Cough prediction device, droplet blocking device, and cough prediction program

Abstract

To provide a device capable of properly coping with occurrence of predicted cough and accurately predicting occurrence of cough of a patient or the like, for preventing splash timely, at only required timing.SOLUTION: A cough predicting device comprises: imaging means for imaging an observation object person; and surface displacement amount calculating means for the observation object person. The surface displacement amount calculating means calculates a surface displacement amount of a neck part of the observation object person from the image acquired by the imaging means, and the cough predicting device predicts occurrence of cough of the observation object person by detecting a peak from a waveform output from the surface displacement amount calculating means. There is also provided a splash blocking device which is used with the cough predicting device, in which the splash blocking device is configured to receive a cough predicting signal output from the cough predicting device and operate a splash blocking mechanism.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a cough prediction device, a droplet blocking device, and a cough prediction program. [Background technology]

[0002] As the spread of infection by Covid19 becomes serious, it is urgent that adequate countermeasures be taken. Various investigations have been conducted into droplet infection, aerosol infection, and contact infection, and it has been reported that droplet infection countermeasures are the deciding factor in preventing infection. Meanwhile, even before the problem of Covid19, there were various general droplet infection countermeasures such as masks, face shields, and protective clothing. Among these, the installation of droplet prevention stands, which are simple in structure but effective in preventing droplet infection, is used in various situations.

[0003] Patent Document 1 describes a splash prevention stand that is installed on counters and tables in various facilities such as pharmacies, hospitals, and restaurants. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Registered Utility Model No. 3227488 Summary of the Invention [Problem to be solved by the invention]

[0005] However, there are often situations where it is difficult to set up a splash prevention stand on a regular basis, such as when taking medical questions at medical facilities or assisting with meals at nursing homes. Even if it is possible to set up a splash prevention stand on a regular basis, it is necessary to frequently wipe the splash prevention panel with alcohol or other disinfectants, and the running costs such as refilling disinfectants and labor costs are not insignificant. In addition, it is no exaggeration to say that the reason why the work environment in medical facilities is said to be tight is actually due to the situation where such casual work is occurring in huge quantities. In such a situation, it would be very convenient if splashes could be prevented at the appropriate time only when necessary, but the prerequisite for this is to accurately predict the occurrence of coughing. In order to solve these problems, the present invention has an objective of providing a device that can accurately predict the onset of coughing in a patient, etc., and a device that can appropriately respond to the predicted onset of coughing. [Means for solving the problem]

[0006] The cough prediction device of the present invention has at least the following configuration. A cough prediction device comprising an imaging means for imaging an observation subject and a surface displacement amount calculation means for the observation subject, the surface displacement amount calculation means calculating a surface displacement amount of the observation subject from an image obtained by the imaging means. horizontal and the cough prediction device calculates a surface displacement amount in a direction output from the surface displacement amount calculation means. Derivative of the surface displacement in the horizontal direction of the neck The method is characterized in that it predicts the occurrence of coughing in the subject by detecting peaks from the waveform.

[0007] In addition, the cough prediction device of the present invention has at least the following configuration. A cough prediction device comprising: an imaging means for imaging a subject; a surface displacement amount calculation means for the subject; a blood flow amount calculation means for calculating a surface displacement amount of the neck of the subject from an image obtained by the imaging means; a tissue perfusion intensity output means for the subject; wherein the surface displacement amount calculation means calculates a surface displacement amount of the neck of the subject from an image obtained by the imaging means; the blood flow amount calculation means calculates or estimates a hemoglobin absorption light amount from the image obtained by the imaging means; the tissue perfusion intensity output means outputs a tissue perfusion intensity obtained based on the product of a spatial derivative of the surface displacement amount and a time derivative of the hemoglobin absorption light amount; and the cough prediction device predicts the occurrence of coughing in the subject by detecting a peak from an output waveform of the tissue perfusion intensity output means.

