Determination device, determination method, and determination program

Abstract

Provided are a determination device, a determination method, and a determination program whereby whether a test subject has a health condition with risks can be easily measured.　The determination device according to one embodiment comprises: an image acquisition unit that acquires an image of a fingernail of the test subject; a determination unit that determines whether the test subject has a health condition with risks on the basis of information pertaining to the appearance of the fingernail obtained from the image acquired by the image acquisition unit and information acquired from changes in images acquired in chronological order; and an output unit that outputs a determination result from the determination unit.

Description

[Technical Field]

[0001] The present invention relates to an apparatus, a method, and a program for determining the health condition of a subject. [Background technology]

[0002] Because nails lack melanin and have little individual variation like skin, they are characterized by their ease of observing health conditions and the ease of monitoring the immediate blood conditions under the nails. Because nails are formed over a relatively long period of time, such as several months, it is believed that nails reflect a person's health over a relatively long period of time, such as when their daily diet is reflected in the unevenness and ridges of the nails. Therefore, the shape and color of the nails can reveal a chronic health condition, which is a health condition that lasts for a relatively long period of time. This is a health condition that is unlikely to change suddenly in a relatively short period of time. In contrast, the blood condition under the nails is believed to reflect the blood condition over a relatively short period of time, such as a few hours to a few days. Therefore, the blood condition under the nails can reveal an acute health condition, which is a health condition that is likely to change suddenly in a relatively short period of time. Therefore, by observing the condition of the nails, both acute and chronic health conditions can be read.

[0003] Thus, as a device for determining health status from the condition of the nail, a fingernail pressure blood flow return measurement device has been known (for example, Patent Document 1), which determines and manages health status by pressing the fingernail to block blood flow to the extremities and then measuring the time it takes for the fingernail to return to its pre-pressure state. The fingernail pressure blood flow return measurement device described in Patent Document 1 automatically measures the time it takes for the pressed fingernail to return to its pre-pressure state stored in a memory unit, thereby accurately measuring the return of blood flow in a constant state without variability between users, allowing for relatively accurate determination and management of health status. The fingernail pressure blood flow return measurement device described in Patent Document 1 is considered to be observing acute health status.

[0004] Furthermore, a roll call support system is known as a system for determining the health condition of a person being called, which performs a health condition assessment of the person being called based on the results of a nail reader that reads the condition of the person's nails and displays the assessment results on a display unit (see, for example, Patent Document 2). According to the roll call support system described in Patent Document 2, the assessment results are displayed on a display device during roll call support processing, eliminating the need for the roll call executor to assess the health condition of the person being called based on the results of the reading device, making it easier for the roll call executor to recognize the health condition of the person being called than in the past. In the roll call support system described in Patent Document 2, the health condition case shown in Figure 4 is an example of a chronic health condition assessment. Furthermore, while examples of myocardial infarction and cerebral infarction assessment are also shown, the explanations and descriptions of nail color used for assessment suggest that these are stages that may lead to these conditions, i.e., a chronic health condition, are inferred based solely on steady-state nail color. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 10-057329 [Patent Document 2] Japanese Patent Application Laid-Open No. 2018-005268 Summary of the Invention [Problem to be solved by the invention]

[0006] As disclosed in Patent Documents 1 and 2, there are devices that monitor either acute or chronic health conditions. However, for example, if a subject has no subjective symptoms, the subject may not know which device to use. Furthermore, since the methods of dealing with these two health conditions are significantly different, a device that can easily monitor both health conditions with one device is needed for daily use, taking into consideration ease of use and ease of dealing with them.

[0007] An object of the present disclosure is to provide a determination device, a determination method, and a determination program that enable easy determination of whether a subject is in a health state that poses a risk. [Means for solving the problem]

[0008] A determination device according to one embodiment of the present disclosure is characterized by having an image acquisition unit that acquires images of the subject's nails, a determination unit that determines whether the subject is in a health condition that poses a risk based on information about the appearance of the nails from the images acquired by the image acquisition unit and information acquired from changes in the images acquired over time, and an output unit that outputs the determination result by the determination unit.

[0009] In an embodiment of the determination device of the present disclosure, the determination unit may include a first factor determination unit that determines, based on a change in an image, whether the subject's health condition is in a state that poses a risk based on a predetermined first factor, and a second factor determination unit that determines, based on information regarding the appearance of the nail, whether the subject's health condition is in a state that poses a risk based on a predetermined second factor that includes a factor different from the first factor.

[0010] In the determination device according to an embodiment of the present disclosure, the first factor may be an acute factor, and the second factor may be a chronic factor.

[0011] In an embodiment of the determination device of the present disclosure, the first factor determination unit may use the pulse wave detected from changes in the image to determine whether the subject's health condition is at risk due to an acute factor, based on the time it takes for the amplitude of the pulse wave, which has decreased due to pressure on the finger, to return to its pre-pressure state after the pressure on the finger is released.

[0012] In an embodiment of the determination device of the present disclosure, the first factor determination unit may measure the period of the pulse wave detected from a change in the image, and determine whether the subject's health condition is at risk due to an acute factor based on the measured pulse wave period.

[0013] In an assessment device according to one embodiment of the present disclosure, a first factor assessment unit may assess whether the subject's health condition is in a state that poses a risk due to an acute factor, and a second factor assessment unit may assess whether the subject's health condition is in a state that poses a risk due to a chronic factor, based on an image acquired during a single measurement operation.

[0014] In a determination device according to one embodiment of the present disclosure, the first factor determination unit determines whether the subject's health condition is at risk due to an acute factor based on a pulse wave detected from changes in images of multiple images acquired over time, and after the first factor determination unit performs a determination operation of whether the subject's health condition is at risk due to an acute factor, the second factor determination unit performs a determination operation of whether the subject's health condition is at risk due to a chronic factor.

[0015] In a determination device according to one embodiment of the present disclosure, the first factor determination unit determines whether the subject's health condition is at risk due to an acute factor based on a comparison of each changing image with a reference image, and after the second factor determination unit performs a determination operation to determine whether the subject's health condition is at risk due to a chronic factor, the first factor determination unit performs a determination operation to determine whether the subject's health condition is at risk due to an acute factor.

[0016] In a determination device according to one embodiment of the present disclosure, when the first factor determination unit determines that the subject's health condition is in a state that poses a risk due to an acute factor, the output unit may output the determination result in a manner different from that when the subject's health condition is determined to be in a state that poses a risk due to a chronic factor.

[0017] In a determination device according to an embodiment of the present disclosure, when a single measurement operation is used to determine whether a subject's health condition is at risk due to acute factors and whether the subject's health condition is at risk due to chronic factors, the image used to determine whether the subject's health condition is at risk due to chronic factors may be an image acquired before finger compression for determining whether the subject's health condition is at risk due to acute factors, or an image acquired after blood flow has recovered from finger compression.

[0018] In a determination device according to an embodiment of the present disclosure, the image acquired after blood flow recovery from finger compression may be an image acquired after blood flow recovery is detected, or an image acquired after a predetermined time has elapsed since finger compression.

[0019] In an assessment device according to one embodiment of the present disclosure, the image acquisition unit acquires an image of the nail by outputting an instruction to the subject to capture an image of the nail, and the output unit may display a guide that specifies how the nail should appear in the image of the nail.

[0020] A determination method according to one embodiment of the present disclosure is characterized in that it includes: an image acquisition unit acquiring images of the subject's nails; a processor determining whether the subject is in a health condition that puts the subject at risk based on information about the appearance of the nails from the images acquired by the image acquisition unit and information acquired from changes in the images acquired over time; and outputting a determination result as to whether the subject is in a health condition that puts the subject at risk.

[0021] A judgment program according to one embodiment of the present disclosure is characterized in that it causes a computer to execute the following steps: an image acquisition unit acquires images of the subject's nails; a judgment unit determines whether the subject is in a health condition that poses a risk based on information regarding the appearance of the nails from the images acquired by the image acquisition unit and information acquired from changes in the images acquired over time; and an output unit outputs a judgment result as to whether the subject is in a health condition that poses a risk. [Effects of the Invention]

[0022] A determination device, a determination method, and a determination program according to an embodiment of the present disclosure enable a test subject to easily determine whether or not the subject is in a health state that poses a risk. [Brief explanation of the drawings]

