RU2193337C2 - Method for studying eyeball movements - Google Patents

Abstract

FIELD: medicine. SUBSTANCE: method involves building video image on computer display screen, transforming every recorded frame into binary form, analyzing eye motion characteristic parameters. An area is selected on the video image within which the eye pupil moves. A figure is found in this area which dimensions and shape match those of the eye pupil. The point placed as far as possible from the figure boundary is selected as its center. To exclude flashes when producing images, video pictures are taken under additional uniform illumination of the eye from all sides with semitransparent tube set between the eye pupil and video camera. Figure center movement spectrum harmonics amplitudes and frequencies are taken as the motion parameters usable for determining nystagmus pattern. EFFECT: high reliability of examination results. 10 dwg

Description

 The invention relates to the field of medicine and biophysics, can be used in ophthalmology for the diagnosis of nystagmus and saccades.

 There is a method of ultrasonic examination of the eye by ultrasonic scanning, in which a series of parallel two-chamber scans that are not graded by brightness are summed up electronically and due to the merging of the matching parts, a brightness gradation is used that is used as the third coordinate when constructing a three-dimensional image of the acoustic section of the eye (see a S.S. USSR 9252245, class A 61 F 9/00, 1980).

 However, this method does not allow to obtain information about the characteristics of the movement of the eyeball.

 A known method of ultrasonic examination of the eye by constructing a three-dimensional acoustic image after electronic summation of a series of consecutive acoustic sections (see RF patent 2128473, class A 61 B 8/10, 1999).

 The disadvantage of this method is the lack of accuracy of the measurements.

 The closest in technical essence to the proposed one is the method of experiments in the field of teleoculometry when turning the head, including receiving video images on a computer screen, binarizing each recorded frame, analyzing the parameters of eye movement, and the eye position measuring channel is made in the form of an optical-electronic converter based on a video camera with a CCD matrix fixedly fixed relative to the eye socket, and the channel for measuring the angle of rotation of the head is in the form of a rotary electromechanical about a measuring transducer with pulse-code modulation (see. Nystagmoscope. // Russian scientific-practical conference "OPTICS AND SCIENTIFIC INSTRUMENT-2000" FTP "Integration". March 29-30, 2000 / Abstract. - St. Petersburg: ITMO, 2000 - 72 p.).

 However, this method is quite laborious to implement from a technical point of view.

 The objective of this method is to reduce the complexity of the study while increasing the reliability (information content) due to the greater number of investigated parameters.

 The problem is achieved in that in the method for studying the movement of the eyeball, including receiving video images on a computer screen, binarizing each recorded frame, analyzing the parameters of eye movement, select the area within which the eye moves, and in this area find a figure with the size and the shape corresponding to the size and shape of the pupil of the eye, for its center take the point farthest from the border of the figure, while to exclude glare when receiving an image, video recording is carried out at additional uniform illumination of the eye from all sides and a translucent tube installed between the eye and the camera, the amplitudes and frequencies of the spectral harmonics of the center of the figure, which determine the nature of the nystagmus, are selected as the motion parameters.

 The originality of the proposed solution lies in the use of a new technique for eliminating glare on the image of the pupil of the eye, the method of tracking its center and determining the parameters of eye movement. A similar set of actions to determine the nature of nystagmus is not known.

 The proposed method is illustrated by drawings.

 Figure 1. Installation diagram: 1 - patient's eye, 2 - tube, 3 - scattered-light annular lamp, 4 - video camera, 5 - analog-to-digital converter (video capture card), 6 - computer, 7 - storage medium that provides data storage.

 Video information processing scheme: FIG. 2 - capture of a video image and selection of an analyzed area, FIG. 3 — image binarization, FIG. 4 — finding the center of the pupil of the eye, FIG. 5 — outputting a motion path on a plane (top left), horizontal movement pattern ( right) and vertically (bottom).

 FIG. 6. The dependence of the displacement of the pupil of the eye along the horizontal axis with the pendulum-like form of nystagmus (period 0.32-0.4 s) with a characteristic saccadic movement (sharp decline lasting 0.16 s).

 FIG. 7. The trajectory of the pupil of the eye with rotatory jerky nystagmus.

 FIG. 8. The dependence of the displacement of the pupil of the eye along the horizontal axis on time.

 FIG. 9. The dependence of the displacement of the pupil of the eye along the vertical axis on time.

 FIG. 10. The range of motion of the pupil of the eye along the horizontal axis with rotatory jerky nystagmus.

 The method is as follows.

 Using, for example, an annular scattered light lamp 3, the patient’s eyeball 1 is uniformly illuminated, and a tube 2 is used to eliminate glare on the pupil. Video camera 4 records the displacement of eye 1 in the orbit. The video image of the moving pupil is introduced into the computer 6 using the analog-to-digital Converter 5 and analyzed using a computer program. The program carries out frame-by-frame threshold processing of the video file and recording the trajectory of the center of the pupil of the eye, as well as its mathematical processing (spectral conversion). The size of the treated area is selected so that it contains the pupil in all its positions during the entire observation time, since the presence in the selected area of foreign objects that are comparable or large in size of the pupil of the eye will lead to inadequate program operation. The program recognizes the center of a figure having a size and shape corresponding to the size and shape of the pupil of the eye. Its center is the point farthest from the boundary of the object.

