JP2015157071A - Health condition evaluation support system and capillary vessel data acquisition method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a health condition evaluation support system and a capillary vessel data acquisition method which can automatically recognize the shape of a capillary vessel and digitize it.SOLUTION: A health condition evaluation support system 10 comprises image processing means 11 which processes an image obtained by imaging a capillary vessel to recognize the shape of a capillary vessel in the image, and calculation means 12 which calculates the abnormality of the capillary vessel from the recognized shape of the capillary vessel. A capillary vessel data acquisition method comprises an image processing step for processing an image obtained by imaging a capillary vessel to recognize the shape of the capillary vessel in the image, and a calculation step for calculating the abnormality of the capillary vessel from the recognized shape of the capillary vessel.

Description

The present invention relates to a health condition evaluation support system and a capillary blood vessel data acquisition method based on the shape of a capillary blood vessel.

It is conventionally known that the shape of capillaries such as the nail epithelium and the movement of blood flow are related to the health condition. FIG. 2A schematically shows a capillary 20 of the nail epithelium. In a healthy state, the capillary 20a is linear (see FIG. 2B), but in an unhealthy state, the capillary 20b tends to meander (see FIG. 2C). Therefore, by observing the shape of the capillary 20 and the like, it can be used for discovery and treatment of various diseases.

Under such circumstances, as a system for observing the shape of capillaries, etc., the density, thickness, shape (number of twisted or deformed blood vessels) and flow velocity of the observed blood vessels are classified and analyzed, and the results are evaluated. A medical diagnosis support system that outputs a comprehensive evaluation of a health condition has been proposed (see Patent Document 1). However, in this medical diagnosis support system, the operator observes the shape, thickness, etc. of the blood vessel and inputs this numerical value, which is performed semi-automatically.

Japanese Patent No. 4743486

The present invention has been made in view of such circumstances, and it is an object of the present invention to provide a health condition evaluation support system and a capillary blood vessel data acquisition method for automatically recognizing the shape of a capillary blood vessel and evaluating the state.

The health condition evaluation support system according to the first invention in accordance with the object described above processes an image in which capillaries are imaged, image processing means for recognizing the shape of the capillaries in the images, and the recognized capillaries. Calculation means for calculating the degree of abnormality of the capillary blood vessel from the shape of the blood vessel is provided. According to the health condition evaluation support system according to the first invention, by providing the image processing means and the calculation means, the shape of the capillary blood vessel is automatically recognized from the image of the capillary blood vessel, and the recognized capillary blood vessel is recognized. The degree of abnormality of the capillary can be calculated from the shape of the blood vessel.

In the health condition evaluation support system according to the first invention, it is preferable that the calculation means calculates the degree of abnormality based on a bending level of the capillaries.

In the health condition evaluation support system according to the first aspect of the invention, it is preferable that the calculation means calculates the degree of abnormality based on the number of bends of the capillaries.

In the health condition evaluation support system according to the first aspect of the present invention, it is preferable that the calculation means calculates the degree of abnormality based on the number of capillaries that intersect with the artery and vein.

In the health condition evaluation support system according to the first aspect of the present invention, it is preferable that the calculation means calculates the degree of abnormality based on the number of branches of the capillaries.

In the health condition evaluation support system according to the first invention, the calculation means further comprises a group consisting of the length, thickness, thickness ratio, definition, width, flow velocity, area, aspect ratio, and length of the capillary. It is preferable to calculate at least one selected data. In this way, a plurality of data including the degree of abnormality can be used for multifaceted and comprehensive evaluation relating to the health condition and the like.

The health condition evaluation support system according to the first invention preferably further comprises an output means for outputting the calculated data as a radar chart. By doing in this way, it becomes easy for an evaluator (doctor etc.), a patient, etc. to perform visually multifaceted, comprehensive evaluation, judgment, etc.

In the health condition evaluation support system according to the first aspect of the invention, it is preferable that the calculation means detects the amount of at least one of red blood cells and white blood cells of the capillaries from the density of the capillaries in the image. Based on the amount of at least one of red blood cells and white blood cells detected, it is possible to know the influence on the human body due to medication, for example, administration of an anticancer agent.