[0008] In addition, the droplet blocking device of the present invention is a droplet blocking device used together with the cough prediction device described above, and is characterized in that the droplet blocking device receives a cough prediction signal output by the cough prediction device and operates a droplet blocking mechanism. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is an explanatory diagram showing the system configuration of a cough prediction device according to an embodiment of the present invention. [Diagram 2] FIG. 1 is an explanatory diagram showing the process of coughing. [Diagram 3] FIG. 13 is a diagram showing how to diagnose the onset of coughing, where (a) shows the amount of displacement in the left-right direction, and (b) shows the amount of displacement in the cranio-caudal direction. [Figure 4] This figure shows the flow rate of the superficial peripheral circulation when coughing occurs, (a) showing the forehead and (b) showing the neck. [Diagram 5] This figure shows a waveform for predicting the onset of coughing, combining indicators of the neck surface shape and neck peripheral circulation. [Figure 6] 1 is an explanatory diagram showing a system configuration of a splash blocking device according to an embodiment of the present invention. [Figure 7] FIG. 1 is an explanatory diagram showing a state in which a droplet blocking device is applied to a medical field. [Figure 8] FIG. 1 is an explanatory diagram showing a state in which the droplet blocking device is applied to a care site. [Figure 9] 11 is an explanatory diagram showing other application scenes of the splash blocking device. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The cough prediction device 100A according to an embodiment of the present invention has an imaging means 19 for imaging the subject, in particular, the larynx in the neck and its surrounding area, and analyzes the obtained image to perform processing related to cough prediction.

[0011] (Overall configuration of the cough prediction device) 1 is an explanatory diagram showing a system configuration of a cough prediction device 100A according to an embodiment of the present invention. As shown in FIG. 1, the cough prediction device 100A according to the present embodiment includes an imaging means 19 and an information processing device 1 (computer).

[0012] The imaging means 19 is provided to image the body surface of a predetermined portion such as the larynx TH of the human body A and its surrounding area.

[0013] For example, a CCD type imaging device, a CMOS type imaging device, a MOS type imaging device, etc. can be used as the imaging means 19. In this embodiment, a stereo camera system using a plurality of cameras is used.

[0014] The information processing device 1 calculates various indices required for predicting the occurrence of coughing based on the signal from the imaging means 19.

[0015] The information processing device 1 (computer) has a communication unit 11, an input unit 12, a display unit 13, a sound output unit 14, a memory unit 15, a control unit 16 (CPU), an interface 18, etc. Each component (the communication unit 11, the input unit 12, the display unit 13, the sound output unit 14, the memory unit 15, and the interface 18) is electrically connected to the control unit 16 (CPU) via a communication line 17.

[0016] Under the control of the control unit 16, the communication unit 11 performs data communication with another computer (not shown) via a wireless or wired communication path.

[0017] The input unit 12 is an operation input unit such as a button, a switch, a keyboard, a mouse, or a touch panel, and outputs a signal according to the input to the control unit 16.

[0018] The display unit 13 is a display device such as an LED or a liquid crystal display device, and performs display under the control of the control unit 16 . The sound generating unit 14 is a sound generating device such as a speaker, and produces sound under the control of the control unit 16 . The storage unit 15 is a storage device such as a RAM, a ROM, a magnetic recording and reproducing device, or an SSD. The storage unit 15 stores programs and various data, and also functions as a work area for executing programs. In addition, the storage unit 15 may be provided with a special storage area 151 for storing a trained prediction model as necessary, for example, when an AI algorithm is used.

[0019] The control unit 16 comprehensively controls each component of the information processing device 1. In addition, the control unit 16 executes a program stored in the storage unit 15 to cause the information processing device 1 (computer) to realize the functions according to the present invention. In detail, the control unit 16 includes a surface displacement amount calculation means 161, a blood flow amount calculation means 162, a cough prediction means 163, and the like. The interface 18 receives a video signal from the image capture means 19 and also outputs an operation signal for operating a splash blocking mechanism (not shown).

[0020] The surface displacement amount calculation means 161 and the blood flow amount calculation means 162 both analyze the images obtained by the imaging means 19 and output the indices required by the cough prediction means 163 .

[0021] (Mechanism of cough occurrence prediction) Here, in order to explain the mechanism of action for predicting the occurrence of coughing, how the inventors of the present invention came to find the index necessary for coughing prediction will be described. As shown in Figure 2, when coughing that generates droplets occurs, it goes through three phases, namely, the inhalation phase, the compression phase, and the exhalation phase, from the normal state. In particular, under the expectation that the closing of the glottis, which is the decisive factor, can be detected by observing the larynx on the surface of the human body, the neck of the subject was observed.

