image with respect to the eyeball viewed from the front. In the example shown here, shooting is performed on each of the shooting planes that are parallel to each other and set for each interval Δy, and L shooting planes are set from the bottom to the top shown as the Y direction. Here, Δy varies depending on the specifications of the photographing apparatus. As Δy is smaller, a higher-resolution image can be photographed, and the photographing range is determined by the number of L. In addition, photographing on each photographing surface is performed from the left to the right indicated as the X direction. FIG. 20B shows an example in which the measurement position of the interference signal that is the source of the image acquired on each imaging plane is acquired as row and column information, with the X direction assigned to the column and the Z direction assigned to the row. The image shown here is composed of M × N interference signals. The distances Δx and Δz between the measurement points vary depending on the specifications of the photographing apparatus as in the above-described Δy, and the photographing range is determined by the number of M or N. Here, the X direction is the same as that in FIG. 20A, and the Z direction corresponds to the depth direction of the paper surface in FIG. Interference signals are acquired in the X direction (left to right in the figure) for each column on each imaging plane, and when the interference signals are acquired for all columns, imaging on that imaging plane is completed. Accordingly, when imaging is completed up to the imaging surface L shown in FIG. 20A, L × M × N interference signals distributed three-dimensionally are acquired.

  From here, selection of the analysis image of S1 in this embodiment is described. A storage unit in the control / analysis unit 201 of the anterior segment image analysis apparatus 200 reads and analyzes a plurality of cross-sectional images from the image data acquisition apparatus 205. Note that the above-described three-dimensional image data can acquire a cross-sectional image at an arbitrary position (described later), and can be handled as corresponding to a plurality of cross-sectional images. The read image data is displayed on the display 208, and the operator selects an image to be analyzed from the images displayed by the mouse 206 or the like. When the displayed image data is three-dimensional image data, the cross-sectional image specified by the input information specifying the corneal apex and the cross-sectional direction is selected. The control / analysis unit 201 constructs a cross-sectional image including the corneal apex based on the input information.

  FIG. 21 illustrates a procedure for acquiring a cross-sectional image of the cross-section CP set based on the position of the apex P of the cornea and the information on the tilt angle with respect to the imaging surface shown in FIG. In the following description, shooting is performed on the shooting planes 1 to L shown in FIG. 20A in a state where the positional relationship between the apparatus and the eyeball does not change (no movement of the eyeball), and each shot image is shot. The criteria shall be the same.

FIG. 21A shows the relationship between the cross section CP set with the angle θ _γ inclined with respect to the imaging surface and the image signal of each imaging surface. In this example, an image photographed on the i-th photographing surface is specified as an image including the apex P of the cornea. The position in the X direction of the apex P of the cornea in the image specified here is specified as the row position in the X direction shown in FIG. Here, description will be made assuming that the p-th column indicated as Col-p in FIG. 20B is specified as the X-direction position of the apex P of the cornea. CrL _i , CrL _j , and CrL _k indicate the intersection positions of the cross section CP and the imaging surfaces i, j, and k.

しかしながら、上述のように導かれた交差位置の情報は各撮影面における画像についてのものであり、このままでは断面ＣＰの画像を取得できない。ここで、断面ＣＰにおける交差位置ＣｒＬ_i、ＣｒＬ_j、ＣｒＬ_kについて、ＣｒＬ_iを基準とする距離をそれぞれＬ_ij、Ｌ_ikと定義すると、図２１（ａ）から幾何学的にＸＬ_ij及びＸＬ_ikと以下の関係になることが導かれる。

これは、ＣｒＬ_iを基準とする各撮影面のＸ方向の距離は１／cosθ_γ倍で断面ＣＰへ投影されることを意味している。図２１（ｅ）は、断面ＣＰ上の交差位置ＣｒＬ_i、ＣｒＬ_j、ＣｒＬ_kの関係を示したものである。ここで、交差位置ＣｒＬ_i、ＣｒＬ_j、ＣｒＬ_kにはそれぞれ撮影面ｉの第ｐ列、撮影面ｊの第ｑ列、撮影面ｋの第ｒ列の画像情報が反映されている。従って、撮影面１から撮影面Ｎについて前述の操作を行なうことにより、図２１（ｅ）に点線で示される前眼部断面の画像が取得でき、Ｓ１における解析画像の選択が完了する。この後、ディスプレイ２０８の表示範囲に合わせて倍率変換された画像が表示される。 FIGS. 21B to 21D show the positions of the aforementioned intersection positions CrL _i , CrL _j , and CrL _k in the cross-sectional images photographed on the photographing surfaces i, j, and k. On the left and bottom outside the frame shown as the shooting range, the horizontal and vertical positions are indicated by the rows and columns in FIG. Here, the positions of CrL _i , CrL _j , and CrL _{k in} the X direction are the p-th column, the q-th column, and the r-th column, respectively. Moreover, XL _ij and XL _ik in FIGS. 21C and 21D define distances in the X direction with reference to CrL _i . This XL _ij or XL _ik can be expressed as follows by the above-mentioned columns.

