JP2023161233A - Calculation method of blood vessel elastic force index - Google Patents

Abstract

To provide a technique for calculating a hemoglobin amount at each time from a dynamic image of a prescribed region of a body surface of a subject, and on the basis of the change of the hemoglobin amount, calculating a blood vessel elastic force index.SOLUTION: A calculation method of a blood vessel elastic force index comprises: an image acquiring step for acquiring a time series image of a prescribed region of a body surface; a hemoglobin amount calculating step for calculating a hemoglobin amount at each time of the time series image; a period calculating step for calculating one period of a heart beat on the basis of a change of the hemoglobin amount calculated; and a blood vessel elastic force index calculating step for calculating a blood vessel elastic force index, on the basis of the change of the hemoglobin amount.SELECTED DRAWING: Figure 1

Description

The present invention relates to a method for calculating a blood vessel elasticity index.

There is a method of measuring blood flow in the skin using a laser speckle method (Patent Document 1).
There is also a method of creating a hemoglobin image using face image data photographed using an RGB camera, creating heart rate data from the hemoglobin image, and performing stress monitoring (Patent Document 2).

Patent No. 6507639 JP2017-29318A

In Patent Document 1, measurement is performed using a light intensity image photographed by irradiating a laser beam, so it is not suitable for a user to measure the blood flow rate in the skin in daily life.
Further, in Patent Document 2, stress monitoring was performed using hemoglobin images, but other biological indicators could not be calculated, so there was room for research.

The present invention has been made in view of the above problems, and provides a blood vessel that calculates the amount of hemoglobin at each time from a moving image of a predetermined area of the body surface of a subject, and calculates a blood vessel elastic force index based on the change in the amount of hemoglobin. Concerning the calculation method of elastic force index. In this specification, "method for calculating blood vessel elasticity index" and "method for evaluating blood vessels and blood flow" refer to calculation of blood vessel elasticity index and evaluation of blood vessels and blood flow of a subject for non-medical purposes. This includes calculation methods and methods for evaluating blood vessels and blood flow that can be performed even by non-experts. Specifically, the method of evaluating blood vessels and blood flow includes a method of giving advice on prevention of lifestyle-related diseases and a method of estimating age.

The present invention includes an image acquisition step of acquiring time-series images of a predetermined region of the body surface, a hemoglobin amount calculation step of calculating the hemoglobin amount at each time of the time-series images, and a heart rate based on the calculated change in the hemoglobin amount. A method for calculating a vascular elasticity index, comprising: a period calculation step of calculating one period of the hemoglobin amount; and a vascular elasticity index calculation step of calculating a vascular elasticity index based on the change in the amount of hemoglobin and the calculated period. Regarding.

The present invention also provides an image acquisition section that acquires time-series images of a predetermined region of the body surface, a hemoglobin amount calculation section that calculates the hemoglobin amount at each time of the time-series images, and a change in the calculated hemoglobin amount. The present invention relates to a blood vessel elasticity index calculation system, including a cycle calculation unit that calculates one cycle of a heartbeat based on the change in the amount of hemoglobin, and a blood vessel elasticity index calculation unit that calculates a blood vessel elasticity index based on the change in the amount of hemoglobin.

Further, the present invention includes at least a processor and a memory, and the processor includes an image acquisition unit that acquires time-series images of a predetermined region of the body surface, and a hemoglobin amount calculator that calculates the amount of hemoglobin at each time of the time-series images. means, a cycle calculation means for calculating one cycle of a heartbeat based on the calculated change in the hemoglobin amount, and a blood vessel elasticity index calculation means for calculating a blood vessel elasticity index based on the change in the hemoglobin amount, The present invention relates to a blood vessel elastic force index calculation device.

According to the method provided by the present invention, the vascular elasticity index can be calculated based on the change in hemoglobin amount at each time calculated from the acquired time-series images (moving images), so the vascular elasticity index can be easily calculated. It can be used for daily health management.

FIG. 2 is a conceptual diagram of a method for calculating a blood vessel elastic force index. FIG. 7 is a diagram showing a comparison result between a case where a blood vessel elastic force index is calculated using LSFG and a case where a blood vessel elastic force index is calculated using a method for calculating a blood vessel elastic force index. FIG. 7 is a diagram showing a comparison result between a case where a blood vessel elastic force index is calculated using LSFG and a case where a blood vessel elastic force index is calculated using a method for calculating a blood vessel elastic force index. It is a figure explaining each blood vessel elastic force index. It is a flowchart which shows the calculation method of a blood vessel elastic force index. It is a flow chart showing a hemoglobin amount calculation process. It is a flowchart which shows a cycle calculation process. It is a flowchart which shows a blood vessel elastic force index calculation process. It is a figure explaining the calculation process of a blood vessel elastic force index. It is a figure showing a calculation result. It is a figure showing a calculation result. It is a figure showing a calculation result. FIG. 1 is a conceptual diagram of a blood vessel elastic force index calculation system. FIG. 2 is a conceptual diagram of a blood vessel elasticity index calculation device. It is a figure showing an evaluation result.

