Method for estimating the shape of internal skin fibers
By using facial imaging parameters to estimate internal skin fiber shape, the method simplifies the evaluation process, facilitating frequent monitoring and accurate assessment of skin fiber changes.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- FUAN KERU
- Filing Date
- 2024-12-06
- Publication Date
- 2026-06-18
Smart Images
Figure 2026098936000001_ABST
Abstract
Description
Technical Field
[0005] , , , , , , ,
[0004] , , , , , , ,
[0001] The present invention relates to a method for estimating the shape of internal skin fibers from an image obtained by imaging a face.
Background Art
[0002] The internal fiber structure of the skin is closely related to skin elasticity. As methods for evaluating the internal fiber structure of the skin, methods such as evaluation with an in vivo confocal laser microscope, evaluation of a section of immunostained skin tissue with an optical microscope, and evaluation with an electron microscope are used. However, with these methods, it is difficult to evaluate the internal fiber structure of the skin three-dimensionally and protein-specifically.
[0003] The applicant of the present application has proposed, in Patent Document 1, a method for evaluating the internal skin structure that numerically analyzes a three-dimensional image of the internal skin structure obtained by combining skin tissue clearing and immunofluorescence staining, and quantifies the fibrin structure in the skin tissue. By the method described in Patent Document 1, it is possible to numerically evaluate the length, diameter, linearity, etc. of fibrin, which is the skeletal fiber of elastic fibers.
Prior Art Documents
Patent Documents
[0004]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0005] The method described in Patent Document 1 requires collection of skin cells, clearing treatment and immunofluorescent staining treatment of the collected skin cells by an expert, imaging of a three-dimensional image of the treated cell tissue, and analysis thereof, and it was difficult to say that it could be easily implemented. An object of the present invention is to provide a method for easily estimating the shape of internal skin fibers.
Means for Solving the Problems
[0006] This invention was developed after diligent research to solve the above problems, and it was found that the shape of internal skin fibers can be estimated using numerical parameters obtained from facial imaging images as indicators. Specifically, the means for solving the problems of the present invention are as follows: 1. An estimation method characterized by estimating the fiber shape index of internal skin fibers using facial shape parameters obtained from facial imaging images as indicators. 2. The estimation method according to 1, characterized in that the facial shape parameter is one or more selected from the group consisting of (1) to (14) below. (1) The length from the center of the nose to the point where a line extended horizontally touches the contour. (2) Within the contour, in the area between the line extended horizontally from (1) and the line extended straight down from the corner of the lips, the longest length from the center of the nose (3) The length from the point where a line drawn straight down from the center of the nose touches the tip of the chin (the point where the line drawn straight down from the center of the nose intersects with the contour). (4) The angle formed by a line drawn horizontally from the center of the mouth and a line connecting the point of contact between this line and the contour and the central tip of the chin. (5) The line extending horizontally from (1) and the line connecting the point of contact with the contour to the central tip of the chin, and the area of the arc formed by the contour. (6) The length from the center of the mouth downwards to the tip of the center of the chin. (7) The angle formed by the line extended horizontally from (1) and the line extended from the point of contact between this line and the contour to the highest point of the cheekbone. (8) Length from the highest point of the cheekbone to the outer corner of the eye (9) Ratio of face shape parameters (1) to (2) (1 / 2) (10) Ratio of face shape parameters (1) to (3) (1 / 3) (11) Ratio of face shape parameters (2) and (3) (2 / 3) (12) Ratio of face shape parameters (8) to (1) (8 / 1) (13) Ratio of face shape parameters (8) to (3) (8 / 3) (14) Maximum value of the differential curve of the brightness value in the nasolabial fold area 3. The estimation method according to 1. or 2., characterized in that the fiber shape index is one or more selected from the group consisting of A to I below. A: Straightness (ratio of elastin fiber length to straight-line distance between vertices (straight-line distance / length)) B: Fiber thickness score (a score obtained by visually determining the thickness of elastin fibers from