However, the information on the intersection position derived as described above is about the image on each photographing plane, and the image of the cross section CP cannot be acquired as it is. Here, regarding the intersection positions CrL _i , CrL _j , and CrL _k in the cross section CP, if distances based on CrL _i are defined as L _ij and L _ik , respectively, XL _ij and XL are geometrically shown in FIG. _The following relationship is derived with _ik .

This distance in the X direction of each imaging surface relative to the CrL _i are meant to be projected into the cross section CP at times 1 / cosθ _γ. FIG. 21 (e) shows the relationship between the intersection positions CrL _i , CrL _j , and CrL _k on the cross section CP. Here, the intersection positions CrL _i , CrL _j , and CrL _k reflect the image information of the p-th column of the imaging plane i, the q-th column of the imaging plane j, and the r-th column of the imaging plane k, respectively. Therefore, by performing the above-described operation on the imaging surface 1 to the imaging surface N, an image of the anterior segment cross section indicated by the dotted line in FIG. 21E can be acquired, and the selection of the analysis image in S1 is completed. Thereafter, an image whose magnification has been converted in accordance with the display range of the display 208 is displayed.

  FIG. 22 shows an example in which the reference straight line in S2 is set not based on the apex of the cornea but on the scleral cape existing in the corner area. This is based on the premise that a cross-sectional image including the apex of the cornea as a target in the present invention is acquired at a position corresponding to the diameter of corners distributed in an annular shape. As shown in FIG. 22 (a), when two scleral capes SS1 and SS2 are recognized in the same cross-sectional image, the length of the segment SS1-SS2 is the diameter of the scleral cape distributed in an annular shape, The midpoint MP of SS1 and SS2 corresponds to the center of the distribution circle of the sclera cape. Therefore, by setting the perpendicular bisector that can be specified only by the position information of SS1 and SS2 in the straight line passing through the midpoint MP of SS1 and SS2 as the reference straight line BL, the corner angle including the scleral cap SS It is possible to easily set a reference reflecting the area. Here, the setting of SS1 and SS2 is performed by designating an arbitrary position in the display image, similarly to the setting of the apex P of the cornea in the first embodiment. FIG. 22B shows the relationship between the set reference straight line BL and the vertical line of the display screen. Similarly to the first embodiment, the eyeball image that can be used to unify the display direction by rotating the image or the like. Tilt information can be acquired. The midpoint MP can be used as the image rotation reference.

  Subsequently, the analysis target point in S3 is set. In this setting, it goes without saying that the display image is designated as in the first embodiment, and the information of SS1 and SS2 set in S2 may be used as they are.

  The setting of the angle mirror information performed in S4 is performed by displaying a list of specifications of the angle mirror as shown in FIG. 23 on the display 208 and designating the angle mirror information assumed by the operator. It is possible to prevent erroneous input. The designation is No. in the list. This is done by matching the input to the line where the information of the angle mirror for selecting the character figure that can be moved to an arbitrary position of the list display in conjunction with the input from the mouse 206 or the keyboard 207 is displayed. The angle mirror information can be registered in advance in a storage unit included in the control / analysis unit 201, or registered information can be read using the image data acquisition device 205.

  Based on the information set in S1 to S4, the analysis straight line AL is set by performing the process of S5 as in the first embodiment. In the case of performing processing based on the two-dimensional coordinate reference in consideration of the circumstances shown in FIG. 15, it is necessary to set the position of the vertex P of the cornea and the curvature of the surface of the cornea in the image. Unlike the first embodiment in which the above-described information is acquired in the setting, in the present embodiment that has not been set up to S4, the setting is performed at the beginning of this step. As in the procedure for setting the reference straight line BL in the first embodiment, the setting of the position of the apex P of the cornea in the image is manually performed, and the curvature of the cornea surface is automatically calculated using the luminance distribution. To do. In the present embodiment, since the reference line BL that has already been specified can be used, the apex P of the cornea is displayed on the screen as a manual setting guide, or the brightness of pixels existing on the reference line BL When the boundary of the cornea surface is identified based on the distribution, the setting can be automatically performed.