[0023] [Figure 1] FIG. 1 is a schematic diagram illustrating a state in which a nail condition is observed using a determination device according to an embodiment of the present disclosure. [Figure 2] FIG. 10 is a diagram showing an example in which an image of a nail is captured using the determination device of embodiment 1, and the determination results of whether the subject's health condition is at risk due to acute factors or not, and whether the subject's health condition is at risk due to chronic factors or not are displayed. [Figure 3] 1 is a block diagram showing the configuration of a determination device according to a first embodiment. [Figure 4] 1A and 1B are graphs showing an example of determining whether a subject is dehydrated or not from a pulse wave acquired from an image of a nail using a determination device according to embodiment 1, where (A) shows an example of a case where the subject is not dehydrated, and (B) shows an example of a case where the subject is dehydrated. [Figure 5] 1 is a flowchart illustrating a procedure for determining, from a nail image, whether a subject's health condition is at risk due to acute factors, and whether the subject's health condition is at risk due to chronic factors, using a determination device according to embodiment 1. [Figure 6] FIG. 1 is a sequence diagram illustrating the procedure for determining whether or not the subject's health condition is at risk due to an acute factor, and whether or not the subject's health condition is at risk due to a chronic factor, by downloading a program from the server side to the determination device and executing the program on the determination device side in the determination device according to the first embodiment. [Figure 7] FIG. 10 is a diagram showing an example of an imaging screen including a guide, which is displayed on the output unit 4. [Figure 8] 10 is a graph showing an example of determining whether a subject has tachycardia or bradycardia from a pulse wave detected from changes in an image of a nail. [Figure 9] FIG. 10 is a diagram showing an example of a screen displayed on the output unit 4 when it is determined that the subject has tachycardia or bradycardia. [Figure 10] FIG. 10 is a block diagram showing the configuration of a determination device according to a second embodiment. [Figure 11] 10(A) to 10(E) are examples of images of a nail captured using the determination device according to the second embodiment, showing an example of a series of steps from an anemic state to a healthy state. [Figure 12] 10 is a flowchart illustrating a procedure for calculating a response time required for returning from an anemic state to a healthy state from a captured image of a nail using the determination device according to the second embodiment. [Figure 13] 10 is a flowchart illustrating the steps from downloading a measurement program using a determination device according to embodiment 2 to outputting a determination result indicating whether the subject's health condition is at risk due to acute factors, and whether the subject's health condition is at risk due to chronic factors. [Figure 14] FIG. 10 is a block diagram showing the configuration of a determination device according to a third embodiment. [Figure 15] FIG. 10 is a block diagram showing the configuration of a determination device according to a fourth embodiment. [Figure 16] FIG. 10 is a sequence diagram illustrating the procedure for uploading a nail image captured by a terminal to a server side, which is the determination device, in a determination device according to embodiment 4, and executing a program in the determination device to determine whether the subject's health condition is at risk due to acute factors, and whether the subject's health condition is at risk due to chronic factors. [Figure 17] FIG. 10 is a block diagram showing the configuration of a determination device according to a fifth embodiment. [Figure 18] 13 is a diagram showing an example of display of an acute mode execution button and a chronic mode execution button in the determination device according to the fifth embodiment. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0024] Hereinafter, a determination device, a determination method, and a determination program according to an embodiment of the present disclosure will be described with reference to the drawings. However, it should be noted that the technical scope of the present invention is not limited to these embodiments, but extends to the inventions set forth in the claims and their equivalents.

[0025] FIG. 1 is a schematic diagram showing a state in which a determination device 101 according to an embodiment of the present disclosure observes the state of a nail 201. The determination device 101, which is, for example, a smartphone or the like, is fixed to a stand 301. A subject holds a hand 200 in front of a camera (not shown) of the determination device 101. The determination device 101 captures an image of the nail 201 and displays the image of the nail 201 on the image display unit 41 of the output unit 4. The determination device 101 determines the health state of the subject from the captured image of the nail 201 and displays the determination result on the determination result display unit 42.

[0026] Here, factors that affect a subject's health condition can be broadly divided into acute factors and chronic factors. "Acute factors" are immediate, short-term factors that affect a subject's health condition. A subject with acute factors is prone to sudden changes in health condition over a relatively short period of time, and acute factors are factors that require emergency treatment. "Chronic factors" are long-term factors that affect a subject's health condition. A subject with chronic factors is prone to gradual changes in health condition over a relatively long period of time, and chronic factors are factors that do not require emergency treatment compared to acute factors. For example, if an elderly person who has difficulty accurately communicating their symptoms exhibits a certain symptom, the cause of the symptom may be an acute or chronic factor, and the method of dealing with it will differ depending on whether it is an acute or chronic factor.

[0027] For example, if a subject's symptoms are "dizziness and fainting," possible acute causes include dehydration, i.e., increased blood viscosity, while possible chronic causes include anemia (cerebral anemia), fatigue (or autonomic nervous system disorders), or high blood pressure (or diabetes), each of which requires different treatment methods. Therefore, it is important to accurately determine whether the subject's current health condition is at risk due to acute factors or chronic factors, and to respond appropriately. It is also possible that increased blood viscosity, an acute factor, is caused by red blood cell abnormalities.

[0028] Nails are suitable sites for observing whether a subject's health condition is at risk due to acute factors or whether the subject's health condition is at risk due to chronic factors. This is because nails contain almost no melanin, and therefore blood flow conditions can be easily observed from the blood flow beneath the nail to determine whether the subject's health condition is at risk due to acute factors without being affected by the condition of the skin. Furthermore, because steady-state appearance characteristics of nails, such as unevenness and streaks, reflect chronic factors over a period of several weeks, observing the appearance of the nails can also determine whether the subject's health condition is at risk due to chronic factors. Therefore, it is possible to determine both whether the subject's health condition is at risk due to acute factors and whether the subject's health condition is at risk due to chronic factors from images of the nails.

[0029] One method for determining whether a subject's health condition is at risk from acute factors based on nail images is to determine whether the subject's health is based on the reversion of nail color. For example, when a finger is compressed, the blood vessels under the nail are compressed, reducing blood flow, causing the nail color to change from pink to white. When the pressure on the finger is released, blood flow returns to normal, causing the nail color to return to its original pink. If the subject is not dehydrated, their blood is sufficiently hydrated and therefore smooth, so blood flow is expected to return to normal quickly when the pressure on the finger is released. On the other hand, if the subject is dehydrated, their blood is thick due to a lack of water in their blood. Therefore, the time it takes for blood flow to return to normal is expected to be longer the lower the blood moisture content. Therefore, by detecting the "reversion time" (time required for blood flow to return to the state before pressure was applied) after pressure on the finger is released, the viscosity of the blood can be determined, allowing the presence or absence of dehydration to be determined. In practice, quantitative evaluation of nail color is difficult, so it is preferable to detect pulse waves that fluctuate in response to changes in nail color, as described below. Because the pulse wave is less affected by the external characteristics of the nail, it is possible to determine whether the subject's health condition is at risk due to acute factors based on the pulse wave, regardless of whether the subject's health condition is at risk due to chronic factors.

[0030] Another method for determining whether a subject's health condition is at risk due to chronic factors from nail images is image assessment using artificial intelligence (AI). This method uses a trained model that has previously learned the characteristics of normal and abnormal nails from images of nails in various disease states using deep learning or other methods. If a subject is healthy, their nails will be pink. However, if the subject is anemic, for example, they will be white, and if they have heart disease, for example, they will be purple. Furthermore, vertical lines may appear on the nails due to aging, and horizontal lines may appear on the nails reflecting the subject's past poor health. In this way, it is possible to determine whether a subject's health condition is at risk due to chronic factors from images of the subject's nails.

[0031] FIG. 1 illustrates an example in which the determination device 101 is a smartphone, but is not limited to this example. The determination device 101 may also be a mobile terminal such as a laptop computer, a desktop computer, or the like. The imaging unit, which is a camera that captures images of the nails, may be built into these devices or may be external. FIG. 1 also illustrates an example in which the determination device 101 is fixed to a stand 301, but is not limited to this example. The subject may hold the determination device 101 with the hand opposite the hand 200 whose nails are to be imaged. FIG. 1 also illustrates an example in which the determination device 101 captures an image of the thumb nail 201 of the hand 200, but is not limited to this example. The determination device 101 may also capture images of the index finger nail 202, middle finger nail 203, ring finger nail 204, or little finger nail 205.

[0032] [Embodiment 1] (Configuration of the determination device) FIG. 2 is a diagram showing an example in which the determination device 101 according to the first embodiment captures an image of a nail 201 and displays the determination results of whether or not the subject's health state is a state in which there is a risk due to an acute factor and whether or not the subject's health state is a state in which there is a risk due to a chronic factor. FIG. 3 is a block diagram showing the configuration of the determination device 101 according to the first embodiment. The determination device 101 according to the first embodiment includes a control unit 10 and an output unit 4, which are connected by a bus 7. The control unit 10 includes an image acquisition unit 1, an acute factor determination unit 2, and a chronic factor determination unit 3. The acute factor determination unit 2 is an example of a first factor determination unit and a determination unit, and the chronic factor determination unit 3 is an example of a second factor determination unit and a determination unit. In the following description, an example will be described in which the first factor determination unit is the acute factor determination unit 2 and the second factor determination unit is the chronic factor determination unit 3. The determination device 101 may further include an imaging unit 11, a storage unit 5, and a communication unit 6. The image acquisition unit 1, the acute factor determination unit 2, and the chronic factor determination unit 3 are functional modules realized by firmware or software executed by a control unit 10 such as a processor (computer).

[0033] The imaging unit 11 captures an image of the subject's nail 201. The imaging unit 11 is a camera, a video camera, or the like. The image acquisition unit 1 acquires the nail image captured by the imaging unit 11. The image acquisition unit 1 may acquire the nail image from outside the determination device 101. In this case, the image acquisition unit 1 acquires the nail image from outside the determination device 101 via the communication unit 6. The nail image may include a moving image of the nail.

