Apparatus for detecting vital functions, control unit and pulse wave sensor

Abstract

An apparatus for detecting vital functions has a pulse wave sensor attachable to a body and a control unit. The control unit checks if amplitude of pulse wave signals produced from the pulse wave sensor varies. The control unit further checks if a large change in the amplitude during a systolic phase of a pulse wave corresponding to the systolic phase of the heart. If a first large change in the amplitude during a diastolic phase of a pulse wave corresponding to the diastolic phase of the heart, it is highly probable that a motion artifact has occurred. Therefore, a motion artifact flag is set. Next, it is checked if the amplitude in the next diastole is changing by more than 30%. If it is presumed that the occurrence of cough is highly probable, a cough flag is set. If it is neither the motion artifact nor the cough, then a yawn flag is set.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description made with reference to the accompanying drawings. In the drawings:

FIG. 1A is a schematic diagram illustrating an apparatus for detecting the conditions of a body and a mounting device therefor according to an embodiment, and FIG. 1B is an enlarged schematic view illustrating a pulse wave sensor used in the embodiment;

FIG. 2A is a graph illustrating a waveform of pulse waves, and FIG. 2B is a graph illustrating a respiration waveform;

FIG. 3A is a graph illustrating a waveform of pulse waves of when there is a motion artifact, and FIG. 3B is a graph illustrating a respiration waveform of when there is a motion artifact;

FIG. 4A is a graph illustrating a waveform of pulse waves of when there is a strong motion artifact, FIG. 4B is a graph illustrating a respiration waveform of when there is a strong motion artifact, FIG. 4C is a graph illustrating a waveform of pulse waves of when there is a motion artifact of an intermediate degree, FIG. 4D is a graph illustrating a respiration waveform of when there is a motion artifact of an intermediate degree, FIG. 4E is a graph illustrating a waveform of pulse waves of when there is a weak motion artifact, and FIG. 4F is a graph illustrating a respiration waveform of when there is a weak motion artifact;

FIG. 5A is a graph illustrating a waveform of pulse waves of when a cough has occurred, and FIG. 5B is a graph illustrating a respiration waveform of when a cough has occurred;

FIG. 6A is a graph illustrating a waveform of pulse waves of when a strong cough has occurred, FIG. 6B is a graph illustrating a respiration waveform of when a strong cough has occurred, FIG. 6C is a graph illustrating a waveform of pulse waves of when a cough of an intermediate degree has occurred, FIG. 6D is a graph illustrating a respiration waveform of when a cough of an intermediate degree has occurred, FIG. 6E is a graph illustrating a waveform of pulse waves of when a weak cough has occurred, and FIG. 6F is a graph illustrating a respiration waveform of when a weak cough has occurred;

FIG. 7A is a graph illustrating a waveform of pulse waves of when a yawn has occurred, and FIG. 7B is a graph illustrating a respiration waveform of when a yawn has occurred;

FIG. 8A is a graph illustrating a waveform of pulse waves of when a strong yawn has occurred, FIG. 8B is a graph illustrating a respiration waveform of when a strong yawn has occurred, FIG. 8C is a graph illustrating a waveform of pulse waves of when a yawn of an intermediate degree has occurred, FIG. 8D is a graph illustrating a respiration waveform of when a yawn of an intermediate degree has occurred, FIG. 8E is a graph illustrating a waveform of pulse waves of when a weak yawn has occurred, and FIG. 8F is a graph illustrating a respiration waveform of when a weak yawn has occurred;

FIG. 9 is a flowchart illustrating processing for setting flags T1, S1 and A1;

FIG. 10 is a flowchart illustrating processing for setting flags T2, S2 and A2;

FIG. 11 is a flowchart illustrating processing for setting flags T3, S3 and A3; and

FIG. 12 is a flowchart illustrating control processing based on a comprehensive determination of the motion artifact, coughing and yawning.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

First, a biometric detection apparatus for detecting conditions of a body is described with reference to one embodiment shown in FIGS. 1A and 1B. The apparatus detects vital functions such as cough or yawn by using a pulse wave sensor 1. This sensor 1 is attached to a portion of a human body 100 such as a finger, a palm or a wrist, where motion is small. The apparatus further uses a control unit 3, which drives the pulse wave sensor 1 and processes the outputs from the pulse wave sensor 1.

Here, the pulse wave sensor 1 is an optical sensor of the reflection type (opto-capacitive pulse wave sensor) comprising a light-emitting element (e.g., light-emitting diode: green LED) 5, a light-receiving element (e.g., photodiode: PD) 7, and a transparent lens 9 which permits light to pass through and also efficiently receives light.