Examples of preferred embodiments of the present invention will be described below with reference to the drawings. Note that the drawings of this embodiment are for explaining the technical concept, configuration, and operation of the present invention, and do not specifically limit the configuration. Further, in all the drawings, similar components are given the same reference numerals, and overlapping explanations will be omitted as appropriate.

<Summary>
An overview of the method for calculating the vascular elastic force index in this embodiment will be explained.
The method for calculating the vascular elasticity index of this embodiment (hereinafter sometimes referred to as the present method) includes an image acquisition step of acquiring time-series images of a predetermined region of the body surface, and a step of acquiring time-series images of a predetermined region of the body surface, and A hemoglobin amount calculation step for calculating the amount of hemoglobin, a period calculation step for calculating one cycle of the heartbeat based on the change in the calculated hemoglobin amount, and a blood vessel elasticity index based on the change in the hemoglobin amount and the calculated period. The method is characterized by including a step of calculating a blood vessel elastic force index.

Until now, when calculating the blood vessel elastic force index on the body surface, LSFG (Laser Speckle Flowgraphy), which is a technology that visualizes blood flow distribution using reflected and scattered laser light, has been used, for example, as in Patent Document 1. It was getting worse. However, since LSFG requires a device that irradiates laser light, it has been difficult for general users to obtain a blood vessel elasticity index on a daily basis at home or the like. Therefore, the present inventors propose a calculation method that allows the vascular elasticity index to be calculated on a daily basis by calculating the hemoglobin amount from moving image data taken of the body surface and calculating the vascular elasticity index.

FIG. 1 shows a conceptual diagram of this method.
As shown in Figure 1, this method uses a visible light camera, such as a web camera attached to a personal computer, a camera built into a tablet terminal, a camera built into a smartphone, or a digital video camera, to scan the subject's body surface ( In this embodiment, the skin of the face is photographed. Since this method calculates the amount of change in hemoglobin amount and extracts the blood flow waveform, the captured images need to be time-series images (moving images).
The captured time-series images contain not only the blood flow (signal) required in this method, but also noise such as the movement of the subject and uneven lighting during imaging. Therefore, by performing independent component analysis on the RGB waveform of each pixel in a time-series image, separating it into hemoglobin, melanin, and shadow components, and performing predetermined filter processing, a waveform containing only the hemoglobin component (blood flow waveform) is extracted. . In this embodiment, when extracting the waveform of only the hemoglobin component, a known technique (for example, a pigment component analysis method from a skin color image (References: Norimichi Tsumura, Nobutoshi Ojima, Kayoko Sato, Mitsuhiro Shiraishi, Hideto Shimizu, Hirohide Nabeshima , Syuuichi Akazaki, Kimihiko Hori, Yoichi Miyake, Image-based skin color and texture analysis/synthesis by extracting hemoglobin and melanin information in the skin, acm Transactions on Graphics, Vol. 22, No. 3.pp. 770-779(2003 ). (Proceedings of ACM SIGGRAPH 2003))).

Next, FIGS. 2 and 3 show comparison results between the case where the vascular elastic force index is calculated by LSFG and the case where the vascular elastic force index is calculated by this method. The measurement conditions here are as follows.
Target: 7 people in their 30s to 50s (total of men and women)
Method: Measuring facial skin blood flow at rest for 10 seconds Camera shooting conditions: 30 [fps], shooting for 10 seconds Image compression: None Illumination: LED (5000K)
Measurement site: Left cheek Measurement area: Approximately 4cm ² (2cm x 2cm)
Number of pixels: approximately 5000 pixels Although FIGS. 2 and 3 show BOT (Blowout Time) among blood vessel elasticity indicators, similar results can be obtained with other blood vessel elasticity indicators as described later.
As shown in Figure 2, when the graphs of the blood flow rate calculated using LSFG and the hemoglobin amount calculated using this method are displayed overlapping, their waveforms match well, and as a result, the It can be seen that the value of BOT calculated by this method allows calculation of a waveform close to the value of BOT calculated using LSFG. Moreover, as shown in FIG. 3, it can be seen that there is a high correlation between the value of BOT calculated by this method and the value of BOT calculated using LSFG.
In this way, the results of calculating the vascular elasticity index using this method and the results of calculating the vascular elasticity index using LSFG are close values. Therefore, it becomes possible to measure (calculate) the vascular elasticity index in daily life, and health management can be performed using the vascular elasticity index.