 In FIG. Figure 2-5 shows a diagram of the analysis of the selected area of the video image. After entering data into the computer from the source of the video information (video camera, VCR) and starting the processing program, the video image is displayed in the window (figure 2). To start the analysis, an analysis area is selected on the image, covering a sufficient area so that during movement the object does not go beyond the boundaries of the selected area (figure 2). The threshold level for binarization (grayscale image conversion to reduce information redundancy) of the image is selected so that the image remains a clear silhouette of the studied object (figure 3). Then there is the geometric center of the pupil of the eye, and its coordinates are recorded in an array (figure 4). Upon completion of the analysis, the screen displays the form of movement vertically (bottom), horizontally (right) and the trajectory on the plane (bottom left) (figure 5). Using the fast Fourier transform technique, the spectrum of the eyeball movement along the horizontal and vertical axes of the eye is calculated. The magnitude of the amplitudes and frequencies of the spectral harmonics of eye movement determines the nature of nystagmus.

 An example of a practical implementation of the method.

 The structure of the installation (figure 1) includes a video camera 4, which records the displacement of the eye 1 in the orbit. At the same time, the patient’s head and video camera are fixed in a special holder to ensure the video camera is stationary relative to the eye socket. Uniform illumination of the eye is provided, for example, by means of an annular lamp of diffused light 3 with the use of a tube 2, since the presence of glare on the pupil displaces the recognized center from the center of the real one. The video image of the moving pupil is entered into the computer 6 using an analog-to-digital converter 5 and analyzed using a program that implements an object recognition algorithm and provides a dialogue with the user. For image input, the Matrox Marvell G400 video capture card is used, which supports full-format video input at a speed of 25 frames per second, and the software (software) supplied with it for capturing video in an AVI file.

 The software part of the PAC includes a specialized program operating in the Windows'98 environment and performing analysis of the input image. Using the software part of the PAC, the position of the center of the pupil of the eye is determined and the recording and spectral analysis of the trajectory of its movement are carried out. The results of the analysis are stored on the storage medium 7.

 The program allows you to analyze the video image of a moving object entered into the computer or part of it: it opens an AVI file, allows you to select the area to be processed and the processing threshold, conducts frame-by-frame threshold processing of the video file and records the trajectory of the center of the object, as well as its mathematical processing (spectral conversion). The size of the treated area is chosen so that it contains the pupil of the eye in all its positions during the entire observation time. It should be borne in mind that an unreasonable increase in the size of the region leads to a significant increase in processing time.

 The program recognizes the center of a figure having a size and shape corresponding to the size and shape of the pupil of the eye. Its center is the point farthest from the boundary of the object.

 Surveys were conducted in patients with severe nystagmus. Patient P., 12 years old, had jerky nystagmus with marked periods of pendulum nystagmus. Figure 6 presents the results of measuring the displacement of the pupil for a time of ~ 2 seconds. As can be seen from the presented figure, there are periodic deviations of the pupil of the eye by a value of 1.2 to 1.6 mm. The period of the pendulum-like nystagmus was 0.32-0.4 s. Observed in FIG. 6, a sharp decline lasting 0.16 s (the pupil displacement is 2.8 mm) corresponds to the saccadic form of movement, which is usually responsible for changing the direction of view. After changing the direction of the gaze (time interval 0.9-1.8 s in Fig. 6), pupil drift is observed, associated with the slow gliding of the person’s gaze from the initial direction and shown in the figure in that each subsequent maximum with the pendulum-like form of pupil motion is higher (or lower ) of the previous one.

 Patient M., 12 years old, had rotatory jerky nystagmus. In FIG. 4 shows a trajectory describing the rotational nature of the eye movement with a center shift, and the amplitude of the pupil along the horizontal (Fig. 8) and vertical (Fig. 9) axes is different.

 The greatest displacement is the movement of the pupil of the eye along the horizontal axis. A spectral representation of eye movement along the horizontal axis is shown in Fig.10. As follows from the results presented in figure 10, in the spectrum of the pupil of the eye there is a harmonic at a frequency of 1.85 Hz with an amplitude exceeding the other harmonics. A period of movement corresponding to this frequency, equal to 0.54 s, characterizes a jerky nystagmus.

 Thus, the described technique allows you to record the movement of the eyeball and determine the nature of nystagmus.

Claims (1)

 A method for studying the movement of the eyeball, including receiving a video image on a computer screen, binarizing each recorded frame of the video image, followed by analyzing the parameters of eye movement, characterized in that before or after binarization on the video image, the region within which the pupil of the eye moves is selected, and in this region find a figure with a size and shape corresponding to the size and shape of the pupil of the eye, the analysis of motion parameters is carried out by analyzing the movement of the center of this figure, for which the notes They take a look at the trajectory of its motion, measure the amplitude and frequency of the spectral components of the motion, which are used to judge the nature of the motion, while the center is taken as the point furthest from the border of the figure, and to exclude glare when receiving an image, video is shot with additional uniform illumination of the eye from all sides installed between the pupil of the eye and the video camera translucent tube.

Images