[0022] Figure 3 shows this. Figure 3(a) shows the waveform in the left-right direction, and Figure 3(b) shows the waveform in the head-tail direction (up-down direction). Since data obtained by a stereoscopic imaging device using multiple cameras is used, the displacement in the depth direction is also taken into account. Figure 3(a) shows the time change in the horizontal spread of the neck as a result of summation while averaging the horizontal data of the neck region and the depth data. Figure 3(b) shows the time change in the head-tail direction as a result of summation while averaging the vertical data of the neck region and the depth data. As can be seen by comparing Figures 3(a) and (b), the head-tail movement shows characteristics only after the onset of coughing, whereas the horizontal spread of the neck shows signs before the onset of coughing. In other words, a certain degree of significant peak is seen before the large peak that occurs when coughing actually occurs.

[0023] However, in FIG. 3(a), there is a peak that can be interpreted as an artifact near the start of the output waveform. Therefore, the inventors of the present invention also focused on the change in blood flow of the subject. FIG. 4 shows blood flow, with FIG. 4(a) showing the waveform of blood flow at the forehead and FIG. 4(b) showing the waveform of blood flow at the neck. In (a), a peak is observed after the onset of coughing, while in (b), a peak is observed before the onset of coughing. In this way, it can be intuitively understood that the output waveform of FIG. 4(b) will be well matched with the output waveform of FIG. 3(a). Therefore, the product of the spatial differentiation (differentiation with respect to the horizontal area) of the output of FIG. 3(a) and the time differentiation of the output of FIG. 4(b) was taken, and the square root of the squared product was taken, and it was found that an index that eliminated the artifact was given. FIG. 5 shows the output waveform obtained in this way. It can be seen that a peak is detected about 0.5 seconds before the onset of coughing.

[0024] In this way, the inventors of the present invention have obtained a new finding that the occurrence of coughing can be accurately predicted by using both the horizontal surface displacement of the neck and the blood flow rate of the neck as an index. However, the inventive idea is not limited to this. If we ignore some archeffects, as explained above, it is possible to predict the occurrence of coughing only from the horizontal surface displacement of the neck, and even if measures to block droplets are taken when coughing does not actually occur, this does not pose a major problem in practice.

[0025] In addition, in the embodiment of the present invention, in order to observe the subject non-invasively, attention is paid to the horizontal component of the neck, which is an easy-to-detect change, but even if the expansion or contraction of the glottis is detected by invasive observation, it can be said that it belongs to the idea of ​​the present invention in a broad sense. In other words, the target to be focused on as the displacement amount of the neck, such as the depth direction, is not limited to the horizontal direction.

[0026] (Specific configurations of the imaging means and the surface displacement amount calculation means) In this embodiment, as described above, a 3D image is generated using multiple cameras and the surface displacement is calculated. A 4K camera with a resolution of 1024 x 576 pixels and a sampling frequency of 30 frames / second manufactured by Logitech was used. Specifically, as a data extraction method, after recognizing a face from an acquired image, the image range of the neck is extracted. However, since a wide range of information is required to create a 3D image, data including 3D depth information is created once for an image including a face, and then data of the neck region is acquired. The horizontal, vertical, and depth directions of the neck are all calculated from the 3D data (displacement) of the neck region, but in the horizontal displacement calculation, the depth data is averaged and summation is performed to estimate the musculoskeletal displacement of the neck in the horizontal direction before and after coughing. The output waveform obtained by this calculation process is shown in FIG. 3(a).

[0027] Instead of using a stereo camera method using multiple cameras, a single camera may be used to obtain three-dimensional depth information using a time-of-flight (TOF) method, a pattern irradiation method, etc. In addition, even if it is a completely ordinary monocular camera, as long as it has high resolution, it is possible to quantify changes in the neck musculoskeletal system structure using a machine-learned prediction model that uses as training data the relationship between change data in face images consisting of many pixel patterns obtained from high resolution and the actual occurrence of coughing.

[0028] Furthermore, it is possible to measure the surface shape of the neck before and after coughing by focusing on subcutaneous muscle contraction. In other words, it is possible to estimate changes in the neck musculoskeletal structure without using non-invasive three-dimensional analysis. It is important to avoid coughing when working face-to-face with the subject or patient, not only during interviews but also during examinations, diagnoses, and treatments, and when a camera method cannot be used, it is possible to predict coughing using an invasive method.