[0034] The acute factor determination unit 2, which is a first factor determination unit, determines whether the subject's health state poses a risk due to a predetermined first factor, based on the images acquired by the image acquisition unit 1. The acute factor determination unit 2 determines whether the subject's health state poses a risk due to a predetermined first factor, based on information acquired from changes in the images acquired in time series by the image acquisition unit 1. The first factor may be an acute factor. The acute factor determination unit 2 includes a nail pulse wave response detection unit 21 and a pulse wave response time determination unit 22. A method for determining whether the subject's health state poses a risk due to an acute factor will be described later.

[0035] The chronic factor determination unit 3, which is the second factor determination unit, determines whether the subject has a health condition that poses a risk due to a second factor different from the first factor, from the image acquired by the image acquisition unit 1. The chronic factor determination unit 3 determines whether the subject's health condition is a condition that poses a risk due to a predetermined second factor that includes a factor different from the first factor, based on a still image included in the image acquired by the image acquisition unit 1. The second factor may be a chronic factor. The chronic factor determination unit 3 has a nail appearance case determination unit 31. A method for determining whether the subject's health condition is a condition that poses a risk due to a chronic factor will be described later. In the following explanation, an example will be described in which the first factor is an acute factor and the second factor is a chronic factor. The still image included in the image acquired by the image acquisition unit 1 is an example of information related to nail appearance.

[0036] The control unit 10 further includes an image recognition unit 12 and a measurement correctness determination unit 13. The image recognition unit 12 determines whether or not an image of a nail is included in the image acquired by the image acquisition unit 1 through image recognition using an image recognition trained model. However, if it is possible to determine whether or not an image of a nail is included in the image acquired by the image acquisition unit 1 without using an image recognition trained model, the image recognition unit 12 may be omitted. For example, if a mark is attached to a finger in advance and the position of the finger can be detected by detecting this mark, the image recognition unit 12 may be omitted.

[0037] The measurement appropriateness determination unit 13 determines whether the captured image of the nail is appropriate for measurement. For example, when a finger is placed within the frame of the imaging unit 11, which is a camera, the measurement appropriateness determination unit 13 may simply determine whether the captured image is appropriate for measurement. When a camera is used, strong specular reflection may cause the image to appear white, making it impossible to capture an image appropriate for measurement. Therefore, the measurement appropriateness determination unit 13 first determines whether the captured image of the nail is appropriate based on the presence or absence of halation, etc. Furthermore, when a pulse wave is detected from the image as described below, the measurement appropriateness determination unit 13 may simply detect the pulse wave by image recognition and determine whether it is possible to detect the pulse wave from the captured image.

[0038] The image recognition unit 12 and the measurement correctness determination unit 13 are functional modules realized by firmware or software executed by the control unit 10, which is a processor (computer) or the like.

[0039] The output unit 4 outputs the determination result by the acute factor determination unit 2, which is the first factor determination unit, and the determination result by the chronic factor determination unit 3, which is the second factor determination unit, i.e., whether or not the subject is in a health state that poses a risk. The output unit 4 is a display device such as a liquid crystal display device or an organic electroluminescence (EL) display device. The output unit 4 displays an image based on display data supplied from the control unit 10. The output unit 4 includes an image display unit 41 and a determination result display unit 42. For example, as shown in FIG. 2, the output unit 4 may display icons such as "Normal" 401, "Slightly Dehydrated" 402, and "Dehydrated" 403 on the determination result display unit 42 as the determination results of whether or not the subject's health state poses a risk due to an acute factor. The output unit 4 may also display icons corresponding to the determination results in a more prominent manner than other icons. The output unit 4 may also display bars 421 to 424 indicating the degrees of "Aging," "Anemia," "Fatigue," and "Health" on the determination result display unit 42 as the determination results of whether or not the subject's health state poses a risk due to a chronic factor. The lengths of the hatched portions of the bars 421 to 424 indicate the degrees of "aging," "anemia," "fatigue," and "health." For example, as shown in FIG. 2, when vertical lines 201a are observed on the nail 201, aging is presumed to be the cause. In this case, the output unit 4 may display the degree of aging by adjusting the lengths of the hatched portions of the bars 421 to 424 according to the degree of the vertical lines.

[0040] The output unit 4 may include an audio output unit (not shown). The audio output unit outputs audio based on audio data supplied from the control unit 10. For example, if the determination result of whether the subject's health condition is a state with a risk due to acute factors is "dehydration," the output unit 4 may output audio advice to improve the dehydration state, such as "Do not exercise. Drink fluids immediately." Alternatively, if the determination result of whether the subject's health condition is a state with a risk due to chronic factors is "anemic" and the certainty of this is 80% or higher, the output unit 4 may output audio advice to improve the anemia state, such as "Be conscious of iron in your daily diet." However, the present invention is not limited to these examples. The output unit 4 may output various audio advice to improve these factors depending on the respective determination results of whether the subject's health condition is a state with a risk due to acute factors and whether the subject's health condition is a state with a risk due to chronic factors. Alternatively, the output unit 4 may display various pieces of advice on the determination result display unit 42 in the form of text or the like.

[0041] The memory unit 5 includes an image memory unit 51 and a trained model memory unit 52. The memory unit 5 is a storage device such as a hard disk or a semiconductor memory. The image memory unit 51 can store images captured by the imaging unit 11. The trained model memory unit 52 can store a trained model for determining whether the subject's health state is a state in which there is a risk due to an acute factor or whether the subject's health state is a state in which there is a risk due to a chronic factor. The trained model memory unit 52 may store a trained model downloaded from an external device each time a measurement is performed, or may store a trained model in advance before a measurement. The memory unit 5 can also store a program for determining whether the subject's health state is a state in which there is a risk due to an acute factor and a program for determining whether the subject's health state is a state in which there is a risk due to a chronic factor.

[0042] (Determining whether a subject's health condition is at risk due to acute factors using pulse waves) The determination device 101 according to the first embodiment determines whether the subject's health condition is at risk due to an acute factor, using a pulse wave detected from an image of the subject's nail. Specifically, the acute factor determination unit 2 uses the subject's pulse wave detected from changes in the image. The acute factor determination unit 2 determines whether the subject's health condition is at risk due to an acute factor, based on the time it takes for the amplitude of the pulse wave, which has decreased due to pressure on the finger, to return to its pre-pressure state after the pressure on the finger is released. Here, the "pre-pressure state" does not necessarily mean that the pulse wave amplitude has completely returned to its pre-pressure state, but also includes a case in which the pulse wave amplitude can be considered to have almost returned to its pre-pressure state. Specifically, the "pre-pressure state" does not necessarily mean that the pulse wave amplitude has returned 100% to its pre-pressure state, but includes, for example, a case in which the pulse wave amplitude has returned to 90% of its pre-pressure state. FIGS. 4A and 4B are graphs illustrating an example of the determination device 101 according to the first embodiment determining whether the subject is dehydrated based on the pulse wave detected from an image of the nail. FIG. 4(A) is a diagram showing an example in which the subject is not dehydrated, and FIG. 4(B) is a diagram showing an example in which the subject is dehydrated.

[0043] First, using FIG. 4(A), an example will be described in which the determination device 101 determines that the subject is not dehydrated based on the pulse wave. At time t1, the nail pulse wave response detection unit 21 starts detecting the pulse wave w1. The pulse wave can be detected from the temporal change in the green (G) signal among the RGB signals of the image of a predetermined area R of the nail (see FIG. 2). This is based on the fact that the G signal intensity changes depending on the amount of hemoglobin contained in the blood flowing through the artery.

[0044] At time t2, the subject blocks blood flow to the finger. For example, by pressing the finger against a fixed object such as a desk, blood flow under the nail is blocked by the pressure. Therefore, the period from time t1 to t2 (e.g., 3 seconds) is the state before the finger is compressed, and the average value ca1 of the envelope of the peak value of the pulse wave detected by nail pulse wave response detection unit 21 during this period is defined as the "return reference value."

[0045] At time t2, the output unit 4 outputs a voice or text message urging the subject to press their finger, such as "Please press your finger." The subject presses their finger in response to this message. Pressing the finger blocks blood flow to the finger, reducing the amplitude of the pulse wave. In the example shown in Figure 4(A), the amplitude increases immediately after the finger is pressed. This is thought to be due to the so-called water hammer phenomenon.

[0046] Next, at time t3, output unit 4 outputs an announcement by voice or text, such as "Please release your finger," encouraging the subject to release the pressure on the finger. In response to this announcement, the subject releases the pressure on the finger. Therefore, the period from time t2 to t3 is the period during which the finger is compressed, and this period is, for example, 10 seconds. At this time, the average value cv1 of the envelope of the peak values ​​of the pulse wave detected by nail pulse wave response detection unit 21 during the three seconds immediately prior to time t3, when the finger is released, is defined as the "average peak value during compression."

[0047] After an announcement encouraging the subject to release their finger, the pulse wave peak value increases when the subject actually releases their finger. The pulse wave response time determination unit 22 determines the intersection of the peak value envelope at that time with the line representing cv1', which is a 20% increase in the average peak value during compression, as the "return start point" ts11. Next, the pulse wave response time determination unit 22 determines the time from ts11 to the "return completion point" ts12, at which the peak value envelope reaches the line representing the average pre-compression peak value ca1, as the "return time" Ts1. Figure 4(A) shows an example in which the subject is not dehydrated, so the return time Ts1 is short, for example, within 2 seconds.