In the health condition evaluation support system according to the first invention, the capillary is preferably a capillary of the nail epithelium. Since the nail epithelium is thin and the aligned capillaries can be seen through, image processing and calculation are relatively easy, and calculation accuracy can be increased.

According to a second aspect of the present invention, there is provided a capillary blood vessel data acquisition method, comprising: processing an image obtained by capturing a capillary blood vessel; recognizing the shape of the capillary blood vessel in the image; A calculating step of calculating an abnormality degree of the capillary blood vessel from the shape of the capillary blood vessel; According to the capillary blood vessel data acquisition method according to the second aspect of the present invention, it is possible to automatically recognize the shape of the capillary blood vessel and calculate the abnormality degree of the capillary blood vessel from the recognized capillary blood vessel shape.

According to the health condition evaluation support system according to the first aspect of the invention and the capillary blood vessel data acquisition method according to the second aspect of the invention, it is possible to automatically recognize the shape of the capillary and calculate the degree of abnormality of the capillary. . Therefore, in addition to diagnosis such as grasping of the health condition, it can be used for evaluation of effects such as treatment, drinking of supplements, use of health equipment, and application of cosmetics and coatings to the skin.

It is a schematic diagram which shows the health condition evaluation assistance system which concerns on the 1st Embodiment of this invention. (A) is a schematic diagram showing capillaries of nail epithelium, (b) is a schematic diagram showing capillaries in a healthy state, and (c) is a schematic diagram showing capillaries in an unhealthy state. (A) is explanatory drawing explaining each data of a capillary vessel, (b) is a radar chart as an output of each data. It is explanatory drawing of the method of searching a blood vessel area | region. (A), (b), (c) is a schematic diagram which respectively shows the capillary vessel of an unhealthy state.

Next, embodiments of the present invention will be described with reference to the accompanying drawings.
As shown in FIG. 1, the health condition evaluation support system 10 according to the first exemplary embodiment of the present invention includes an image processing unit 11, a calculation unit 12, and an output unit 13.

As the image processing unit 11 and the calculation unit 12, a known computer 14 can be used. Specifically, the computer 14 includes a control unit including a CPU, a ROM, a RAM, a hard disk, a keyboard, and the like. The computer 14 functions as an image processing unit 11 by processing an image input based on a computer program stored in a ROM, a hard disk or the like, and functions as a calculation unit 12 by calculating based on a computer program. It is configured as follows.

The image processing unit 11 receives an image (see FIG. 3A) obtained by capturing the capillaries 20. The image processing means 11 processes this image and recognizes the shape of the capillary 20 in the image. The captured image may be a color image or a monochrome image. Moreover, although a still image or a moving image may be sufficient, when measuring the flow velocity, it is preferable that it is a moving image. Capillary blood vessel 20 can be imaged by a known blood flow observation device (for example, Bscan manufactured by Toku Co., Ltd.). Although it is not particularly limited to which part of the living body the capillary blood vessel is imaged, the capillary blood vessel of the nail epithelium can be preferably imaged. The input of the image to the image processing means 11 may be input in real time by connecting the blood flow observation apparatus or the like and the image processing means 11 or may take in the image data stored in the recording medium.

Image processing by the image processing means 11 (recognition of the shape of the capillary 20 in the image) can be performed, for example, by the following procedure (algorithm).