[0029] (Specific Configuration of Blood Flow Calculation Means) The blood flow rate is also calculated by analyzing the image obtained by the imaging means 19. In this embodiment, the blood flow rate (absorption amount of hemoglobin) of the skin surface tissue is calculated by extracting the G signal of the RGB signal. The reason for this is that the light irradiated to the skin of a living body is scattered and absorbed in the subcutaneous tissue, and a part of it returns to the outside through the surface of the skin as reflected light. At this time, the light is absorbed by the hemoglobin contained in the blood, so the intensity of the reflected light varies depending on the increase or decrease in the amount of blood under the skin. In the case of visible light, the green luminance component, which has a high absorption rate of hemoglobin in blood, shows a particularly remarkable reaction, and the element for acquiring data in RGB is inexpensive, and it is easy to extract and use only the G signal. Of course, the R signal and the B signal may be used in addition to the G signal, or a 5-band camera with two sensitivities of cyan (C) and orange (O) may be used in addition to the normal RGB camera. Alternatively, infrared luminance analysis of the video signal under infrared light illumination may be used. In short, it is sufficient to image the red light absorption by hemoglobin.

[0030] The output waveforms obtained by applying the necessary signal processing to the image obtained from the imaging means 19 are shown in Figures 4(a) and (b). As mentioned above, the waveform of the blood flow rate on the surface tissue of the neck skin in (b) shows signs of coughing before it occurs, and this is what we intend to utilize.

[0031] (Specific configuration of cough prediction means) Next, the configuration of the tissue perfusion intensity output means, which is the main component of the cough prediction means 163, will be described. The tissue perfusion intensity is calculated using two indices, namely, the horizontal surface displacement of the neck and the hemoglobin tissue blood flow. As described above, the magnitude of the displacement or blood flow itself is calculated for both the horizontal surface of the neck and the hemoglobin tissue blood flow. However, when calculating the tissue perfusion intensity, the horizontal surface displacement is differentiated with respect to the horizontal area for the former, and the red shading change of the skin surface tissue blood flow is differentiated with respect to time for the latter. In particular, the latter is based on the hypothesis that the musculoskeletal system operates after sufficient blood flow is supplied to the tissue that realizes the coughing action, and the point where the acceleration change of the tissue blood flow becomes large indicates a functional change of the part that contributes to coughing, and utilizes the anatomical structure change of the musculoskeletal system before and after coughing, which utilizes the characteristic that the peripheral blood volume increases before coughing and changes after coughing. As the actual index, we used a function of time calculated by taking the square root of the product of the spatial or time derivatives of the two indexes, as shown in [Equation 1].

number

[0032] In the output waveform obtained by this calculation, a peak can be confirmed as a symptom of coughing about 0.5 seconds before the onset of coughing, as shown in Fig. 5. The device that uses this peak as a prediction signal for the onset of coughing is the cough prediction device 100A according to this embodiment.

[0033] The cough prediction means 163 according to this embodiment uses tissue perfusion intensity as an index, but if the influence of artefacts can be ignored to a certain extent or the blood flow calculation means 162 cannot be used, the cough prediction means 163 may use a peak in the output from the surface displacement calculation means 161 as a prediction signal for the occurrence of coughing. Even such a cough prediction device 100A is included in the technical idea of ​​the present invention.

[0034] (Configuration of splash protection device) So far, we have explained the cough prediction device 100A, which accurately predicts the occurrence of coughing in patients, etc. Next, we will explain the droplet blocking device, which is a device for appropriately dealing with the occurrence of coughing predicted using the cough prediction device 100A. 6 is an explanatory diagram showing the system configuration of the droplet blocking device 200A according to this embodiment. As shown in Fig. 6, the droplet blocking device 200A according to this embodiment is configured by connecting a droplet blocking mechanism 2 to a cough prediction device 100A. The cough prediction device 100A has an imaging means 19 and an information processing device 1 (computer), and the information processing device 1 (computer) has a communication unit 11, an input unit 12, a display unit 13, a sound output unit 14, a memory unit 15, a control unit 16 (CPU), an interface 18, etc. The output line of the interface 18 is connected to the droplet blocking mechanism 2. Details of each element other than the interface 18 are the same as those already explained, so explanations will be omitted.