  Subsequently, the visibility analysis of S6 is performed. In the first embodiment, an example is shown in which visibility is evaluated based on the luminance distribution of pixels located on the analysis straight line AL. In this embodiment, the front surface of the iris that directly affects the visibility of the analysis target point. The boundary position is obtained as an approximate curve from the luminance information of the image, and the visibility is evaluated based on the relationship with the aforementioned analysis straight line. This procedure will be described with reference to FIG. 24 and FIG.

  FIG. 24 shows a state before and after specifying the boundary position of the front surface of the iris on the display screen. FIG. 24A shows a state in which the range for specifying the boundary is set as IFL and IFR. This setting is performed by designating an arbitrary position in the image in the same manner as S2 or S3 described above. The boundary position of the iris front surface existing between the set two points is specified based on the luminance information, and an approximate curve is obtained from the specified boundary information. FIG. 24B shows the relationship between the analytical straight line AL obtained in S5 and the approximate curve AC obtained from the luminance information. The visibility of the analysis target point is evaluated by the presence / absence of an intersection between the analysis straight line AL and the approximate curve AC.

  FIG. 25 illustrates a procedure for identifying the front boundary of the iris based on the luminance information between two points of the IFL and IFR shown in FIG. FIG. 25A shows the luminance information of the display screen in the range shown as region E in FIG. Here, a pixel displayed with a circle has a luminance equal to or higher than a predetermined value, and a pixel displayed with a black circle indicates that the luminance is lower than a predetermined value. Note that the lower left pixel in this figure corresponds to the IFL in FIG. FIG. 25B shows an example in which pixels from which luminance information is acquired in specifying the boundary position are grouped. In this example, seven pixels that are consecutive in column units are grouped. The number of pixels of 7 is an example and is not limited to this. Here, a central pixel surrounded by a dotted circle shown as a reference pixel is a pixel that becomes a reference when a column for acquiring luminance information is changed, and its role will be described in the following description.

  FIG. 25C shows a method for specifying a pixel located on the front boundary of the iris based on the luminance information acquired by the pixel group shown in FIG. Here, a state is shown in which the boundary position is specified while moving the column in the right direction from the leftmost column, which is the n column. In addition, (circle) represents the pixel which has a brightness | luminance more than a predetermined value, and ● represents the pixel of a brightness | luminance below a predetermined value. First, luminance information is acquired for pixels in ± 3 rows centering on a reference pixel located in the M row in the n columns, and the pixel located at the end of the pixels having a luminance equal to or higher than a predetermined value is specified. However, if the luminance of all the pixels is equal to or higher than a predetermined value or lower than the predetermined value, the boundary cannot be specified, and the evaluation of the column is terminated and the next column is moved. Since all the pixels in the n-th column have a luminance equal to or higher than a predetermined value, the boundary is not specified and the next n + 1 column is moved. At this time, since the boundary position is not specified, the row position of the reference pixel is set to the same M row as the previous n columns. Since the luminance of the n + 1 column has pixels with luminance equal to or lower than a predetermined value up to M + 1 rows, the M + 2 rows of pixels surrounded by dotted squares are specified as boundary positions and stored in the storage unit of the control / analysis unit 201. Store row and column information. Subsequently, the luminance information is acquired in the same manner by moving to the n + 2 column, but the position of the reference pixel is set to the M + 2 row specified as the boundary position in the previous column. In the n + 2 column, the boundary position is specified as a pixel of M rows. Therefore, the reference pixel in the n + 3 column is set to M rows, and the boundary specified by the luminance information is the pixel of M-2 rows. FIG. 25D shows the result when the above-described procedure is performed for the region F in FIG. 25A with the IFL as the start position. A dotted circle is a reference pixel of each column, and ◎ is a pixel specified as a boundary. By repeating this until the pixel set as the IFR in FIG. 24A is reached, the boundary of the front surface of the iris from the IFL to the IFR is specified. Thereafter, it is possible to obtain an approximate curve based on the row and column information of the specified boundary pixel. The order of the approximate curve may be set in advance, or the order in which the positions of all the specified boundary pixels fall within a predetermined error range may be determined by sequentially increasing from the lower order. Visibility can be evaluated by determining the presence or absence of the intersection of the specified approximate straight line AL and the analytical straight line AL.