(1) Gray scale conversion Gray scale conversion is processing for converting a color image into a monochrome image (gray scale image) having only lightness when the captured image is color.
(2) Binarization Binarization is processing that eliminates an intermediate value (gray value) from a grayscale image and converts it into a complete white or black binary image. A binary image is obtained, for example, by converting a pixel value equal to or greater than a certain threshold value to a white pixel and a pixel value less than the threshold value to a black pixel. The threshold value can be appropriately set with a scale bar.
(3) Noise processing Noise is processed by raster scanning and labeling. The raster scan is a scan in which the pixel is examined from the left end to the right starting from the upper left of the image, and when reaching the right end, the row is lowered by one and the pixel is examined from the left end to the right. Labeling is a process of assigning the same number (label) to each pixel having the same connected component. As noise processing, if the labeled area (white pixel) is less than a certain area, the area is erased, and if the area of black pixel is less than a certain area, it is complemented (filled).
(4) Edge detection Edge detection is a process of extracting a contour line of an object captured as an image. As the edge detection algorithm, Kenny's edge detection algorithm can be used.
(5) Feature point extraction With feature point extraction, end points, intersections, and corners of lines (capillary vessels 20 in the present embodiment) in an image are detected. As the feature point extraction algorithm, Harris corner detection or the like can be used.
(6) Rearrangement of feature points The extracted feature points are rearranged in ascending order with respect to the x coordinate, for example.
(7) Classification of feature points The feature points on the vein side and the artery side are classified. For this classification, a support vector machine, which is one of the two-class classifiers, can be used.
(8) Rearrangement of post-classification data The post-classification feature points are rearranged in ascending order with respect to the y coordinate, for example.
(9) Complementary processing Complementary processing is performed between adjacent feature points.

As an image processing method different from the above, a method (algorithm) for searching a blood vessel region can also be used. This method will be described with reference to FIG. An extraction starting point A is arranged at a point that is equal to or greater than a predetermined pixel value (for example, 0.8) and exists at the end of the image, and a blood vessel region is estimated using the gray value of the image and sequentially searched. That is, the pixel D having the highest gray value in the range of the sector C having a predetermined central angle centered on the tip B (initially the starting point) of the target blood vessel region is searched, and the pixel D is assumed to be a blood vessel region. This is a method of searching a blood vessel region.

In the present embodiment, the image processing unit 11 has a function of specifying the position of the capillary 20 in the captured image based on data (for example, coordinates) manually input by an input device. This is effective when it is difficult to automatically specify the shape of the capillary 20 by image processing because the captured capillary 20 is unclear.
Further, the computer 14 stores the shape of the capillary 20 specified by the image processing unit 11 in a storage medium. The computer 14 is provided with a shape comparison means (not shown) comprising a computer program. The shape comparison means obtains a plurality of data of the shape of the designated capillary 20 from a storage medium, and is imaged on different days. The shapes of the capillaries 20 are compared. By checking whether or not the shape of the capillary blood vessel 20 has improved after a predetermined period of time, it is possible to know the effect of introducing the drug or supplement.

The calculating means 12 calculates data (index) such as the abnormal degree, length, thickness, thickness ratio, sharpness, width, flow velocity, area, aspect ratio, and portrait of the capillary 20 from the recognized shape of the capillary 20. ) Is calculated. Hereinafter, description will be made with reference to FIG. Moreover, the arrow in Fig.3 (a) shows the direction of a blood flow.

The degree of abnormality of the capillary 20 is one of the bending level of the capillary 20, the number of bends of the capillary 20, the number of branches of the capillary 20, and the number of crossings of the artery and vein of the capillary 20, or Calculated based on multiple combinations.
In the present embodiment, an integrated value of angle change θ per unit length of capillary blood vessel 20 (hereinafter also referred to as “flexion degree”) is adopted as the bending level of capillary blood vessel 20. The degree of bending can be calculated by the following formula from the coordinates of each feature point, for example. The degree of flexion is related to the degree of blood flow stagnation. If the degree of flexion is large, there is a possibility of hyperlipidemia, inflammation, abnormal tissue growth, and the like.

Here, the calculation means 12 may employ an integrated value of the bending angle of the center line drawn between the artery and vein of the capillary 20 as the bending level of the capillary 20 instead of the bending degree. The center line is drawn, for example, at a position where the distance between the center line and the artery of the capillary 20 and the distance between the center line and the vein of the capillary 20 are equal. It means the change in angle per unit length.