[0035] As explained above, the tissue perfusion intensity output means gives a peak output value about 0.5 seconds before the onset of coughing. Using this, the cough prediction device 100A outputs an operation signal to the droplet blocking mechanism 2 via the interface 18. Upon receiving the operation signal, the droplet blocking mechanism 2 operates. This is an outline of the droplet blocking device 200A.

[0036] Next, a specific application scene of the droplet blocking device 200A according to this embodiment will be described. FIG. 7 shows a state when a doctor is interviewing a patient in a medical field. In the current medical field, it is not uncommon for partitions to be installed everywhere to prevent droplet infection, but they can be an obstacle when conducting an interview, and if the partitions are always installed, the partitions must be wiped frequently with alcohol disinfectant, and medical staff will be busy with that work. Therefore, as shown in the lower diagram of FIG. 7, the patient is automatically detected by the imaging means 19 attached in front of the seat, and the droplet blocking mechanism 2 is activated before the onset of coughing. Specifically, in the example of FIG. 7, a transparent partition such as acrylic is automatically raised. Since the partition is designed to appear only when it is actually necessary, the disinfection work is only required when the device is activated. In addition, since the partition is not usually blocking the view, the doctor and the patient can become closer and a relationship of trust between them can be fostered.

[0037] FIG. 8 shows a scene where a nurse is assisting a care recipient with eating at a nursing care site. Although it is possible to always install a partition during a doctor's interview, it is not realistic to install a partition during meal assistance. In such a scene, the droplet blocking device 200A according to the present embodiment is effective. As shown in the lower diagram of FIG. 8, the image capturing means 19 attached in front of the seat automatically detects the care recipient, and the droplet blocking mechanism 2 is activated before the onset of coughing. Specifically, in the example of FIG. 8, the air flow like an air curtain is used to prevent the spread of coughing. If a physical partition were to come down, there would be a risk of contacting the care recipient and causing injury, but with an air curtain, such concerns are eliminated.

[0038] FIG. 9 shows an application example of the droplet blocking device 200A according to the present embodiment in yet another scene. The upper figure shows a scene where several people in need of care are gathered in a nursing home, and the lower figure shows a scene of serving customers in a bar. The point here is that the subjects of observation are several people. In order to extract the image range of the neck, the imaging means 19 performs face recognition processing from images acquired over a relatively wide range. In other words, since the field of view of the imaging means 19 is relatively wide, it is possible to capture images of several people, perform face recognition, and observe several people at once. Therefore, even if there are several subjects, the droplet blocking mechanism 2 can be activated to appropriately block droplets.

[0039] (Cough prediction program and droplet blocking program) Although the embodiments of the present invention have been described as a cough prediction device or a droplet blocking device, the technical idea of ​​the present invention can also be understood as an invention of a program that causes the information processing device 1 to function. In other words, the invention can be conceived as a program that causes a cough prediction computer or a droplet blocking device computer to function as necessary means.

[0040] (Automatic infectious disease diagnosis device / system) It is also possible to construct an automatic diagnostic analysis device or system for pathological conditions with the cough prediction device of the present invention as the core. The cough prediction device monitors and analyzes the neck surface displacement and circulatory dynamics fluctuation as a sign of coughing, and in addition to this, it also monitors and analyzes and records the actual occurrence of coughing, fever, and autonomic nerve fluctuation. Specifically, a monitor camera in a waiting area such as an outpatient waiting room records the "coughing," "fever," "circulatory dynamics fluctuation," and "heart rate fluctuation" of a waiting patient, as well as motion data and blood flow information, and time-series information on the center of gravity sway by a thermograph or a pressure sensor installed in the seat. These elements are used as teacher data, and a machine-learned prediction model is used to link them to the definitive diagnosis of the patient made thereafter, allowing automatic diagnostic analysis of pathological conditions accompanied by fever and coughing in respiratory infections such as Covid-19. In other words, it is possible to make a diagnosis using the prediction of coughing, the actual result, and other data.