images of elastin fibers detected by autofluorescence) C: Fiber curl score (a score that quantifies the curl of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) D: Fiber width (measured value) (Width of elastin fibers measured from images of elastin fibers detected by autofluorescence) E: Linearity score of elastin fibers (a score that quantifies the linearity of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) F: Elastin visual score (average of B, C, and E) G: Collagen concentration (sum of brightness values in images of collagen fibers detected by SHG light) H: Collagen volume (sum of voxel counts in collagen regions extracted by binarizing images of collagen fibers detected by SHG light) I: ratio (average value of the brightness of collagen regions extracted by binarizing images of collagen fibers detected by SHG light) [Effects of the Invention]
[0007] The estimation method of the present invention makes it possible to estimate the shape of internal skin fibers very easily and inexpensively. By estimating the shape of internal skin fibers more frequently than before using the estimation method of the present invention, it is possible to grasp changes in the state of internal skin fibers, and thus to more accurately understand the impact of daily life on internal skin fibers. [Brief explanation of the drawing]
[0008] [Figure 1] A diagram illustrating facial shape parameters. [Figure 2] A diagram illustrating facial shape parameters. [Figure 3] Figure showing an example of criteria for visual determination of fiber shape index B: fiber fineness, C: fiber crimp, and E: fiber straightness of internal fibers of the skin using facial shape parameters obtained from a captured facial image as an index.
Embodiment for Carrying Out the Invention
[0009] The estimation method of the present invention estimates the fiber shape index of internal fibers of the skin using facial shape parameters obtained from a captured facial image as an index. The captured facial image may be a still image captured from the front or a moving image captured from various angles.
[0010] In the estimation method of the present invention, the combination of the facial shape parameters used as an index and the fiber shape index of the internal fibers of the skin to be estimated is not particularly limited as long as they have a correlation. For example, the following (1) to (14) can be mentioned as facial shape parameters, and the following A to I can be mentioned as fiber shape indices. <(
[0011] "Facial Shape Parameter" Facial shape parameters (1) to (8) are shown in FIGS. 1 and 2. (1) Length until a line extending horizontally from the center of the nose reaches the contour (2) Among the contours, the longest length from the center of the nose in the range sandwiched between the line extending horizontally from (1) and the line extending vertically downward from the end of the lip (3) Length until a line extending vertically downward from the center of the nose reaches the center tip of the jaw (the contact point between the line extending vertically downward from the center of the nose and the contour) <( (4) Angle formed by a line extending horizontally from the center of the mouth and a line connecting the contact point between this line and the contour and the center tip of the jaw (5) Area of the arc formed by the contour and a line connecting the contact point between the line extending horizontally from (1) and the contour and the center tip of the jaw (6) Length until a line extending vertically downward from the center of the mouth reaches the center tip of the jaw (7) Angle formed by the line extending horizontally from (1) and a line extending from the contact point between this line and the contour to the highest point of the cheekbone (8) Length from the highest point of the cheekbone to the outer corner of the eye (9) Ratio of facial shape parameter (1) to (2) (1 / 2) (10) Ratio of face shape parameters (1) to (3) (1 / 3) (11) Ratio of face shape parameters (2) and (3) (2 / 3) (12) Ratio of face shape parameters (8) to (1) (8 / 1) (13) Ratio of face shape parameters (8) to (3) (8 / 3) (14) Maximum value of the differential curve of the brightness value in the nasolabial fold area
[0012] In this specification, the vertical and horizontal (left-right) directions of the face are based on the state of the face when a person is standing upright, that is, with the top of the head pointing upwards and the chin pointing downwards. The center of the nose is the highest point on the midline between the left and right sides of the nose. In (2), the area of the contour between the line extended horizontally in (1) and the line extended straight down from the corner of the lips is the part of the contour shown in Figure 2 where the darker lines overlap. In (14), the point where the differential curve of the luminance value is at its maximum is the position where the nasolabial fold is deepest.