[0048] Figure 4(B) shows an example of a pulse wave when the subject is dehydrated. As described above, using pulse wave w2, the average value cv2 of the envelope of the pulse wave peak values ​​detected by nail pulse wave response detection unit 21 during the three seconds immediately preceding time t3, when the finger is released, is defined as the "average compression peak value." After an announcement encouraging the subject to release their finger is made, pulse wave response time determination unit 22 determines the intersection of the peak value envelope at the time the subject actually releases their finger with the line representing cv2', which is a 20% increase from the average compression peak value, as the "return start point" ts21. Pulse wave response time determination unit 22 measures the time from ts21 to the "return completion point" ts22, at which the peak value envelope reaches the line representing the average pre-compression peak value ca2, as the "return time" Ts2. Because Figure 4(B) shows an example when the subject is dehydrated, the return time Ts2 is long, e.g., 3 seconds or more.

[0049] As described above, the acute factor determination unit 2 measures the return time to determine whether the subject's health state is dehydrated, which is an example of a state that poses a risk due to an acute factor. For example, if the return time is less than 2 seconds, the acute factor determination unit 2 determines that the subject is in a "normal" state, not dehydrated. If the return time is between 2 and 3 seconds, the acute factor determination unit 2 determines that the subject is in a "slightly dehydrated" state. If the return time is 3 seconds or more, the acute factor determination unit 2 determines that the subject is in a "dehydrated" state. However, the above-mentioned reference times for determining whether the subject is dehydrated are merely examples, and are not limited to these values.

[0050] (Operation procedure) Next, the operation procedure of the determination device 101 according to the first embodiment will be described. In the determination device 101 according to the first embodiment, the acute factor determination unit 2 determines whether or not the subject's health state is a state in which there is a risk due to an acute factor, and the chronic factor determination unit 3 determines whether or not the subject's health state is a state in which there is a risk due to a chronic factor, based on an image acquired during one measurement operation. FIG. 5 shows a flowchart for explaining the procedure by which the determination device 101 according to the first embodiment determines, from a nail image, whether or not the subject's health state is a state in which there is a risk due to an acute factor, and whether or not the subject's health state is a state in which there is a risk due to a chronic factor. In the flowchart shown in FIG. 5, one measurement operation is executed from step S104 to step S112.

[0051] First, in step S101, the determination device 101, which is a smartphone or the like, accesses a predetermined website. Next, in step S102, the determination device 101 downloads an acute factor determination program from a predetermined server via the predetermined website. Furthermore, in step S103, which is performed in parallel with step S102, the determination device 101 starts downloading a chronic factor determination program and a trained model for determining whether the subject's health condition is a chronic factor-related risk state from the predetermined server via the predetermined website. The trained model for determining whether the subject's health condition is a chronic factor-related risk state has a large file size and may take a long time to download. Therefore, it is preferable to complete downloading the trained model while the determination device 101 is determining whether the subject's health condition is an acute factor-related risk state. In this case, the determination device 101 determines whether the subject's health condition is an acute factor-related risk state in the first half of the operation procedure, and determines whether the subject's health condition is a chronic factor-related risk state in the second half of the operation procedure. This reduces the total determination time, which is the sum of the time required to determine whether the subject's health condition is at risk due to acute factors and the time required to determine whether the subject's health condition is at risk due to chronic factors.

[0052] Next, in step S104, the determination device 101 starts determining whether the subject's health condition is at risk due to an acute factor. The determination device 101 activates the imaging unit 11, which is a camera. Next, in step S105, the determination device 101 waits for a predetermined waiting time. The waiting time is, for example, about 5 seconds. This waiting time is set to ensure that the nail deformation caused by pressure on the finger is reliably restored to its original state when measurements are performed multiple times in succession.

[0053] Next, in step S106, the determination device 101 instructs the subject to capture an image of the nail and issues a message encouraging the subject to compress and release the finger. When the subject holds their nail over the smartphone camera and the determination device 101 recognizes the nail, the imaging unit 11 captures images of the nail over time, for example, at a frame rate of 30 fps. The image acquisition unit 1 acquires the nail images from the imaging unit in time series. The nail pulse wave response detection unit 21 detects the intensity of the green signal contained in each image. The nail pulse wave response detection unit 21 then detects the pulse wave from the temporal change in the green signal that accompanies changes in each image acquired in time series. When the imaging unit 11 captures nail images at a frame rate of 30 fps, the nail pulse wave response detection unit 21 detects the intensity of the green signal at 33 msec intervals and detects the pulse wave based on the intensity of the green signal that changes every 33 msec. Next, in step S107, pulse wave response time determination unit 22 measures return time Ts from the detected pulse wave. The method for measuring return time from a pulse wave is as described above. Whether or not the subject's health state is at risk due to an acute factor is determined from the magnitude of the return time. In this way, pulse wave response time determination unit 22 generates a determination result as to whether or not the subject's health state is at risk due to an acute factor.

[0054] Next, in step S108, the determination device 101 completes the download of the chronic factor determination program and the trained model. Next, in step S109, the determination device 101 waits for a predetermined waiting time. The waiting time is, for example, approximately 5 seconds. This waiting time ensures that the nail deformation caused by pressure on the subject's finger is reliably restored to its original state when a measurement is performed to determine whether the subject's health state is a risk state based on an acute factor, followed by a measurement to determine whether the subject's health state is a risk state based on a chronic factor.

[0055] Next, in step S110, the determination device 101 instructs the subject to capture an image of the nail again. The image captured here is an image taken after the pressure on the finger is released and blood flow has recovered. This image is an image taken after blood flow recovery is detected or an image taken a predetermined time after the finger was compressed. That is, in step S107, the pulse wave response time determination unit 22 measures the return time Ts and confirms that the pulse wave has returned to normal. Therefore, the image captured in step S110 can be said to be an image taken after blood flow recovery is detected. Furthermore, because a wait time is set in step S109 to ensure that blood flow has returned to normal, the image captured in step S110 is an image taken a predetermined time after the finger was compressed. That is, in step S106, the subject's finger is released from compression, and further, because a wait time is set in step S109, a predetermined time has passed since the finger was compressed by step S110. Therefore, the image captured at this time is guaranteed to be an image in a state where blood flow has returned to normal. The predetermined time from when the finger is pressed may be preset as a fixed time to suit people who require a long time for the pulse wave to return to normal after the finger is pressed.

[0056] Next, in step S111, the nail appearance case determination unit 31 uses the downloaded trained model to determine whether the subject's health condition is a state in which there is a risk due to a chronic factor. The nail appearance case determination unit 31 generates a determination result as to whether the subject's health condition is a state in which there is a risk due to a chronic factor. Next, in step S112, the output unit 4 displays the determination results as to whether the subject's health condition is a state in which there is a risk due to an acute factor and whether the subject's health condition is a state in which there is a risk due to a chronic factor, and advice according to the determination results, on the determination result display unit 42. The advice according to the determination results may be output by voice.

[0057] Next, in step S113, the determination device 101 ends the operation of the imaging unit 11. This completes the series of operation procedures.

[0058] As described above, in the determination device 101 according to the first embodiment, the acute factor determination unit 2 determines whether the subject's health condition is a state in which there is a risk due to an acute factor, based on a pulse wave detected from changes in images obtained over time from a plurality of images. After the acute factor determination unit 2 determines whether the subject's health condition is a state in which there is a risk due to an acute factor, the chronic factor determination unit 3 determines whether the subject's health condition is a state in which there is a risk due to a chronic factor. The determination device according to the first embodiment can easily determine both whether the subject's health condition is a state in which there is a risk due to an acute factor and whether the subject's health condition is a state in which there is a risk due to a chronic factor, based on images of the subject's nails.

[0059] (Server configuration) The determination device 101 of embodiment 1 adopts a web application format that uses application software based on web technology, in which the determination device 101 downloads a measurement program and a trained model from the server 501 each time a measurement is performed, and determines on the determination device 101 whether the subject's health condition is in a state where there is a risk due to acute factors, and whether the subject's health condition is in a state where there is a risk due to chronic factors.

[0060] FIG. 6 shows a sequence diagram for explaining the procedure in which the determination device 101 according to embodiment 1 downloads a program from the server 501, executes the program, and determines whether the subject's health condition is in a state in which there is a risk due to an acute factor, and whether the subject's health condition is in a state in which there is a risk due to a chronic factor.

[0061] First, in steps S201 to S203, the server 501 stores the trained model for image recognition, the program for chronic factor determination, and the program for acute factor determination.

[0062] Next, in step S204, the server 501 transmits the chronic factor determination program, the acute factor determination program, and the trained model to the determination device 101.

[0063] Next, in step S205, the determination device 101 executes the program for determining acuteness factors. At this time, since the file size of the trained model for image classification is relatively large, downloading may not be complete.