  Note that by setting a plurality of analysis target points and repeatedly performing S4 to S6 on each analysis target point, it is also possible to evaluate the visibility of different parts in the image at a time. FIG. 26 shows such an example. FIG. 26 (a) shows a corner angle that is normally opened, and FIG. 26 (b) shows an image of the scleral cap SS and Schwalbe line SL in a corner image with a small opening. A point is set. In FIG. 26 (a), it is evaluated that both the scleral cap SS and the Schwalbe line SL can be observed, but in FIG. 26 (b), only the Schwalbe line SL can be observed and the scleral cap SS cannot be observed. Be evaluated. Here, in addition to these two points, analysis target points may also be set in the trabecular meshwork TM between the scleral cap SS and the Schwalbe line SL.

  When there are a plurality of images related to different cross sections of the same eyeball, and it is possible to associate the position information of each cross section with the analysis result, it is possible to acquire more useful information in the diagnosis of glaucoma. FIG. 27 shows an example in which the results of analysis of the analysis target points set in each of the cross-sectional images moved at predetermined angles with the axis passing through the apex of the cornea as the rotation axis are displayed in a listable manner. In this example, the scleral cap SS, Schwalbe line SL set as analysis target points in the images acquired every angle of 5 °, and the visibility of one point of trabecular meshwork TM located between the two points described above, and The reliability is displayed. The analysis target point determined to be observable is displayed as ◯, and the analysis target point determined as unobservable is displayed as ×. Further, as shown in FIG. 28, it is also possible to obtain the ratio of the number that can be observed to the number of analysis target points that have been analyzed. FIG. 28 displays the result of obtaining the observability rate for each part where the observation target point is set for the same data as FIG. The observable rate of the fiber columnar band TM obtained here is equivalent to the narrow-corner angle evaluation parameter defined in Non-Patent Document 1, and when this value is 25% or less, the narrow-corner angle is suspected. It is possible to infer that there is. Furthermore, FIG. 29 is an example showing the analysis result graphically. For the same data as FIG. 27, FIG. 29 (a) shows the scleral cap SS, FIG. 29 (b) shows the scleral cap SS, the Schwalbe line SL, The possibility of observation acquired for the fiber column band TM is plotted in correspondence with the acquisition position of the cross-sectional image. By performing such display, it is possible to grasp the visibility of the corner area distributed in an annular shape in a two-dimensional manner.

  As mentioned above, although each embodiment of the present invention has been described in detail, the present invention is not limited in any way by the specific description in the embodiment, and various changes, modifications, and modifications based on the knowledge of those skilled in the art. Needless to say, the present invention can be implemented in a mode with improvements and the like, and all such modes are included in the scope of the present invention without departing from the gist of the present invention.

It is the figure which showed the anterior eye part area | region of the eyeball. It is the figure which showed the corner area | region of the anterior eye part. It is the figure which showed the characteristic site | part of the corner area | region. It is the figure which showed the structure of the angle mirror (indirect inspection type). It is the figure which showed the relationship between the eyeball at the time of a corner mirror inspection, and a corner mirror. It is the figure which showed the cross-sectional image at the time of rotating an imaging position. It is the figure which showed the cross-sectional image at the time of parallel-moving an imaging position. It is the figure which showed the external appearance of the eyeball imaging device concerning this invention. It is the figure which showed the block diagram of the eyeball imaging device concerning this invention. It is the figure which showed the processing flow concerning this invention. It is the figure which showed an example of the setting method of a reference | standard straight line. It is the figure which showed an example of the setting method of an analysis object point. It is the figure which showed an example of the input screen of corner mirror information. It is the figure which showed the influence by the inclination of a cross-sectional image. It is the figure which showed the state by which the straight line for analysis was set to the cross-sectional image. It is the figure which showed the luminance distribution on the straight line for analysis. It is the figure which showed the state by which the some analysis straight line was set to the cross-sectional image. It is the example which showed the analysis result by the numerical value. It is the figure which showed another structure of the apparatus concerning this invention. It is the figure which showed the example of the acquisition method of the image data generally employ | adopted in the cross-sectional image imaging device using optical interference. It is the figure which showed the example which acquires the cross-sectional image of arbitrary positions from the several cross-sectional image which constructs | assembles a three-dimensional image. It is the figure which showed the other example of the setting method of a reference | standard straight line. It is the figure which showed the other example of the input screen of angle mirror information. It is the figure which showed the state before and behind specifying the boundary of the structure | tissue contained in a display image. It is the figure which showed the example which specifies the boundary of the structure | tissue contained in a display image based on luminance information. It is the figure which showed the straight line for analysis of the cross-sectional image in which the some analysis object point was set. It is the other example which showed the analysis result by the numerical value. It is the other example which showed the analysis result by the numerical value. It is the example which showed the analysis result by the graphic.