<Calculation method of vascular elasticity index>
The method for calculating the vascular elasticity index will be explained.
"Body surface" refers to the surface of the body, including skin, skin, lips, etc. In addition, parts such as the eyeballs, palpebral conjunctiva, and gums that are not exposed (cannot be touched) when the eyes or mouth are closed can be exposed by opening the eyes or mouth. In other words, parts that can be observed from the outside without medical treatment such as surgery are included in the body surface. The "predetermined area on the body surface" may be any area on the body surface, and in this embodiment, the predetermined area is provided on the cheek or forehead of the face.
A "time series image" is an image obtained by photographing a predetermined region of the body surface for a predetermined period of time. The number of frames per second of time-series images may be 5 [fps] or more, preferably 30 [fps] or more. Further, the photographing time of the time-series images may be 2 seconds or more, preferably 10 seconds or more. A time-series image can also be said to be a moving image composed of multiple images.
"Image acquisition process" is a process of acquiring time-series images of a predetermined region of the body surface, such as from a predetermined medium, via a network, or directly from an imaging device (imaging directly from an imaging device). ) etc., the source of the time-series images does not matter.
"Hemoglobin amount at each time" is the value obtained by extracting only the hemoglobin component from each frame of the acquired time-series image and calculating the hemoglobin amount of each pixel. "Hemoglobin amount calculation step" is the calculation step It is. The calculation method will be described later.
"Change in the amount of hemoglobin" refers to a change in the amount of hemoglobin at each time, and the change is grasped based on the value calculated for each frame of the time-series image.
"One cycle of a heartbeat" is one cycle of a heartbeat, which is a heartbeat, and the period from contraction to expansion to the next contraction (or , the period from expansion to contraction to the next expansion). "Calculating one cycle of a heartbeat" refers to cutting out one cycle of a heartbeat based on a change in the amount of hemoglobin, since the amount of hemoglobin changes depending on the heartbeat. Specifically, the start and end points of a cycle are identified based on predetermined criteria (for example, selecting the rising time when the hemoglobin amount changes from decreasing to increasing) from waveform information of the hemoglobin amount that is repeated approximately periodically, and one cycle is determined. One example is to cut out the . At this time, the length of one heartbeat cycle may be acquired as real time, or the length of one heartbeat cycle may be acquired as a predetermined dimensionless time (for example, 100). The "cycle calculation step" is a step of cutting out one cycle of a heartbeat.
The "vascular elastic force index" may be any information that can be used as an index regarding the flow (flow velocity or flow rate) of body fluid (mainly blood) in the skin tissue on the body surface, such as Fluctuation, Skew (deviation of blood flow waveform). ), Blowout Score (BOS) (steady flow index of blood flow), BOT (Blowout Time), Rising rate (rate of increase in blood flow), Falling rate, Flow Acceleration Index (FAI), Acceleration Time Index (ATI), RI (Resistivity Index), etc. Although the details will be described later, by evaluating the blood vessel elasticity index, especially BOT, Falling rate, or ATI, it is possible to determine whether the blood vessel is in a flexible state (able to expand and contract) or hard (able to expand or contract).・It is possible to understand (estimate) whether it is not possible to contract or not. The "vascular elasticity index calculation step" is a step of calculating at least one of the blood vessel elasticity indexes.

Based on FIG. 4, the vascular elastic force index will be explained.
BOT (Blowout Time): Sustainability of high blood flow is an indicator of persistence of high MBR value, and BOT value is the half width (time) (W) of MBR value and one heartbeat width shown in Figure 4 It is expressed by the following formula using (time) (F) and proportionality constant "C".
(BOT)=C・(W)/(F)
Rising rate: This is a value that indicates the time variation in the rate of increase in blood flow, and is expressed by the following formula ("C" is a constant, and "S ₁ " and "Sall" are the areas shown in FIG. 4. be).
Rising rate=C×(S ₁ /Sall)
Falling rate: Time variation in blood flow rate is an index indicating the time variation in the rate of fall from the maximum MBR value in the time waveform of the MBR value per heartbeat. Using (Sall) and (S2) shown in FIG. 4, Fallin rate can be expressed by the following formula. "C" below is a proportionality constant.
(Falling rate)=C・(S2)/(Sall)
FAI: Maximum rising acceleration when blood flow rises indicates the instantaneous maximum blood flow when blood flow rises.
ATI: Blood flow peak position is the ratio of the time to reach the maximum MBR value (MBRmax in FIG. 4) of the MBR time waveform in one heartbeat.
RI (Resistivity Index): Peripheral vascular resistance is an index that indicates peripheral vascular resistance, and the RI value is the difference between the maximum MBR value (MBRmax in Figure 4) and the minimum MBR value (MBRmin in Figure 4) with the maximum MBR value. Obtained by dividing.

5 to 8 show flowcharts of the method. Note that FIGS. 5 to 8 show only the characteristic processing of this method.
FIG. 5 is the main flowchart of the method. This method includes an image acquisition step, a hemoglobin amount calculation step, a period calculation step, and a blood vessel elastic force calculation step.