[0064] Next, in step S206, the determination device 101 captures an image of the subject's nails, and determines whether the subject's health condition is at risk due to an acute factor in step 207. Steps S205 to S207 correspond to steps S104 to S107 of the operation procedure shown in FIG.

[0065] Since it takes a predetermined measurement time to determine whether the subject's health state is at risk due to an acute factor, in step S208, the determination device 101 completes downloading the trained model. Next, in step S209, the determination device 101 executes a chronic factor determination program, and in step S210, determines whether the subject's health state is at risk due to a chronic factor using the trained model. Steps S209 to S210 correspond to steps S109 to S111 of the operation procedure shown in FIG. 6.

[0066] Next, in step S211, the determination device 101 displays the determination result of whether the subject's health condition is a state in which there is a risk due to an acute factor, and whether the subject's health condition is a state in which there is a risk due to a chronic factor.

[0067] The determination device 101 according to the first embodiment executes respective determination programs for determining whether the subject's health state is a state in which there is a risk due to an acute factor and whether the subject's health state is a state in which there is a risk due to a chronic factor. Therefore, the determination device 101 has the advantage of preventing the processing load for determining whether the subject's health state is a state in which there is a risk due to an acute factor and whether the subject's health state is a state in which there is a risk due to a chronic factor from being concentrated on the server 501. On the other hand, the determination device 101 needs to download an acute factor determination program, a chronic factor determination program, and a trained model at the start of measurement. Because the trained model has a relatively large file size, downloading it may take several minutes. The determination device 101 first determines whether the subject's health state is a state in which there is a risk due to an acute factor and downloads the trained model in parallel, thereby shortening the overall measurement time.

[0068] The determination device 101 may have the following modified example.

[0069] [Variation 1] When an image of the nail is captured, a guide that defines the position, size, direction, etc. of the nail in the image may be displayed on the output unit 4.

[0070] 7 is a diagram showing an example of an imaging screen including a guide, which is displayed on the output unit 4. The imaging screen is displayed by the image acquisition unit 1 controlling the output unit 4 when the camera serving as the imaging unit 11 is activated in step S104 of the operation procedure of the determination device 101 shown in FIG. 5, and when an instruction to capture an image of the nail is output to the test subject in step S110. As shown in FIG. 7, the image display unit 41 of the output unit 4 displays an image of the nail 201 captured by the imaging unit 11 in real time, and also displays a guide H superimposed on the image of the nail 201.

[0071] The guide H includes a guide figure H1 and text information H2. The guide figure H1 is displayed so as to surround a predetermined area on the image display unit 41. The predetermined area is an area including a nail position that is suitable when the nail pulse wave response detection unit 21 detects a predetermined area R (see FIG. 2) of the nail in S106 or when the nail appearance case determination unit 31 performs image determination in S111, and is, for example, a central area on the image display unit 41. The text information H2 displays a character string to instruct the subject how the nail should be displayed. In the example shown in FIG. 7, the text information H2 indicates that the nail 201 should be positioned inside the guide figure H1 and that the fingertip should be facing upward on the image display unit 41 as the direction of the nail 201. The subject adjusts the position of the determination device 101 or the nail 201, etc., so that the nail is displayed in accordance with the instructions of the text information H2, and then captures an image of the nail.

[0072] In this way, the output unit 4 of the determination device 101 according to the first modification displays a guide that defines how the position, size, direction, etc. of the nail appears in the image. This improves the detection accuracy when the nail pulse wave response detection unit 21 detects the predetermined area R (see FIG. 2 ) of the nail and the determination accuracy when the nail appearance case determination unit 31 performs image determination, thereby enabling highly accurate determination of whether the subject's health condition is at risk.

[0073] Instead of or in addition to displaying the text information H2, the output unit 4 may output a sound corresponding to the text information H2 from the sound output unit.

[0074] [Variation 2] Whether the subject's health condition is at risk due to an acute factor may be determined by determining whether the subject has tachycardia or bradycardia. Tachycardia and bradycardia, which are arrhythmias, can cause dizziness, which is an acute physical condition, and are therefore classified as acute factors. Therefore, whether the subject's health condition is at risk due to an acute factor is determined based on whether the subject has tachycardia or bradycardia.

[0075] FIG. 8 is a graph showing an example of determining whether a subject has tachycardia or bradycardia from a pulse wave detected from changes in an image of the nail. Between time t1, when detection of a pulse wave w1 begins, and time t2, when blood flow to the finger is blocked, nail pulse wave response detection unit 21 measures the pulse wave period. In the example shown in FIG. 8, the pulse wave period is the time from the time when the amplitude of the pulse wave reaches a minimum to the time when the amplitude of the pulse wave next reaches a minimum. This example is not limited to this example, and the pulse wave period may be the time from the time when the amplitude of the pulse wave reaches a maximum to the time when the amplitude of the pulse wave next reaches a maximum. Furthermore, nail pulse wave response detection unit 21 measures the pulse wave period multiple times. Preferably, nail pulse wave response detection unit 21 measures the pulse wave period three or more times. In the example shown in FIG. 8, nail pulse wave response detection unit 21 measures the pulse wave period three times, and periods a1, a2, and a3 are obtained as measurement results.

[0076] Tachycardia refers to a pulse rate of more than 100 beats per minute, and bradycardia refers to a pulse rate of less than 50 beats per minute. Nail pulse wave response detection unit 21 calculates the subject's pulse rate per minute based on the measured pulse wave period. For example, nail pulse wave response detection unit 21 calculates the average value of pulse wave periods measured multiple times. In the example shown in FIG. 8, nail pulse wave response detection unit 21 calculates the average value of periods a1, a2, and a3. Nail pulse wave response detection unit 21 calculates the pulse rate per minute by dividing 60 seconds by the calculated average value. Note that nail pulse wave response detection unit 21 may calculate any representative value, such as the median, instead of the average value of the pulse wave period.

[0077] Based on the pulse rate per minute, nail pulse wave response detection unit 21 determines whether the subject has tachycardia, bradycardia, or neither tachycardia nor bradycardia. For example, nail pulse wave response detection unit 21 determines that the subject has tachycardia when the calculated pulse rate per minute exceeds 100 beats per minute. Furthermore, nail pulse wave response detection unit 21 determines that the subject has bradycardia when the pulse rate per minute is below 50 beats per minute. Furthermore, nail pulse wave response detection unit 21 determines that the subject has neither tachycardia nor bradycardia when the pulse rate per minute is between 50 and 100 beats per minute.

[0078] The pulse wave period is measured during the wait time in step S105 of the operational procedure in Figure 5. Because the pulse wave period is typically within the range of 0.5 to 1.5 seconds, nail pulse wave response detecting unit 21 can measure the pulse wave period at least three times during the approximately five-second wait time.

[0079] FIG. 9 is a diagram showing an example of a screen displayed on the output unit 4 when it is determined whether the subject has tachycardia or bradycardia. The output unit 4 displays whether the subject has tachycardia, bradycardia, or neither tachycardia nor bradycardia. In the example shown in FIG. 9, the determination result display unit 42 of the output unit 4 displays, in addition to the screen shown in FIG. 2, a tachycardia icon 404 indicating whether the subject has tachycardia and a bradycardia icon 405 indicating whether the subject has bradycardia. The tachycardia icon 404 is displayed in a different display mode depending on whether the subject has tachycardia. The bradycardia icon 405 is displayed in a different display mode depending on whether the subject has bradycardia. In the example shown in FIG. 9, the tachycardia icon 404 is displayed in an emphasized manner, and the bradycardia icon 405 is displayed without being emphasized, indicating that the subject has tachycardia.

[0080] In this way, in the determination device 101 according to Modification 2, the acute cause determination unit 2 determines whether the subject has tachycardia or not and whether the subject has bradycardia or not. This enables the determination device 101 according to Modification 2 to determine whether the subject's health condition is at risk due to an acute cause from a perspective different from that of dehydration.

[0081] In the above description, the determination device 101 according to the second modification determines both whether the subject is dehydrated and whether the subject has tachycardia or bradycardia, but the present invention is not limited to this example. The determination device 101 may determine only whether the subject has tachycardia or bradycardia.

[0082] [Embodiment 2] FIG. 10 is a block diagram showing the configuration of a determination device 102 according to the second embodiment. The determination device 102 according to the second embodiment differs from the determination device 101 according to the first embodiment in that the acute factor determination unit 2 determines whether the subject's health condition is at risk due to an acute factor by comparing each changing image with a reference image, and that the chronic factor determination unit 3 determines whether the subject's health condition is at risk due to a chronic factor, and then the acute factor determination unit 2 determines whether the subject's health condition is at risk due to an acute factor. That is, the acute factor determination unit 2 of the determination device 102 uses information obtained by comparing each changing image with a reference image as information obtained from changes in the images acquired in time series by the image acquisition unit 1. To achieve this, the acute factor determination unit 2 of the determination device 102 according to the second embodiment includes a frame rate acquisition and counting unit 23 and a nail appearance dynamic determination unit 24. Other configurations of the determination device 102 according to the second embodiment are similar to the corresponding configurations of the determination device 101 according to the first embodiment, and therefore detailed description thereof will be omitted.

