An eyeground observation device, an ophthalmologic image processing unit, an ophthalmologic image processing program and an ophthalmologic image processing method

Abstract

First image forming part forms a two-dimensional surface image of a fundus oculi of an eye based on optically obtained data. Second image forming part forms tomographic images of fundus oculi based on data obtained by optically scanning a region of the surface of fundus oculi corresponding to at least part of two-dimensional image. Accumulated image generating part generates an accumulated image by accumulating the formed tomographic images in a depth-wise direction. Extracting part extracts first vascular territory corresponding to a fundus oculi vessel from two-dimensional image formed by first image forming part, and also extracts second vascular territory corresponding to a fundus oculi vessel from accumulated image generated by accumulated image generating part. Specification part specifies a position of a vascular cross sectional region corresponding to a cross section of a fundus oculi vessel in the tomographic image based on extracted first vascular territory and extracted second vascular territory.

Description

technical field [0001] The present invention relates to a fundus observation device, an ophthalmic image processing device, an ophthalmic image processing program, and an ophthalmic image processing method for observing the state of the fundus of an eye to be inspected. Background technique [0002] As a fundus observation device, a fundus camera has conventionally been widely used. FIG. 19 shows an example of an external structure of a conventional general fundus camera, and FIG. 20 shows an example of an optical system built therein (for example, refer to Japanese Patent Application Laid-Open No. 2004-350849). In addition, the term "observation" includes at least observation of a captured image of the fundus (in addition, observation of the fundus with the naked eye may also be included). [0003] First, the appearance structure of a conventional fundus camera 1000 will be described with reference to FIG. 19 . The fundus camera 1000 includes a gantry 3 mounted on a base ...

AI Technical Summary

Problems solved by technology

[0036] When a conventional optical image measuring device is used to obtain a tomographic image of the fundus, the image area directly below the fundus blood vessels (the position in the +z direction shown in FIG. The state of the layer boundary cannot be seen with the eyes

[0037] For example, as shown in FIG. 21, when the tomogram G' or layer (in the image) L1' of the fundus is taken from the fundus blood vessel (sectional image) V, due to the influence of the fundus blood vessel V, the directly below The image area V' becomes unclear, and it is impossible to grasp the state of the boundary g2' between the layer L1' and the layer L2', the state of the boundary g3' between the layer L2' and the layer L3', the state of the layer L3' and its lower layer (not shown). ) state of the boundary g4'

Therefore, the thicknesses of the layers L1', L2', L3', etc. in the forward direction of the fundus blood vessel V cannot be measured.

Therefore, the layer thickness at this position will be recorded as "0" or "unable to measure", which leads to the problem that the layer thickness of the entire acquired image cannot be measured

[0038] When the user visually recognizes the tomographic image G' to explore the image region V', it must take a considerable amount of time and labor, which is quite difficult in practical applications.

In addition, since it is difficult to analyze the tomographic image G' and specify the fundus blood vessel V, it is difficult to automatically extract the image region V' from the tomographic image G'. Generally speaking, the actual blood vessel cannot be clearly specified in this way.) In addition, the symbol LS in FIG.

Images