The image acquisition step (step S100) is a step of acquiring time-series images obtained by photographing a predetermined region of the body surface for a predetermined period of time. In this step, the time-series images may be acquired from a predetermined medium, via a network, from an imaging device, etc., regardless of the source.

The hemoglobin amount calculation step (step S200) is a step of calculating the hemoglobin amount of each pixel from the acquired time-series image data. FIG. 6 shows a flowchart of the hemoglobin amount calculation process.

Step S210 calculates Red, Green, and Blue information for each pixel for each frame of the acquired time-series image data. This method uses images taken by a visible light camera such as a web camera attached to a personal computer, a camera built into a tablet terminal, a camera built into a smartphone, or a digital video camera, so the RGB information of each pixel is used. It is possible to calculate and extract the RGB waveform.

Step S220 is a process of extracting hemoglobin component values from the RGB waveform (extracting the waveform of the amount of hemoglobin).
Here, a method for calculating hemoglobin components using this method will be explained. Oxygenated hemoglobin present in the blood of blood vessels (arteries) has the property of absorbing specific incident light, so by utilizing this property, it is possible to monitor the blood flow rate, which changes with the pulsation of the heart. The extracted RGB waveform includes a hemoglobin component, a melanin component, a shadow component, and the like, and in this method, components other than the hemoglobin component become noise. Therefore, independent component analysis is performed on the RGB waveform to extract only the hemoglobin component. By extracting only the hemoglobin component (creating a waveform of only the hemoglobin component) in this way, blood flow can be monitored and a blood vessel elasticity index can be calculated. Note that in this embodiment, as described above, only the hemoglobin component is extracted from the RGB waveform using the technique of analyzing pigment components from a skin color image.

Step S230 performs inversion processing (positive/negative inversion processing) of the waveform of the extracted hemoglobin amount. This is reflected light when blood flow is monitored using LSFG, and hemoglobin absorbance is used when blood flow is monitored using an image taken with a visible light camera as in this method. Since reflected light and absorbance need to have an inverse relationship, accurate evaluation cannot be made by comparing the vascular elastic force index calculated using the extracted waveform as is. In this embodiment, in order to compare the vascular elasticity index calculated by this method and the vascular elasticity index calculated using LSFG, and evaluate the effectiveness of the method for calculating the vascular elasticity index by this method, Performs inversion processing on the extracted waveform.

Returning to FIG. 5, the cycle calculation step (step S300) is a step of cutting out one cycle of heartbeat using the waveform of the amount of hemoglobin and calculating the average one heartbeat waveform. FIG. 7 shows a flowchart of the cycle calculation process. The processing contents of the period calculation step will be explained in conjunction with the graph shown in FIG.

Step S310 performs waveform peak detection and processing to cut out each heartbeat waveform.
FIG. 9(1) shows a waveform of the amount of hemoglobin (a waveform subjected to inversion processing in step S230). The negative peak position of the waveform of the amount of hemoglobin is detected as the minimum value, and it is extracted for each heartbeat. In the case of FIG. 9(1), waveforms for 11 heartbeats can be extracted. In this embodiment, only a predetermined value within 20% of the median number of frames constituting one heartbeat is used. This is to eliminate waveforms that are distorted due to measurement errors to obtain the vascular elasticity index. The data range (predetermined value 20%) used for processing may be set as appropriate.

In step S320, normalization processing is performed for each heartbeat waveform extracted in step S310. The normalization process may be performed on each of the hemoglobin component amount (vertical axis) and the time (horizontal axis) for one heartbeat waveform, or may be performed on either one. In this embodiment, normalization processing is performed on the hemoglobin component amount (vertical axis).
FIG. 9(2) shows a graph in which the waveforms cut out for each heartbeat waveform are normalized and then superimposed.

Step S330 performs processing to calculate an average one heartbeat waveform. Since the normalization process is performed for each heartbeat waveform in step S320, the average calculation is performed for each time.

Step S340 performs processing to normalize the calculated average one heartbeat waveform. In this embodiment, the normalization process in this step is performed on the hemoglobin component amount (vertical axis) and time (horizontal axis). FIG. 9(3) shows the calculated average one heartbeat waveform.

In this manner, the period calculation step (e.g., step S300 corresponds) calculates multiple periods of heartbeat (e.g., step S310 corresponds to), and calculates the waveform indicating the change in hemoglobin amount for the calculated multiple periods of the heartbeat. The method includes a waveform normalization step (for example, step S320 is equivalent) of normalizing the waveform by the amount, and one cycle of the heartbeat is calculated based on the normalized waveform (for example, step S310 is equivalent). Furthermore, the cycle calculation step (e.g., step S300 corresponds to this) includes an average waveform calculation step (e.g., step S330 corresponds to) which averages the normalized waveform of a plurality of periods to obtain an average waveform of the hemoglobin amount. .