The number of capillary bends is the number of times the capillaries are bent (for example, the number of times the angle has changed by a predetermined angle or more per unit length), and the number of capillaries of the S-shaped capillaries 21 shown in FIG. In this case, the number of bends is 2.
The number of capillary branches is 3 in the case of the capillary 22 shown in FIG. 5B, and the number of intersections of the capillaries is 1 in the case of the capillary 23 shown in FIG. In the case of a normal capillary blood vessel, the number of branches is 1, that is, the unbranched state shown in FIG. As shown in FIG. 5 (c), when the capillaries intersect, an artery and a vein intersect, and when one region of the artery intersects another region of the artery, one region of the vein and the vein This includes the case where other regions intersect.
The degree of abnormality of the capillary 20 is, for example, K * (number of bends of the capillary 20) + L * (bend of the center line drawn between the artery and vein of the capillary 20) with K, L, M, and P as coefficients. It can be calculated as (integrated value of angle) + M * (number of branches of the capillary 20) + P * (number of intersections of the artery and vein of the capillary 20).

The length and area S of the capillary blood vessel 20 serve as indices indicating the smoothness of the entire blood flow, but the area S can be expressed more sensitively. The length can be obtained as an integrated value of the distance between feature points. Instead of the length, a vertically long (height) h may be used as this index. Here, the vertically long h means the length of the capillary blood vessel 20 in the y-axis direction (direction along the finger) as shown in FIG. 3A, and the value of the vertically long h increases as the blood flow improves. Tend.
The area S may be an area surrounded by a capillary and a line segment connecting both ends of the capillary, or may be a product of a width W and a longitudinal length h. The length of the capillary 20 is about 200 to 800 μm.

The thickness of the capillary 20 is related to the stagnation of the arteries and veins and the degree of venous congestion. As for the thickness of the capillary vessel 20, for example, the values of the arterial part, the venous part, and the tip part which is the boundary between them can be measured. The thickness ratio of the capillary vessel 20 is the thickness R2 (R2 / R1) of the arterial portion with respect to the thickness R1 of the vein portion, and is related to the degree of venous congestion.

The definition of the capillary 20 is related to the accumulation of fatigue substances and the degree of metabolism. The sharpness can be obtained, for example, as the difference between the luminance of the background region and the luminance of the blood vessel region.

The width W of the capillary 20 is related to the degree of hyperlipidemia. The width W is the maximum distance between the arterial side and the venous side, and can be obtained as the maximum difference in x coordinate values within a fixed y coordinate.

The flow velocity (velocity of blood flow) in the capillary blood vessel 20 can be obtained from a change in shading or a moving object existing in the blood vessel.
Here, since the state in which red blood cells, white blood cells, or plasma is flowing through the capillary 20 can be detected by the change in shading, one of the red blood cells, white blood cells, and plasma based on the shading change, Alternatively, a plurality of ratios may be derived so that the amount can be detected. When detecting a change in shading, the image input to the image processing means 11 is preferably a moving image. Detecting the proportion and amount of a specific component of blood is effective in evaluating the health condition. For example, by obtaining a decreased level of red blood cells and white blood cells that are decreased by administration of an anticancer drug, the health condition of the human body is obtained. It can be useful for inspection.
Depending on the reagent introduced into the body, the distribution of the reagent can be confirmed by a change in the concentration of the capillary 20, and further, the concentration of the capillary 20 can be obtained by dyeing specific cells (for example, cancer cells) with the reagent. From the change, the number and concentration of the specific cells can be detected.

The data (index) is not limited to the above-described index, and other indices may be used, or an index that combines the indices by addition / subtraction / multiplication / division or the like. Examples of the combined index include aspect ratio (= width W / length h), roundness (= area of circle with diameter as width W / inside area), and the like. In addition, it is better that both the aspect ratio and the roundness are small.

The output means 13 is a known monitor, printer, or the like, and is connected to a computer 14 (image processing means 11 and calculation means 12). Various data calculated by the image processing unit 11 and the calculation unit 12 are output to the output unit 13. The output data can be appropriately selected as necessary. The output form of the output means 13 includes a radar chart in addition to a table. Specifically, for example, as shown in FIG. 3B, it can be displayed as a radar chart with six items of abnormality, thickness ratio, definition, width, flow velocity, and area. In FIG. 3B, two types of results are output. When the degree of bending is an item of the radar chart, the degree of bending may be displayed as it is on the radar chart, or the degree of bending may be expressed as a straightness with the degree of bending as an inverse. In addition, the degree of abnormality may not be displayed on the radar chart. In the radar chart, each item may be configured to output a normalized value by defining a normal range, a standard value, etc. from a literature value, a sample value, or the like.