[0083] Next, a procedure will be described in which the determination device 102 according to the second embodiment determines whether or not the subject's health state is at risk due to an acute factor, based on a comparison between each changing image and a reference image. Figures 11(A) to 11(E) show images of nails captured using the determination device 102 according to the second embodiment, illustrating a series of steps from an anemic state to a healthy state. Figure 12 shows a flowchart for explaining a procedure in which the determination device 102 according to the second embodiment calculates the recovery time from an anemic state to a healthy state from the captured nail images.

[0084] Unlike the determination device 101 according to embodiment 1, which uses a pulse wave response to determine whether the subject's health condition is at risk due to acute factors, the determination device 102 according to embodiment 2 determines whether the subject's health condition is at risk due to acute factors by comparing each changing image with a reference image. That is, the determination device 102 according to embodiment 2 utilizes the fact that an image of the nail when the finger is compressed to block blood flow is equivalent to an image of the nail when the finger is anemic. The determination device 102 according to embodiment 2 is characterized in that it performs the same AI determination of the nail image for each frame of the video as in determining whether the subject's health condition is at risk due to chronic factors, measures the return time from the number of frames required for the finger to return from an anemic state to normal when the finger is compressed and then released, and determines whether the subject's health condition is at risk due to acute factors based on the return time. Therefore, unlike the determination device 101 according to embodiment 1, which uses a pulse wave response, the determination device 102 assumes that the initial state before the finger is compressed is normal (healthy).

[0085] Therefore, it is preferable that the determination device 102 according to the second embodiment first determines whether the subject's health condition is at risk due to chronic factors, and if the subject's health condition is determined to be normal (healthy) with a reliability equal to or greater than a predetermined value, then determines whether the subject's health condition is at risk due to acute factors.

[0086] The determination device 102 according to the second embodiment calculates the return time from the frame rate of the image. However, when a general-purpose terminal such as a smartphone is used as the determination device 102, the frame rate may change while the determination using the trained model is being performed depending on the performance of the terminal. Therefore, the frame rate acquisition and counting unit 23 acquires the video frame rate of the imaging unit 11 while the terminal is performing the determination using the trained model. The frame rate acquisition and counting unit 23 counts the number of frames of the moving image from when the pressure on the finger is released to when the finger returns to its original state. The frame rate acquisition and counting unit 23 calculates the actual return time from the acquired video frame rate and the counted number of frames. For example, if the normal frame rate is 30 fps and the video frame rate drops to 10 fps after performing image determination using the trained model, the frame rate acquisition and counting unit 23 converts the time per frame to 100 msec using this frame rate.

[0087] First, in step S301, the frame rate acquisition and counting unit 23 determines whether the initial state of the subject's nail is healthy. FIG. 11A shows an example of the determination result for the initial state of the nail. In FIG. 11A, the captured nail image is displayed on the image display unit 41 at the top of the output unit 4, which is a display device, and the determination result as to whether the subject's health condition is a risk state due to chronic factors is displayed on the determination result display unit 42 at the bottom of the output unit 4. Examples of the determination result as to whether the subject's health condition is a risk state due to chronic factors are displayed as bars 421-424, each representing a level of "aging," "anemia," "fatigue," and "health." These levels correspond to the reliability of image recognition using a trained model. Additionally, a bar 425 may be displayed indicating the level of halation or the level of "unknown," which indicates that the captured image is not a target for image recognition, such as when a finger is not displayed in the captured image.

[0088] 11(A), the display of bar 424 corresponding to the "healthy" state indicates that the health level (health degree) is 90% or higher. Therefore, in step S301, the frame rate acquisition and counting unit 23 can confirm that the initial state of the nail is healthy.

[0089] Next, in step S302, the determination device 102, which is a smartphone, issues an announcement encouraging the subject to block blood flow, such as "Please press your finger against the surface." When the subject follows this announcement and presses their finger against a desk or the like, blood flow is blocked from the bottom of the nail, resulting in anemia. At this time, the determination device 102 captures an image of the nail and determines whether the subject's health condition is at risk due to chronic factors. As shown in FIG. 11(B), a bar 422 indicating the anemic state is displayed as extended. In the example shown in FIG. 11(B), the degree of anemia (anemia level) is 70% or higher. At this time, in step S303, the frame rate acquisition and counting unit 23 confirms that the anemia level has reached "70% or higher" due to blockage.

[0090] Next, the assessment device 102, which is a smartphone, issues an announcement urging the subject to release the pressure on their finger, saying, "Please relax your finger." When the subject releases the pressure on their finger in response to this announcement, the blood flow to the bottom of the nail is released, and the finger attempts to return from its anemic state to its original state. Figure 11(C) shows an image of the nail captured immediately after the finger was released, along with the results of assessing the anemia state from the captured image of the nail. The length of the bar 422 indicating the degree of anemia has shortened, and the degree of anemia has decreased to approximately 60%. Furthermore, the length of the bar 424 indicating the degree of health has lengthened, and the health level is now approximately 30%.

[0091] Next, in step S305, the frame rate acquisition and counting unit 23 starts counting the number of frames at a frame where the anemia level has dropped by 10% from "anemia level of 70% or more." A frame where the anemia level has dropped by 10% from "anemia level of 70% or more" refers to a frame where the anemia level is first observed to be 10% or more lower than the anemia level in a frame where the anemia level was confirmed to be 70% or more in step S303. For example, as shown in FIG. 11(B), if the anemia level during occlusion is 80%, the frame rate acquisition and counting unit 23 starts counting the number of frames when the anemia level, which is 10% lower than this value, reaches 70%.

[0092] As time passes from the state shown in Fig. 11(C), the "degree of anemia" percentage drops further (step S306), as shown in Fig. 11(D). At this point, the percentage indicating the degree of health is about 50%, and the patient has not yet returned to the original healthy state, so the number of frames continues to be counted.

[0093] As time passes from the state shown in Fig. 11(D), the percentage indicating the health level becomes 80% or more, as shown in Fig. 11(E). Therefore, in step S307, the frame rate acquisition and counting unit 23 determines that the state has returned to the original "health level of 80% or more" state, and ends counting the number of frames.

[0094] Next, in step S308, the frame rate acquisition and counting unit 23 calculates the return time from the number of frames from steps S305 to S307. For example, if the frame rate of the video is 10 fps and the number of frames is 20, the return time is 2 seconds. As described above, in the determination device 102 according to the second embodiment, the acute cause determination unit 2 can determine whether the subject's health condition is at risk due to an acute cause based on a comparison between each changing image and a reference image.

[0095] Next, the operation procedure of the determination device 102 according to embodiment 2 will be described. Fig. 13 shows a flowchart for explaining the procedure from downloading a measurement program using the determination device 102 according to embodiment 2 to outputting a determination result indicating whether or not the subject's health state is a state in which there is a risk due to an acute factor, and whether or not the subject's health state is a state in which there is a risk due to a chronic factor. In the flowchart shown in Fig. 13, one measurement operation is executed from step S404 to step S414.

[0096] First, in step S401, the determination device 102, which is a smartphone, accesses a predetermined website. Next, in step S402, the determination device 102 announces the wait time and the measurement content to the subject. The determination device 102 according to the second embodiment uses a trained model both to determine whether the subject's health state is a state in which there is a risk due to a chronic factor and to determine whether the subject's health state is a state in which there is a risk due to an acute factor. Therefore, downloading the trained model requires a certain amount of time. Therefore, the subject is notified that a predetermined wait time will occur.

[0097] Next, in step S403, the determination device 102 downloads the chronic factor determination program, the acute factor determination program, and the trained model. This downloading takes, for example, 1 to 2 minutes.

[0098] Next, in step S404, the determination of whether the subject's health condition is a state with a risk due to a chronic factor is started, and the camera, which is the imaging unit 11, is activated. The determination device 102 according to the second embodiment performs the determination of whether the subject's health condition is a state with a risk due to a chronic factor before determining whether the subject's health condition is a state with a risk due to an acute factor. As described above, the determination device 102 according to the second embodiment determines whether the subject's health condition is a state with a risk due to an acute factor by compressing the finger to create a blood flow obstruction state, and then determining the time from the release of the blood flow obstruction until the nail color returns to normal from the image of the nail. Therefore, it is essential that the color of the subject's nail before the blood flow obstruction is clearly distinguishable from the color of the nail during compression, which is a color that indicates a healthy state. Therefore, as a prerequisite for determining whether the subject's health condition is a state with a risk due to an acute factor, the determination device 102 determines whether the subject's health condition is a state with a risk due to a chronic factor before the determination, thereby confirming that the subject is healthy.

[0099] In contrast, the determination device 101 according to the first embodiment uses a pulse wave detected from a change in an image of the nail to determine whether the blood flow returns to its original state after the release of the blood flow obstruction caused by finger compression. In this case, the color of the nail does not substantially affect the determination of the health condition unless the color change is so extreme that the pulse wave cannot be detected. Therefore, the determination device 101 according to the first embodiment does not use the determination of whether the subject's health condition is a state in which there is a risk due to a chronic factor as a prerequisite for determining whether the subject's health condition is a state in which there is a risk due to an acute factor.