Returning to FIG. 5, the blood vessel elastic force calculation step (step S400) is a step of calculating a blood vessel elastic force index from the acquired average one heartbeat waveform. The indicators shown in FIG. 4 are calculated. Note that although it is possible to calculate all indicators, it is not necessary to calculate all indicators. Necessary indicators may be calculated according to the evaluation content.
FIG. 8 shows a flowchart of the blood vessel elasticity calculation process. Steps S410 to S470 in FIG. 8 differ in the types of feature amounts related to changes in the amount of hemoglobin extracted from one cycle of heartbeat, and it is only necessary to calculate the necessary feature amounts depending on the blood vessel elasticity index to be calculated. (It is sufficient to perform only the necessary processing from step S410 to step S470), and it is not necessary to perform all the processing.

Step S410 is a process of calculating a first time, which is the half width of the maximum value of the hemoglobin amount, based on the average one heartbeat waveform.
Step S420 is a process of calculating a second time, which is the total length from the rise to the fall of the hemoglobin amount, based on the average one heartbeat waveform.
Step S430 is a process of calculating a third time period from when the hemoglobin amount rises to when it reaches its maximum value based on the average one heartbeat waveform.
Step S440 is a process of calculating the fourth time period from the peak time to the fall time of the hemoglobin amount based on the average one heartbeat waveform.
Step S450 is a process of calculating the maximum value of hemoglobin amount based on the average one heartbeat waveform.
Step S460 is a process of calculating the minimum value of the hemoglobin amount based on the average one heartbeat waveform.
Step S470 is a process of calculating the maximum increase in hemoglobin amount per predetermined time based on the average one heartbeat waveform.
Step S480 is a process of calculating a blood vessel elasticity index (for example, at least one of BOT, ATI, Falling rate, RI, and FAI).
BOT is calculated using the first time and the second time.
ATI is calculated using the second and third hours.
Falling rate is calculated using the fourth time and the maximum value.
RI is calculated using the maximum value and minimum value.
FAI is calculated using the maximum increase.

In this way, the feature amount calculation step further includes a feature amount calculation step (e.g., steps S410 to S440) that calculates a feature amount related to the hemoglobin amount and a change in the hemoglobin amount from one cycle of the heartbeat. A first time that is the half width of the maximum value (for example, step S410), a second time that is the entire length from the rise to the fall of the hemoglobin amount (for example, step S420), and a second time that is the half width of the hemoglobin amount from the rise to the maximum. One or more of the third time from the peak time of the hemoglobin amount to the peak time (for example, step S430) and the fourth time from the peak time to the fall time of the hemoglobin amount (for example, step S440) is calculated as a feature quantity. In the vascular elasticity index calculation step, the vascular elasticity index (for example, BOT, ATI, Falling rate is equivalent) is calculated using one or more of the first time, second time, third time, or fourth time. (e.g., step S480 corresponds to this).
Further, it further includes a feature amount calculation step (e.g., steps S450 to S470 correspond) of calculating a feature amount related to a change in hemoglobin amount from the hemoglobin amount and one cycle of the heartbeat, and the feature amount calculation step includes a maximum value of the hemoglobin amount. (e.g., step S450), a minimum value of the hemoglobin amount (e.g., step S460), and a maximum increase in the amount of hemoglobin per predetermined time (e.g., step S470). In the vascular elasticity index calculation step, the vascular elasticity index (for example, RI, FAI (e.g., step S480 is equivalent).

Next, calculation results in this embodiment will be explained.
<Measurement conditions>
Target: Male and female healthy subjects in their 20s to 70s N = 100 (before exclusion) Analyzed region: cheek Imaging conditions for this method: 30 [fps] 10 second imaging Image compression: None Illumination: LED (5000K)
Measurement area: Forehead Measurement area: Approximately 4cm ² (2cm x 2cm)
Number of pixels of measurement site: approximately 10,000 pixels Figure 10 (1-1) is a graph plotting the BOT value calculated using LSFG on the vertical axis and the subject's age on the horizontal axis, and (1-2) It is a graph in which the BOT value calculated using the present method is plotted on the vertical axis and the age of the subject is plotted on the horizontal axis.
There was a negative correlation between the BOT value and age in the case of (1-1), and there was also a negative correlation between the BOT value and age in the case of (1-2).
FIG. 10(2) shows the correlation coefficient between each index calculated using LSFG and the subject's age, and the correlation coefficient between each index calculated using this method and the subject's age.
In the case of (2-1), the content is the same as that shown in FIG. 10(1), so the explanation will be omitted.
(2-2) shows the rising rate; in the case of LSFG, there is a weak negative correlation between the rising rate and age, and in the case of this method, there is a negative correlation between rising rate and age. there were.
(2-3) shows the falling rate; in the case of LSFG, there is a weak positive correlation between the falling rate and age, and in the case of this method, there is a positive correlation between the falling rate and age. there were.
(2-4) shows RI; in the case of LSFG, there was a positive correlation between RI and age, and in the case of this method, there was a weak positive correlation between RI and age.
From these facts, it can be said that the vascular elasticity index obtained by this method corresponds to the vascular elasticity index obtained using LSFG.