The radar chart items (indicators) are not limited to the above six items, and the number of items is not limited to six items. In addition to or in place of the above six items, items such as portrait or other combinations (for example, aspect ratio) may be used. For example, the radar chart may include six items, and other items (vertical length, aspect ratio, etc.) may be output as additional items outside the radar chart. Any item may be used as an element of a radar chart, or may be used as an additional item other than an element of a radar chart. As an evaluation method using the output data, comprehensive evaluation may be performed from the items of the radar chart, or simple evaluation using some indexes (for example, combination indexes) may be performed. What is necessary is just to select suitably the number and kind of data (item) used for evaluation according to the objective.
The calculating means 12 can also calculate the capillary age based on the value of each item displayed as a radar chart. Examples of the calculation method include averaging the values of the items displayed on the radar chart, and calculating the total value after weighting the values of the items.

Next, as a second embodiment of the present invention, a data acquisition method for capillaries 20 using the health condition evaluation support system 10 will be described. In this data acquisition method, an image of the capillary 20 is processed, an image processing step for recognizing the shape of the capillary 20 in the image, and the degree of abnormality of the capillary from the recognized shape of the capillary 20; In addition, there is a calculation step for calculating length, thickness, thickness ratio, definition, width, flow velocity, area, aspect ratio, length, and the like. In the health condition evaluation support system 10, the image processing step is performed by the image processing unit 11, and the calculation step is performed by the calculation unit 12. Each calculated data is output by the output means 13 in the form of a table or a radar chart.

The data on the capillaries obtained in this way can be used as a basis for diagnosis by doctors, as well as evaluation of the effects of treatment, drinking supplements, using health equipment, and applying cosmetics and coatings to the skin. Can be useful. In addition, each data can be observed at a fixed point and the change with time can be examined.

The present invention is not limited to the above-described embodiment, and the configuration thereof can be changed without changing the gist of the present invention.

The health condition evaluation support system and the capillary blood data acquisition method according to the present invention are used as medical devices, as evaluation devices when developing new drugs and supplements, as health devices that measure the effects of training, and the like. It can be used as a data acquisition method.

10: health condition evaluation support system, 11: image processing means, 12: calculation means, 13: output means, 14: computer, 20, 20a, 20b, 21, 22, 23: capillary, A: starting point, B: tip , C: Fan shape, D: Pixel

Claims (10)

Image processing means for processing an image in which capillaries are captured, recognizing the shape of the capillaries in the image, and calculating means for calculating the degree of abnormality of the capillaries from the recognized capillaries. Health condition evaluation support system provided. The health condition evaluation support system according to claim 1, wherein the calculation unit calculates the degree of abnormality based on a bending level of the capillaries. 2. The health condition evaluation support system according to claim 1, wherein the calculation means calculates the degree of abnormality based on the number of bends of the capillaries. 2. The health condition evaluation support system according to claim 1, wherein the calculation means calculates the degree of abnormality based on the number of intersections of arteries and veins of the capillaries. 2. The health condition evaluation support system according to claim 1, wherein the calculation means calculates the degree of abnormality based on the number of branches of the capillaries. The health condition evaluation support system according to any one of claims 1 to 5, wherein the calculation means further includes a length, a thickness, a thickness ratio, a sharpness, a width, a flow velocity, an area, a length and a width of the capillary. A health condition evaluation support system that calculates at least one type of data selected from a group consisting of a ratio and a portrait. 7. The health condition evaluation support system according to claim 6, further comprising output means for outputting the calculated data as a radar chart. The health condition evaluation support system according to any one of claims 1 to 7, wherein the calculation unit detects the amount of at least one of red blood cells and white blood cells of the capillaries from the density of the capillaries in the image. Health condition evaluation support system. The health condition evaluation support system according to any one of claims 1 to 8, wherein the capillary is a capillary of a nail epithelium. An image processing step of processing an image in which a capillary is imaged, recognizing the shape of the capillary in the image, and a calculation step of calculating a degree of abnormality of the capillary from the recognized shape of the capillary Capillary blood vessel data acquisition method.

Images