[0013] The facial features used to identify the facial shape parameters described above may be determined visually by a human from the captured image, mechanically, or using machine learning-trained AI. Furthermore, points determined by AI may be manually corrected by a human. When determined mechanically, for example, the point with the highest brightness value within a specific area of the captured image, i.e., the point where light is most strongly reflected, can be determined as the highest point. Additionally, before identifying the facial features, the tilt and distortion of the facial image can be corrected, for example, so that the central tip of the chin is directly downwards. Face shape parameters can be defined using only one side of the face, or both sides. When using both sides, the average of the values obtained from each side is used.
[0014] "Fiber shape index" A: Straightness (ratio of elastin fiber length to straight-line distance between vertices (straight-line distance / length)) B: Fiber thickness score (a score obtained by visually determining the thickness of elastin fibers from images of elastin fibers detected by autofluorescence) C: Fiber curl score (a score that quantifies the curl of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) D: Fiber width (measured value) (Width of elastin fibers measured from images of elastin fibers detected by autofluorescence) E: Linearity score of elastin fibers (a score that quantifies the linearity of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) F: Elastin visual score (average of B, C, and E) G: Collagen concentration (sum of brightness values in images of collagen fibers detected by SHG light) H: Collagen volume (sum of voxel counts in collagen regions extracted by binarizing images of collagen fibers detected by SHG light) I: ratio (average value of the brightness of collagen regions extracted by binarizing images of collagen fibers detected by SHG light)
[0015] The fiber shape index is a value measured by collecting and observing actual tissue. There are no particular restrictions on the method of measuring the fiber shape index, but for example, excised skin tissue can be imaged with a confocal laser microscope, and the fiber shape index can be calculated using image analysis techniques. The estimation method of the present invention makes it possible to obtain an estimated value of the fiber shape index with high accuracy without performing actual measurements that involve skin tissue sampling, etc.
[0016] In the estimation method of the present invention, the method for estimating the fiber shape index of internal skin fibers using facial shape parameters as indicators is not particularly limited, and known methods can be used. For example, facial shape parameters may be subjected to simple linear regression analysis or multiple linear regression analysis. The multiple linear regression analysis may be linear multiple linear regression analysis or nonlinear multiple linear regression analysis. When performing multiple linear regression analysis, the number of facial shape parameters used as indicators is not particularly limited and can be 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, or 14. In the estimation method of the present invention, the absolute value (|r|) of the correlation coefficient between the estimated fiber shape index of internal skin fibers and the facial shape parameter which is the index is preferably 0.2 or higher, more preferably 0.3 or higher, even more preferably 0.4 or higher, even more preferably 0.5 or higher, even more preferably 0.6 or higher, even more preferably 0.7 or higher, and even more preferably 0.8 or higher. [Examples]
[0017] "Acquisition of facial shape parameters" For 40 Japanese women (ages 32-58), frontal still images of their faces were captured using a full-face camera (Visia), and one expert visually identified the position of each image to determine the facial shape parameters (1) to (14) described above.
[0018] "The fibrous shape of internal skin fibers" Forty Japanese women from whom facial shape parameters were obtained, internal skin tissue samples were collected from the cheek area. Using a multiphoton scanning laser biomicroscope (MPT), collagen and elastin fibers from the skin of the cheek were imaged. From the obtained images, the volume, brightness, and ratio values of collagen fibers, and the straightness and width of elastin fibers were measured using image analysis software (Image J). Furthermore, the shape of the elastin fibers was visually evaluated by three experts, and the thinness, curliness, and linearity of the fibers were scored on a 5-point scale. An example of the criteria for visual evaluation of fiber shape indices B: fiber thinness, C: fiber curliness, and E: fiber linearity is shown in Figure 3. "Simple linear regression analysis" Simple regression analysis was performed on each facial shape parameter and the fiber shape index. For those where a correlation was found, the correlation coefficient r, p-value, and the equation of the simple regression line (Y=aX+b, where X is the facial shape parameter value and Y is the fiber shape index) are shown in Tables 1 and 2.