FIG. 11 is the same as the measurement conditions in FIG. 10 except for the measurement site, and is the result of analysis for two measurement sites: the forehead and the cheek. In (1), the analysis target is the forehead, and in (2), the analysis target is the cheek. In case (1), there was a negative correlation between BOT value and age, and in case (2), there was a negative correlation between BOT value and age. Moreover, the correlation coefficients of (1) and (2) were close values.
From these facts, it can be said that the vascular elasticity index can be calculated not only for the forehead but also for other parts of the face (for example, the cheeks).

As described with reference to FIG. 7, in this embodiment, when calculating the average one heartbeat waveform in the cycle calculation process, normalization processing is performed for the hemoglobin component amount and time (see steps S320 to S340). This is because a higher correlation coefficient can be obtained by performing the normalization process as described below than by not performing the normalization process.
FIG. 12 shows the calculation results of the BOT value with and without normalization processing under the same measurement conditions as in FIG. 10. In (1), the hemoglobin component amount and time are not normalized, and the BOT value is calculated using this method, and the BOT value is calculated using LSFG, and (2) is normalized for the hemoglobin component amount (intensity). (3) is a case where the BOT value is calculated using this method and a case where the BOT value is calculated using LSFG. It is a graph showing the correlation when the BOT value is calculated using. As is clear from (1) to (3), when normalization processing is performed, there is a strong correlation with the BOT value calculated using LSFG. Furthermore, as is clear from (2) and (3), there is a strong correlation with the BOT value calculated using LSFG, both when normalizing the hemoglobin component amount and when normalizing the time. It can be said that there is. For these reasons, in this method, the vascular elasticity index is calculated using the average one heartbeat waveform that has been normalized for hemoglobin amount and/or time.

<Vascular elasticity index calculation system>
The blood vessel elasticity index calculation system 200 will be explained using FIG. 13.
The vascular elasticity index calculation system 200 in this embodiment includes an image acquisition section 210, a hemoglobin amount calculation section 220, a period calculation section 230, and a vascular elasticity index calculation section 240. The information processing terminal is also equipped with an information processing terminal (not shown) capable of executing various processes, and the information processing terminal is equipped with an image acquisition section 210, a hemoglobin amount calculation section 220, a cycle calculation section 230, and a blood vessel elasticity index calculation section 240. ing. The information processing terminal includes an input device such as a keyboard and a pointing device, an arithmetic processing unit, a storage unit, and the like. Further, although it is preferable that the information processing terminal includes a display device, the display device may be provided outside the information processing terminal and connected via a network. Further, the information processing terminal may be equipped with a built-in or external camera (visible light camera).

The image acquisition unit 210 is a means for acquiring time-series images of a predetermined region of the body surface, and the time-series images may be acquired from any source, such as from a predetermined medium, via a network, or from an imaging device. . The image acquisition unit 210 also acquires visible light time-series images captured by a digital camera or a camera included in a mobile terminal.
The hemoglobin amount calculation unit 220 is a means for separating only the hemoglobin component from each frame of the acquired time-series image and calculating the hemoglobin amount of each pixel.
The period calculation unit 230 is a means for calculating one period of a heartbeat based on a change in the amount of hemoglobin.
The blood vessel elastic force index calculation unit 240 is a means for calculating a blood vessel elastic force index.
The processes performed by the image acquisition section 210, hemoglobin amount calculation section 220, cycle calculation section 230, and blood vessel elasticity index calculation section 240 are the same as those described above for the calculation method of the blood vessel elasticity index.

<Vascular elasticity index calculation device>
The blood vessel elastic force index calculating device 300 will be explained using FIG. 14.
The vascular elasticity index calculation device 300 in this embodiment includes a processor 301 and a memory 302, and the processor 301 includes an image acquisition section 310, a hemoglobin amount calculation section 320, a period calculation section 330, and a vascular elasticity index calculation section. 340 is included. The memory 302 stores data necessary for processing by the processor 301 (processing for calculating a blood vessel elastic force index), data necessary in the processing process, and the like. Although not shown, the blood vessel elasticity index calculating device 300 preferably includes an input section such as a keyboard or a pointing device, an input/output section for exchanging data with an external device, a display section, and the like. Further, the blood vessel elasticity index calculation device 300 may be equipped with a built-in or external camera (visible light camera).