[0100] Next, in step S405, the determination device 101 waits for a predetermined waiting time. The waiting time is, for example, about 5 seconds. By providing such a waiting time, it is ensured that the nail deformed by the pressure of the finger is returned to its original state.

[0101] Next, in step S406, the determination device 102, which is a smartphone, issues an announcement to the subject prompting the subject to take an image of the nail portion. The subject follows the announcement and takes an image of the nail using the camera of the smartphone.

[0102] Next, in step S407, the chronic factor determination unit 3 inputs the captured image into the trained model to determine whether the subject's health condition is at risk due to a chronic factor.

[0103] Next, in step S408, the determination device 102 determines whether the subject is in a healthy state based on the determination result of whether the subject's health state is a state in which there is a risk due to a chronic factor.

[0104] If the result of determining whether the subject's health condition is a state with a risk due to chronic factors indicates that the subject is not in a healthy state, the condition that the subject's nails are in a healthy state, which is a prerequisite for determining whether the subject's health condition is a state with a risk due to acute factors, is not met. Therefore, the determination of whether the subject's health condition is a state with a risk due to acute factors is not performed. In this case, in step S409, the output unit 4 displays only the determination result of whether the subject's health condition is a state with a risk due to chronic factors and issues advice to the subject according to the determination result of whether the subject's health condition is a state with a risk due to chronic factors. Thereafter, in step S410, the determination device 102 terminates the operation of the imaging unit 11. This completes the series of operational procedures.

[0105] On the other hand, if the determination result of whether the subject's health condition is a state with a risk due to a chronic factor indicates that the subject's nail condition is healthy, the determination device 102 starts determining whether the subject's health condition is a state with a risk due to an acute factor. In step S411, the determination device 102 waits for a predetermined waiting time. The waiting time is, for example, about 5 seconds. This ensures that the pressure on the finger is completely released in order to determine whether the subject's health condition is a state with a risk due to a chronic factor.

[0106] Next, in step S412, the determination device 102 outputs an instruction to the subject to capture an image of the nail, and issues an announcement to encourage the subject to compress and release the finger.

[0107] Next, in step S413, as shown in FIGS. 11 and 12, the frame rate acquiring and counting unit 23 measures the return time Ts from the blood flow obstructed state to the original state based on the change in the image of the nail.

[0108] Next, in step S414, the output unit 4 displays the results of the determination of whether the subject's health condition is a state of risk due to acute factors and whether the subject's health condition is a state of risk due to chronic factors, and issues advice to the subject based on these determination results. When a single measurement operation determines whether the subject's health condition is a state of risk due to acute factors and whether the subject's health condition is a state of risk due to chronic factors, the image used to determine whether the subject's health condition is a state of risk due to chronic factors is preferably an image acquired before the finger compression for determining whether the subject's health condition is a state of risk due to acute factors (e.g., the image in FIG. 11(A)) or an image acquired after blood flow recovery from the finger compression (e.g., the image in FIG. 11(E)). This is because these images are taken without finger compression and are suitable for determining whether the subject's health condition is a state of risk due to chronic factors.

[0109] Thereafter, in step S415, the operation of the imaging unit 11 is terminated. This completes the series of operation procedures.

[0110] The determination device 102 according to the second embodiment can easily obtain the return time by measuring the return time from the number of frames required to return from an anemic state to a healthy (normal) state.

[0111] [Embodiment 3] Next, a determination device 103 according to a third embodiment will be described. FIG. 14 is a block diagram illustrating the configuration of the determination device 103 according to the third embodiment. The determination device 103 according to the third embodiment differs from the determination device 101 according to the first embodiment in that, when the acute factor determination unit 2 determines that the subject's health condition is a state in which there is a risk due to an acute factor, the output unit 4 outputs the determination result in a different manner from when the subject's health condition is a state in which there is a risk due to a chronic factor. For example, in step S112 of FIG. 5 or step S414 of FIG. 13, it is preferable that the determination results be displayed in different manners when the subject's health condition is a state in which there is a risk due to an acute factor and when the subject's health condition is a state in which there is a risk due to a chronic factor. For this purpose, the determination device 103 according to the third embodiment may include an emergency notification unit 8. Furthermore, the determination device 103 according to the third embodiment may be connected to a management center 300 via the Internet 400. The other configurations of the determination device 103 according to the third embodiment are similar to those of the determination device 101 according to the first embodiment, and therefore detailed description thereof will be omitted.

[0112] Because acute health conditions such as severe dehydration in the elderly are life-threatening, it is preferable that the output unit 4 not only displays the determination result of whether the subject's health condition is at risk due to acute factors and whether the subject's health condition is at risk due to chronic factors, but also notifies the subject that the subject's health condition is at risk due to acute factors in a more understandable manner. For example, if the subject's health condition is determined to be at risk due to acute factors, the output unit 4 preferably displays a message indicating an emergency, such as "Do not exercise. Drink fluids immediately." On the other hand, if the subject's health condition is determined to be at risk due to chronic factors, the output unit 4 preferably displays a message indicating that the emergency is not an emergency, such as "Eat a balanced diet and get sufficient rest daily." If the subject's health condition is determined to be at risk due to chronic factors, the output unit 4 preferably displays a message indicating that the emergency is not an emergency, such as "Eat a balanced diet and get sufficient rest daily."

[0113] Furthermore, the determination device 103 according to the third embodiment preferably includes, in addition to the output unit 4, an emergency notification unit 8 for notifying that the subject's health condition is at risk due to an acute factor. The emergency notification unit 8 is, for example, an alarm buzzer that notifies people around the subject of an abnormality. For example, if the subject's health condition is determined to be at risk due to an acute factor, the emergency notification unit 8 may sound the alarm buzzer, and if the subject's health condition is determined to be at risk due to a chronic factor, the emergency notification unit 8 may not sound the alarm buzzer.

[0114] Furthermore, it is also assumed that elderly people or the like may have difficulty contacting external organizations on their own. Therefore, the determination device 103 may notify an external management center 300 or medical institution that provides care for the subject that the subject's health condition is at risk due to an acute factor via a communication network such as the Internet 400. On the other hand, if the subject's health condition is at risk due to a chronic factor, the determination device 103 may not notify the management center 300 or medical institution.

[0115] In this way, the determination device 103 can quickly notify a caregiver, a management center, etc. that the subject's health condition is at risk due to an acute factor, and therefore, a quick response can be made to the subject's health condition that is at risk due to an acute factor.

[0116] [Embodiment 4] Next, a determination device according to a fourth embodiment will be described. In the first embodiment, an example has been described in which a determination device 101, which is a mobile terminal such as a smartphone, determines whether a subject's health condition is a state in which there is a risk due to an acute factor and whether the subject's health condition is a state in which there is a risk due to a chronic factor. However, the present invention is not limited to this example. For example, an image of the subject's nail may be captured by a mobile terminal, the captured image may be transmitted to a server, and the server may determine whether the subject's health condition is a state in which there is a risk due to an acute factor and whether the subject's health condition is a state in which there is a risk due to a chronic factor. In the fourth embodiment, an example will be described in which a server or the like is operated as the determination device. FIG. 15 is a block diagram showing the configuration of a determination device 502 according to the fourth embodiment. The determination device 502 according to the fourth embodiment includes a control unit 110, a storage unit 15, and a communication unit 16, which are connected by a bus 17. The control unit 110 includes an image acquisition unit 111, an acute factor determination unit 120, and a chronic factor determination unit 130. On the other hand, the terminal 104 includes an imaging unit 11, an output unit 4, and a communication unit 6.

[0117] The control unit 110 of the determination device 502 may further include an image recognition unit 112 and a measurement correctness determination unit 113.

[0118] The acute factor determination unit 120 has a nail pulse wave response detection unit 121 and a pulse wave response time determination unit 122. The chronic factor determination unit 130 has a nail appearance case determination unit 131. The functions of the acute factor determination unit 120 and the chronic factor determination unit 130 in the fourth embodiment are similar to the functions of the acute factor determination unit 2 and the chronic factor determination unit 3 in the first embodiment, and therefore detailed description thereof will be omitted.

[0119] The storage unit 15 includes an image storage unit 151 and a trained model storage unit 152.

[0120] FIG. 16 shows a sequence diagram for explaining the procedure in which the terminal 104 uploads a captured image of the nail to a determination device 502, which is a server or the like, and the determination device 502 executes a program to determine whether the subject's health condition is in a state where there is a risk due to an acute factor, and whether the subject's health condition is in a state where there is a risk due to a chronic factor.

[0121] First, in step S501, the terminal 104 captures an image of the subject's nail, and in step S502, the terminal 104 transmits the captured image data to the determination device 502, which is a server or the like.

[0122] Next, in steps S503 to S505, the determination device 502 stores the trained model for image recognition, the program for chronic factor determination, and the program for acute factor determination in the storage unit 15. Note that the trained model for image recognition, the program for chronic factor determination, and the program for acute factor determination may be stored in the storage unit 15 in advance.

[0123] Next, in step S506, the determination device 502 executes the acute factor determination program. In step S507, the determination device 502 determines whether the subject's health condition is at risk due to an acute factor.