[0019] [Table 1]
[0020] [Table 2] The facial shape parameters and fiber shape indices shown in Tables 1 and 2 showed a correlation (|r|≧0.2).
[0021] Multiple Regression Analysis 1 Multiple regression analysis was performed on the fiber shape index using 3 to 4 facial shape parameters as explanatory variables. For those for which a correlation was found, the equation of the multiple regression line (Y = a1X1 + a2X2 + ... + b, where Xn is the facial shape parameter value and Y is the fiber shape index) is shown in Table 3. [Table 3]
[0022] The fiber shape indices A to E showed a high correlation (|r|≧0.4) with 3 to 4 facial shape parameters as explanatory variables.
[0023] Multiple Regression Analysis 2 Multiple regression analysis was performed for each fiber shape index, using 14 facial shape parameters as explanatory variables. The equations of the multiple regression lines (Y = a1X1 + a2X2 + ... + b, where Xn is the facial shape parameter value and Y is the fiber shape index) are shown in Tables 4-6. [Table 4] [Table 5] [Table 6]
[0024] As shown in Tables 4-6, a higher correlation (|r|≧0.5) was observed with the fiber shape index when facial shape parameters were combined.
Claims
1. An estimation method characterized by estimating the fiber shape index of internal skin fibers using facial shape parameters obtained from facial imaging images as indicators.
2. The estimation method according to claim 1, characterized in that the facial shape parameter is one or more selected from the group consisting of (1) to (14) below. (1) The length from the center of the nose to the point where a line extended horizontally touches the contour. (2) Within the contour, in the area between the line extended horizontally in (1) and the line extended straight down from the corner of the lips, the longest length from the center of the nose (3) The length from the point where a line drawn straight down from the center of the nose touches the tip of the center of the chin (the point where the line drawn straight down from the center of the nose intersects with the contour). (4) The angle formed by a line drawn horizontally from the center of the mouth and a line connecting the point of contact between this line and the contour and the central tip of the chin. (5) The line connecting the point of contact between the horizontally extended line of (1) and the contour to the central tip of the chin, and the area of the arc formed by the contour. (6) The length from the center of the mouth downwards until the line touches the tip of the center of the chin. (7) The angle between the line drawn horizontally from (1) and the line drawn from the point of contact between this line and the contour to the highest point of the cheekbone. (8) The length from the highest point of the cheekbone to the outer corner of the eye (9) Ratio of facial shape parameters (1) to (2) (1 / 2) (10) Ratio of face shape parameters (1) to (3) (1 / 3) (11) Ratio of face shape parameters (2) and (3) (2 / 3) (12) Ratio of facial shape parameter (8) to (1) (8 / 1) (13) Ratio of facial shape parameters (8) to (3) (8 / 3) (14) Maximum value of the differential curve of the brightness value in the nasolabial fold area
3. The estimation method according to claim 1 or 2, characterized in that the fiber shape index is one or more selected from the group consisting of A to I below. A: Straightness (ratio of elastin fiber length to straight-line distance between vertices (straight-line distance / length)) B: Fiber thickness score (a score obtained by visually determining the thickness of elastin fibers from images of elastin fibers detected by autofluorescence) C: Fiber curl score (a score that quantifies the curl of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) D: Fiber width (measured value) (Width of elastin fibers measured from images of elastin fibers detected by autofluorescence) E: Linearity score of elastin fibers (a score that quantifies the linearity of elastin fibers by visual inspection from images of elastin fibers detected by autofluorescence) F: Elastin visual score (average of B, C, and E) G: Collagen concentration (sum of brightness values in images of collagen fibers detected by SHG light) H: Collagen volume (sum of voxel counts in collagen regions extracted by binarizing images of collagen fibers detected by SHG light) I: ratio (The average value of the brightness levels of the collagen region extracted by binarizing the image of collagen fibers detected by SHG light)