The image acquisition unit 310 is a means (corresponding to image acquisition means) for acquiring time-series images of a predetermined region of the body surface, and is a means for acquiring time-series images of a predetermined region of the body surface, such as by acquiring from a predetermined medium, via a network, or from a photographing device. The source of the image does not matter. The image acquisition unit 310 also acquires visible light time-series images captured by a digital camera or a camera included in a mobile terminal.
The hemoglobin amount calculation unit 320 is a means (corresponding to a hemoglobin amount calculation means) that separates only the hemoglobin component from each frame of the acquired time-series image and calculates the hemoglobin amount of each pixel.
The period calculation unit 330 is a means (corresponding to a period calculation means) that calculates one period of a heartbeat based on a change in the amount of hemoglobin.
The vascular elasticity index calculation unit 340 is a means for calculating a vascular elasticity index (corresponding to a vascular elasticity index calculation means).
The processes performed by the image acquisition section 310, hemoglobin amount calculation section 320, cycle calculation section 330, and blood vessel elasticity index calculation section 340 are the same as those described above for the calculation method of the blood vessel elasticity index.

<Evaluation method of blood vessels and blood flow>
The method for evaluating blood vessels and blood flow includes an image acquisition step, a hemoglobin amount calculation step, a cycle calculation step, a blood vessel elasticity index calculation step, and an evaluation step. The image acquisition step, the hemoglobin amount calculation step, the cycle calculation step, and the blood vessel elasticity index calculation step are the same as those described above in the method for calculating the blood vessel elasticity index.
In the evaluation step, vascular elasticity or skin blood flow age is evaluated based on the calculated vascular elasticity index. In order to facilitate evaluation by the evaluator, it is preferable to display the calculated vascular elasticity index on a display device, such as a graph display or a comparison display with other subjects. FIG. 15 shows a display example of the calculated blood vessel elasticity index. As shown in Figure 15, the values of the vascular elasticity index (BOT in Figure 15) used in the evaluation are shown, and the calculated BOT values are compared with those of the same age as the subject. What position is the blood vessel elasticity index in Figure 15 located? What is the age of skin blood flow estimated based on the calculated BOT value (corresponds to the skin blood flow age in Figure 15)? Make it visible.
Generally, it is possible to understand the risk of lifestyle-related diseases by observing changes in the value of the vascular elasticity index. For example, blood vessel elastic force index values can be used to determine the inflow of blood into tissues, the continuity of blood flow to tissues, the outflow of blood from tissues, the variability of blood flow, etc. It becomes possible to understand the reaction of blood vessels to If the continuity of blood flow to the tissue is high, it can be assumed that the blood vessel is flexible and can expand and contract. If the continuity of blood flow to the tissue is low, it can be assumed that the blood vessel is hard and cannot expand or contract. Furthermore, since it is known that there is a correlation between the continuity of blood flow to tissues and age, it is also possible to estimate whether the blood vessel age is appropriate for the chronological age from the value of the blood vessel elastic force index. Therefore, by observing changes in the value of the vascular elasticity index, we can predict the risk of lifestyle-related diseases associated with deterioration of vascular conditions, provide lifestyle advice to improve vascular conditions, and improve vascular age. It can also be used to present lifestyle advice for people, and can be used to prevent lifestyle-related diseases and promote health.
In addition, by observing the condition of blood vessels and blood flow on the body surface (just under the skin), it is possible to understand (evaluate) the moisture condition of the skin, the elasticity of the skin, and the condition of the gums. It can also be used to present advice for improving conditions.

<Modified example>
In this embodiment, the hemoglobin component value is extracted from the RGB waveform (extracting the hemoglobin amount waveform) by performing independent component analysis on the RGB waveform and extracting only the hemoglobin component (see step S220). Not exclusively. For example, it is said that hemoglobin strongly absorbs green light with a wavelength of 540 nm, so blood flow can be monitored by using pixel information of the Green waveform of the RGB waveform. Therefore, as a process for extracting only the hemoglobin component, only the Green waveform may be extracted from the RGB waveform to extract the hemoglobin component (extracting the waveform of the amount of hemoglobin). Even in this case, the blood vessel elasticity index can be calculated in the same way as when the RGB waveform is subjected to independent component analysis and only the hemoglobin component is extracted.

In the present embodiment, after each heartbeat waveform is extracted, the normalization process (see step S320) is performed for each heartbeat waveform, but the present invention is not limited to this. For example, after performing normalization processing on the acquired hemoglobin component waveform, each heartbeat waveform may be extracted.

In the present embodiment, the waveform cut out for each heartbeat waveform is normalized to calculate the average one heartbeat waveform (see steps S320 to S340), but the present invention is not limited to this. For example, the average one heartbeat waveform may be calculated from the waveforms cut out for each heartbeat waveform, and then normalized. In this case, the time may be set to 0 at the beginning, the maximum cycle of the target heartbeat waveform may be set to 100, and the averaging may be performed.