Explanation of symbols

10: eyeball 11: cornea 12: sclera 13, 13 ′: iris 14, 14 ′: crystalline lens 21: housing 22: mirror 100: eyeball imaging device 101: control circuit 102: arithmetic circuit 103: memory 104, 104 ′: Probe 105: Connector 106: Rotation knob 107: Touch panel 108: Liquid crystal display 109: External storage means 110: Printer 200: Anterior eye image analysis device 201: Control / arithmetic circuit 206: Mouse 207: Keyboard 208: Display 210: Printer 220: Eyeball photographing device SS, SS1, SS2: Cape sclera TM: Trabecular belt SL: Schwalbe line CL: Corner mirror central axis Pv: Angle mirror observation plane θ _M : Angle mirror mirror angle VB, VB ′, VB1, VBn: Observation beam BL: (Analysis) reference straight line AL: Analysis straight line AC: Near (front of iris) Curve

Claims (7)

In the analysis method of the cross-sectional image of the anterior segment including the corneal vertex and the corner area of the eye to be examined,
Selecting a cross-sectional image to be analyzed;
Setting information of a cross-sectional position of the selected cross-sectional image;
Setting a reference straight line in the cross-sectional image;
Setting an analysis target point in a characteristic part of the corner region;
Setting angle mirror information for the reference straight line;
Setting a straight line passing through the analysis target point based on the reference straight line and the angle mirror information as an analysis straight line;
And a step of analyzing whether or not a predetermined tissue exists on the analysis straight line based on the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis straight line. An anterior segment cross-sectional image analysis method.   2. The method according to claim 1, further comprising: associating the information on the cross-sectional position with the analysis result acquired in the step of performing the analysis, and classifying the analysis result according to a part where the analysis target point is set. Analysis method for cross-sectional images of the anterior segment of the eye. In an eyeball imaging apparatus having an imaging means capable of acquiring a cross-sectional image of an anterior ocular segment including a corneal vertex and a corner area of a subject eye,
Cross-sectional image selection means for selecting a cross-sectional image to be analyzed;
Display means for displaying the cross-sectional image;
Cross-sectional position information setting means for setting information on the cross-sectional position of the cross-sectional image displayed on the display means;
Reference line setting means for setting a reference line on the cross-sectional image displayed on the display means;
Analysis target point setting means for setting an analysis target point in a characteristic part of the corner area;
Angle mirror information setting means for setting angle mirror information for the reference straight line;
An analysis line setting means for setting, as an analysis line, a straight line passing through the analysis target point, the angle intersecting with the reference line is determined based on the angle mirror information;
Characterized by comprising analysis means for analyzing whether or not a predetermined tissue exists on the analysis straight line based on the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis straight line. Eyeball photographing device.   4. The method according to claim 3, further comprising analysis result classification means for associating the information on the cross-sectional position with the analysis result acquired by the analysis means, and for classifying the analysis result according to a region where an analysis target point is set. The eyeball photographing apparatus described.   5. The eyeball photographing apparatus according to claim 4, wherein the analysis result classified by the analysis result classification means is displayed on the display means by numerical value or graphic for each analysis target point. In a computer-processable recording medium in which a program for analyzing a cross-sectional image of the anterior segment including the corneal apex and corner area of the eye to be examined is recorded,
Selecting a cross-sectional image to be analyzed;
Setting information of a cross-sectional position of the selected cross-sectional image;
Setting a reference straight line in the cross-sectional image;
Setting an analysis target point in a characteristic part of the corner region;
Setting angle mirror information for the reference straight line;
Setting a straight line passing through the analysis target point, which is determined based on the angle mirror information, an angle intersecting the reference straight line as an analysis straight line;
Performing a step of analyzing whether or not a predetermined tissue exists on the analysis line based on the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis line. Recording media that can be processed by a computer. In a computer-processable program for analyzing a cross-sectional image of the anterior segment containing the corneal apex and corner area of the eye to be examined,
Selecting a cross-sectional image to be analyzed;
Setting information of a cross-sectional position of the selected cross-sectional image;
Setting a reference straight line in the cross-sectional image;
Setting an analysis target point in a characteristic part of the corner region;
Setting angle mirror information for the reference straight line;
Setting a straight line passing through the analysis target point, which is determined based on the angle mirror information, an angle intersecting the reference straight line as an analysis straight line;
Performing a step of analyzing whether or not a predetermined tissue exists on the analysis line based on the position or luminance information of the pixels constituting the cross-sectional image existing on the analysis line. A computer processable program.

Images