[0041] The cough prediction device and droplet blocking device according to the embodiments of the present invention have been described in detail above with reference to the drawings. However, the specific configurations are not limited to these examples, and the present invention includes design changes that do not deviate from the gist of the present invention. In particular, even if the cough prediction device is not combined with a droplet blocking device, it can be used in an embodiment of the present invention to avoid coughing when working face-to-face with a subject or patient during examination, diagnosis, or treatment. Regarding the calculation of the surface displacement of the observed subject and the calculation of the blood flow rate of the surface tissue of the neck skin as exemplified above, it should be well understood that the significance of the present invention is given as long as it can provide an index that can ultimately lead to tissue perfusion intensity, regardless of the embodiment. [Explanation of symbols]

[0042] 1. Information processing device 11 Communications Department 12 Input section 13 Display section 14 Pronunciation section 15 Storage section 16 Control section 161 Surface displacement calculation means 162 Blood flow calculation means 163 Cough prediction tools 17 Communication lines 18 Interface 19. Shooting Method 2. Droplet blocking mechanism

Claims

1. A cough prediction device comprising: an imaging means for imaging an observed subject; and a surface displacement calculation means for calculating a surface displacement amount of the observed subject, the surface displacement amount calculation means calculates a surface displacement amount in a horizontal direction of the neck of the subject from the image obtained by the imaging means, The cough prediction device predicts the occurrence of coughing in the observed subject by detecting a peak from a differential waveform of the surface displacement amount in the horizontal direction of the neck outputted from the surface displacement amount calculation means. A cough prediction device comprising:

2. A cough prediction device comprising: an imaging means for imaging an observation subject; a surface displacement amount calculation means for the observation subject; a blood flow rate calculation means for calculating a blood flow rate of a skin surface tissue of the neck of the observation subject; and a tissue perfusion intensity output means for the observation subject, the surface displacement amount calculation means calculates a surface displacement amount of the neck of the subject from the image obtained by the imaging means, the blood flow calculation means calculates or estimates an amount of light absorbed by hemoglobin from the image obtained by the imaging means; the tissue perfusion intensity output means outputs a tissue perfusion intensity obtained based on a product of a spatial derivative of the amount of surface displacement and a time derivative of the amount of hemoglobin absorbed light, The cough prediction device predicts the occurrence of coughing in the observed subject by detecting a peak from an output waveform of the tissue perfusion intensity output means. A cough prediction device comprising:

3. The blood flow calculation means estimates the amount of light absorbed by hemoglobin from RGB decomposition data of the image obtained from the imaging means. The cough prediction device according to claim 2 .

4. The photographing means is a plurality of cameras, The surface displacement amount calculation means generates three-dimensional image data from the images captured by the plurality of cameras, and calculates a surface displacement amount in a horizontal direction of the neck of the subject based on the three-dimensional image data. The cough prediction device according to claim 3 .

5. The photographing means is a camera, The surface displacement amount calculation means calculates a surface displacement amount in a horizontal direction of the neck of the subject based on two-dimensional image data captured by the one camera. The cough prediction device according to claim 3 .

6. the observed subjects are a plurality of people included in the photographing range of the photographing means, The photographing means recognizes and identifies the necks of the multiple subjects by face recognition. The cough prediction device according to any one of claims 1 to 5.

7. A droplet blocking device for use with the cough prediction device according to any one of claims 1 to 6, The droplet blocking device receives a cough prediction signal output by the cough prediction device and operates a droplet blocking mechanism. A splash blocking device characterized by:

8. The splash barrier mechanism is a movable partition. The splash barrier device according to claim 7 .

9. The splash blocking mechanism is an air curtain. The splash barrier device according to claim 7 .

10. A computer of a cough prediction device having an imaging means for imaging a subject to be observed, a surface displacement amount calculation means for calculating a surface displacement amount in a horizontal direction of the neck of the subject from the image obtained by the photographing means; a prediction means for predicting the occurrence of coughing in the observed subject by detecting a peak from a differential waveform of the surface displacement amount in the horizontal direction of the neck outputted from the surface displacement amount calculation means; A cough prediction program characterized by functioning as follows.

11. A computer of a cough prediction device having an imaging means for imaging a subject to be observed, a surface displacement amount calculation means for calculating a surface displacement amount of the neck of the subject from the image obtained by the imaging means; a blood flow rate calculation means for calculating or estimating the amount of light absorbed by hemoglobin from the image obtained by the imaging means; a tissue perfusion intensity output means for outputting a tissue perfusion intensity obtained based on the product of the spatial derivative of the surface displacement amount and the time derivative of the amount of hemoglobin absorbed light; a prediction means for predicting the occurrence of coughing in the observed subject by detecting a peak from the output waveform of the tissue perfusion intensity output means; A cough prediction program characterized by functioning as follows.

Images