Further, the means constituting the blood vessel elasticity index calculation system may be provided at different locations. Specifically, the vascular elasticity index calculation system is configured such that captured time-series image data can be acquired by an information processing terminal included in the vascular elasticity index calculation system 200 via a network line, medium, etc. Similarly, the calculation results may be acquired via a network line, medium, etc. by an information processing terminal (not shown) different from the information processing terminal included in the blood vessel elasticity index calculation system 200. . In this way, a user (subject) who is in a location different from the information processing terminal included in the vascular elasticity index calculation system 200 photographs a predetermined area of himself/herself with a smartphone or the like, and the vascular elasticity index calculation system 200 uses the photographed image. The information processing terminal calculates the vascular elastic force index, and the calculation results (including the evaluation results) are sent to another information processing terminal via a network line, medium, etc. good. In this way, the user can easily grasp the vascular elasticity index and check the evaluation results, making it easier to use for daily health management.

200 Blood vessel elasticity index calculation system 210 Image acquisition unit 220 Hemoglobin amount calculation unit 230 Cycle calculation unit 240 Blood vessel elasticity index calculation unit 300 Blood vessel elasticity index calculation device 301 Processor 302 Memory 310 Image acquisition unit 320 Hemoglobin amount calculation unit 330 Cycle calculation Section 340 Blood vessel elasticity index calculation section

Claims (7)

an image acquisition step of acquiring time-series images of a predetermined region of the body surface;
a hemoglobin amount calculation step of calculating the hemoglobin amount at each time in the time-series images;
a cycle calculation step of calculating one cycle of a heartbeat based on the calculated change in the amount of hemoglobin;
a vascular elasticity index calculation step of calculating a vascular elasticity index based on the change in the amount of hemoglobin and the calculated period;
Calculation method of vascular elastic force index, including. The period calculating step calculates a plurality of periods of heartbeat,
a waveform normalization step of normalizing a waveform indicating a change in hemoglobin amount over a plurality of cycles of the calculated heartbeat by the hemoglobin amount,
2. The method for calculating a blood vessel elastic force index according to claim 1, wherein one cycle of a heartbeat is calculated based on a normalized waveform. The period calculation step further includes an average waveform calculation step of averaging the normalized waveforms of the plurality of periods to obtain an average waveform of hemoglobin amount,
3. The method for calculating a vascular elasticity index according to claim 2, wherein the vascular elasticity index calculation step calculates the vascular elasticity index from an average waveform of the hemoglobin amount. further comprising a feature amount calculation step of calculating a feature amount related to a change in the hemoglobin amount from the hemoglobin amount and one cycle of the heartbeat,
The feature value calculation step includes:
a first time that is the half width of the maximum value of the hemoglobin amount;
a second time that is the entire length from the time when the hemoglobin amount rises to the time when it falls;
a third time from the time when the hemoglobin amount rises to the peak time when the hemoglobin amount reaches its maximum value;
A fourth time from the peak time to the fall time of the hemoglobin amount, and calculate any one or more of the following as the feature quantity,
The vascular elastic force index calculating step is characterized in that the vascular elastic force index is calculated using any one or more of the first time, the second time, the third time, or the fourth time. The method for calculating a blood vessel elastic force index according to any one of claims 1 to 3. further comprising a feature amount calculation step of calculating a feature amount related to a change in the hemoglobin amount from the hemoglobin amount and one cycle of the heartbeat,
The feature value calculation step includes:
the maximum value of the hemoglobin amount; the minimum value of the hemoglobin amount;
Calculating any one or more of the maximum increase amount of the increase amount of the hemoglobin amount per predetermined time as the feature amount,
The vascular elasticity index calculating step is characterized in that the vascular elasticity index is calculated using any one or more of the maximum value of the hemoglobin amount, the minimum value of the hemoglobin amount, or the maximum increase in the hemoglobin amount. The method for calculating a blood vessel elastic force index according to any one of claims 1 to 3. an image acquisition unit that acquires time-series images of a predetermined region of the body surface;
a hemoglobin amount calculation unit that calculates the hemoglobin amount at each time in the time-series images;
a cycle calculation unit that calculates one cycle of a heartbeat based on the calculated change in the amount of hemoglobin;
a vascular elasticity index calculation unit that calculates a vascular elasticity index based on the change in the amount of hemoglobin;
A blood vessel elastic force index calculation system, including: comprising at least a processor and a memory;
The processor includes:
image acquisition means for acquiring time-series images of a predetermined region of the body surface;
Hemoglobin amount calculation means for calculating the hemoglobin amount at each time in the time-series images;
a cycle calculation means for calculating one cycle of a heartbeat based on the calculated change in the amount of hemoglobin;
Blood vessel elasticity index calculation means for calculating a blood vessel elasticity index based on the change in the amount of hemoglobin;
A blood vessel elastic force index calculation device, including:

Images