[0124] Next, in step S508, the determination device 502 executes the chronic factor determination program. In step S509, the determination device 502 determines whether the subject's health condition is a state in which there is a risk due to a chronic factor.

[0125] Next, in step S510, the determination device 502 transmits to the terminal 104 the determination results of whether the subject's health condition is in a state where there is a risk due to acute factors and whether the subject's health condition is in a state where there is a risk due to chronic factors.

[0126] Next, in step S511, the terminal 104 displays the determination result received from the determination device 502 on the output unit 4.

[0127] In the above description, the determination device 502 determines whether the subject's health state is a state of risk due to acute factors in the first half and whether the subject's health state is a state of risk due to chronic factors in the second half. However, this is not limited to this example. The determination device 502 may determine whether the subject's health state is a state of risk due to chronic factors in the first half and whether the subject's health state is a state of risk due to acute factors in the second half. This is because the determination device 502 stores the trained model for image recognition and does not need to download it from an external device. That is, the determination device 502 according to the fourth embodiment does not require a waiting time for downloading the trained model for image recognition from an external device, as in the determination device 101 according to the first embodiment. Therefore, even if the determination of whether the subject's health state is a state of risk due to chronic factors is performed in the first half, there is no risk of the measurement time being prolonged.

[0128] As described above, in the fourth embodiment, the terminal 104 only captures an image of the nail using the imaging unit 11 and transmits the image data to the determination device 502 as a file or via streaming. The determination device 502 performs all measurements and determinations and returns only the determination results to the terminal 104. In this case, it is preferable that the determination device 502 has a certain level of processing power and that the communication speed of the image data from the terminal 104 to the determination device 502 is acceptably high. Furthermore, since the terminal 104 does not execute a determination program for determining whether the subject's health state is a state in which there is a risk due to an acute factor and whether the subject's health state is a state in which there is a risk due to a chronic factor, the terminal 104 does not necessarily need to have advanced processing power. For example, an inexpensive smartphone or the like can be used as the terminal 104.

[0129] [Embodiment 5] In the above description, an example has been described in which the determination device determines both whether the subject's health state is a state in which there is a risk due to an acute factor and whether the subject's health state is a state in which there is a risk due to a chronic factor during a single measurement operation, but this is not limited to such an example. The determination device may be configured to independently perform each measurement operation to determine whether the subject's health state is a state in which there is a risk due to an acute factor and whether the subject's health state is a state in which there is a risk due to a chronic factor. FIG. 17 shows a block diagram illustrating the configuration of a determination device 105 according to a fifth embodiment. FIG. 18 shows an example of the display of the acute mode execution button 43 and the chronic mode execution button 44 on the determination device 105 according to the fifth embodiment.

[0130] Since acute factors can have a significant impact on health status, it is preferable to frequently check whether the subject's health status is at risk due to acute factors. In contrast, chronic factors reflect long-term changes in physical condition, such as dietary habits, and are therefore thought to be less likely to change in the short term than acute factors. Therefore, only whether the subject's health status is at risk due to acute factors may be determined on a daily basis, and whether the subject's health status is at risk due to chronic factors may be determined, for example, approximately once a week.

[0131] Furthermore, determining whether the subject's health condition is at risk due to acute factors involves pressing a finger against a desk or the like, which may not be possible if there is no desk or the like. In such cases, there may be cases where only the determination of whether the subject's health condition is at risk due to chronic factors is performed. Therefore, a mode for determining whether the subject's health condition is at risk due to acute factors and a mode for determining whether the subject's health condition is at risk due to chronic factors may be provided. The mode may be switched by the subject operating a button or the like.

[0132] Therefore, the determination device 105 according to the fifth embodiment has a mode operation unit 9 for individually determining whether or not the subject's health state is a state in which there is a risk due to an acute factor, and determining whether or not the subject's health state is a state in which there is a risk due to a chronic factor. The mode operation unit 9 may be configured to display an acute mode execution button 43 and a chronic mode execution button 44 on the output unit 4. When the subject presses the acute mode execution button 43, a determination is made as to whether or not the subject's health state is a state in which there is a risk due to an acute factor, and when the subject presses the chronic mode execution button 44, a determination is made as to whether or not the subject's health state is a state in which there is a risk due to a chronic factor.

[0133] According to the determination device 105 of embodiment 5, it is possible to perform the necessary determinations according to the request of the subject, such as whether the subject's health condition is in a state where there is a risk due to acute factors, and whether the subject's health condition is in a state where there is a risk due to chronic factors. [Explanation of symbols]

[0134] 1 Image acquisition unit 2. Acute Factor Assessment Section 3 Chronic Factor Assessment Section 4 Output section 5 Storage section 6. Communications Department 7. Bus 8 Emergency Notification Department 9 Mode operation section 10 Control Unit 11 Imaging unit 21 Nail pulse wave response detector 22 Pulse wave response time determination unit 23 Frame rate acquisition counter 24 Nail appearance dynamic determination section 31 Nail Appearance Case Assessment Section 41 Image display unit 42 Judgment result display section 43 Acute mode execution button 44 Chronic mode execution button 51 Image storage unit 52 Trained model memory 101 Judgment device 300 Management Center 400 Internet 501 Server

Claims

1. an image acquisition unit for acquiring an image of the subject's nail; a determination unit that determines whether the subject is in a health condition at risk based on information about the appearance of the nail and information obtained from changes in the images acquired over time from the images acquired by the image acquisition unit; and an output unit that outputs a determination result by the determination unit; A determination device comprising:

2. The determination unit has a first factor determination unit that determines whether or not the subject's health state is in a state in which there is a risk based on a predetermined first factor, based on a change in the image, and a second factor determination unit that determines whether or not the subject's health state is in a state in which there is a risk based on a predetermined second factor including a factor different from the first factor, based on information on the appearance of the nail. The determination device according to claim 1 .

3. The determination device according to claim 2 , wherein the first factor is an acute factor, and the second factor is a chronic factor.

4. 4. The determination device according to claim 3, wherein the first factor determination unit uses the pulse wave detected from the change in the image to determine whether the subject's health state is at risk due to the acute factor, based on the time it takes for the amplitude of the pulse wave, which has decreased due to compression of the finger, to return to its state before compression after the compression of the finger is released.

5. 5. The determination device according to claim 3, wherein the first factor determination unit measures a pulse wave period detected from the change in the image, and determines whether the subject's health state is at risk due to the acute factor based on the measured pulse wave period.

6. Based on the images acquired during one measurement operation, The first factor determination unit determines whether the health state of the subject is a state in which there is a risk based on the acute factor, and the second factor determination unit determines whether the subject's health condition is a state in which there is a risk based on the chronic factor; The determination device according to claim 3 or 4.

7. the first factor determination unit determines whether the subject's health state is at risk due to the acute factor based on a pulse wave detected from changes in images obtained over time; the first factor determination unit performs an operation of determining whether or not the subject's health state is in a state in which there is a risk due to the acute factor, and then the second factor determination unit performs an operation of determining whether or not the subject's health state is in a state in which there is a risk due to the chronic factor; The determination device according to claim 3 or 4.

8. the first factor determination unit determines whether the subject's health state is at risk due to the acute factor based on a comparison between each changing image and a reference image; the second factor determination unit performs an operation of determining whether or not the subject's health state is in a state in which there is a risk due to the chronic factor, and then the first factor determination unit performs an operation of determining whether or not the subject's health state is in a state in which there is a risk due to the acute factor; The determination device according to claim 3 or 4.

9. When the first factor determination unit determines that the health condition of the subject is a state in which there is a risk based on the acute factor, the output unit outputs the determination result in a manner different from that when the health condition of the subject is determined to be a state in which there is a risk based on the chronic factor. The determination device according to claim 3 or 4.

10. 7. The determination device of claim 6, wherein when determining whether the subject's health condition is in a state where there is a risk due to the acute factor and whether the subject's health condition is in a state where there is a risk due to the chronic factor by a single measurement operation, the image used to determine whether the subject's health condition is in a state where there is a risk due to the chronic factor is an image acquired before finger compression for determining whether the subject's health condition is in a state where there is a risk due to the acute factor, or an image acquired after blood flow has recovered from finger compression.

11. The determination device according to claim 10 , wherein the image acquired after blood flow recovery from finger compression is an image acquired after blood flow recovery is detected or an image acquired a predetermined time after finger compression.

12. the image acquisition unit acquires an image of the nail by outputting an instruction to the subject to capture an image of the nail; the output unit displays a guide that defines how the nail appears in the nail image. The determination device according to any one of claims 1 to 4.

13. The image acquisition unit acquires an image of the subject's nails, The processor: determining whether the subject is in a health condition at risk based on information about the appearance of the nail from the images acquired by the image acquisition unit and information acquired from changes in the images acquired over time; outputting a determination result as to whether or not the subject is in a health state at risk; A determination method comprising:

14. On the computer, An image acquisition unit acquires an image of the subject's nail; a step in which a determination unit determines whether or not the subject is in a health condition at risk based on information on the appearance of the nail from the images acquired by the image acquisition unit and information acquired from changes in the images acquired over time; an output unit outputting a determination result as to whether the subject is in a health state at risk; A determination program characterized by executing the above.

Images