The pulse wave sensor 1 has a ring-like buffer member (e.g., a sponge having a rugged end) 11 that serves as a spacer surrounding the lens 9 on the skin side so that the lens 9 will not be pushed onto the skin 100a with an excess of pressure, and a spring 13 on the rear end side of the pulse wave sensor 1. This makes it possible to set the pressure for pushing the lens 9 onto the skin 100a to be not larger than 10 gw/cm². The pulse wave sensor 1 is fixed to the wrist or the like by using a band 15. Therefore, the spring 13 is arranged between the band 15 and the pulse wave sensor 1.

When the pulse wave sensor 1 is to be used, a driving electric power is supplied from a drive unit 17 in the control unit 3 and light is projected to the human body from the light-emitting element 5. Part of the light hits capillary vessels (capillary arteries) in the human body, is absorbed mostly by hemoglobin in the blood flowing through the capillary vessels, while rest of the light scatters repetitively and partly falls on the light-receiving element 7. Due to the pulsation of blood at this moment, the amount of hemoglobin in the capillary vessels vary periodically like a wave and, therefore, the light absorbed by the hemoglobin varies like a wave, too.

As a result, the amount of light absorbed by the capillary vessels varies and, accordingly, the amount of light received or detected by the light-receiving element 7 varies. A change in the amount of the received light is output as pulse wave data (sensor output which is a voltage signal representing the pulse wave) to the control unit 3.

The control unit 3 includes the drive unit 17, a detector unit 19 that receives a sensor output, a microcomputer 21 which produces a control signal to the drive unit 17 and to an output unit 23 and receives signals from the detector unit 19 and from an input unit 25 to execute various processing, the output unit 23 that sends control signals to various actuators 27, and the input unit 25 that receives a signal from a manual switch 29.

The microcomputer 21 is an electronic circuit including known CPU, ROM, RAM and the like, and incorporates a program for detecting the coughing and yawning by processing pulse wave signals applied from the pulse wave sensor 1.

The pulse wave sensor 1 may be attached to any part of the body but is desirably attached to an arm, hand, finger, forehead or foot that is less affected by the motion artifact.

Next, the principle for detecting the coughing and yawning will be described with reference to FIGS. 2A to 8F. In these figures, the abscissa represents the passage of time and the ordinate represents the magnitude (intensity or variation) of the signals.

Referring first to FIG. 2A, the pulse wave sensor 1 produces pulse wave signals having peaks corresponding to the systolic phase and to the diastolic phase of the heart. Specifically, varying peaks appear in the upper side in the figure during the systolic phase of the heart, and varying peaks appear in the lower side in the figure during the diastolic phase of the heart.

During the ordinary calm state, i.e., when there is no motion artifact, cough or yawn, and there is only a very mild variation of a large period due to the motion of the blood vessels, variation of the blood pressure in the blood vessels is slow. As a result, the peaks of signals on p-diastole side do not become lower than a predetermined low-limit level LL (line corresponding to the lowest blood pressure).

As a method of setting the lower-limit level LL, there can be exemplified an approximated line found, for example, from a plurality of peaks on p-diastole side representing the lowest blood pressure.

A line connecting the centers of the upper and lower peaks of pulse wave signals is called the base level BL (index related to an average blood pressure). Further, a line connecting the peaks of signals on p-systole is called a pulse wave envelope (first envelope) A, and a line connecting the peaks of the first envelope is called a double envelope (second envelope) B. Referring to FIG. 2B, a waveform found by subtracting the second envelope B from the first envelope A is called a respiration waveform (respiration curve). The respiration waveform is a signal corresponding to the intra-thoracic pressure.

The principle of the processing of the embodiment conducted by using the above signals is described next.

(1) Method of Detecting Motion Artifact

Referring to FIG. 3A, when a motion artifact has occurred, the pulse wave on p-diastole side, first, so varies as to becomes lower than (fall below) the lower-limit level LL from the ordinary (calm) pulse waves, i.e., from the state of the regular sinusoidal pulse wave signals. The frequency of pulse wave signals of this period becomes smaller than the frequency of the ordinary pulse wave signals (of when there is no motion artifact) (e.g., state of high-frequency noise) and the waveforms, in many cases, greatly vary from the sinusoidal waves.

If the signal on p-diastole side exceeds the lower-limit level LL, therefore, it can be so determined that the motion artifact has occurred. Here, if the waveform varies or the frequency of pulse wave signals becomes small, it is highly probable that the motion artifact has occurred.

Referring to FIGS. 4B, 4D and 4F, further, the respiration waveform found from the peaks on p-systole differs depending upon the intensity (strength) of the motion artifact. Therefore, the intensity of motion artifact can be determined from the state of the respiration waveform. If the motion artifact is great, for example, the respiration waveform falls greatly and a curve thereof becomes sharp as shown in FIG. 4B. If the motion artifact is medium, the respiration waveform falls greatly but a curve thereof becomes loose. If the motion artifact is small, the respiration waveform falls little.

Accordingly, the motion artifact can be accurately detected by each of the following determining methods (algorithms) or by a combination of the determining methods. The following determining methods have all been confirmed through the practically conducted experiments (the same also holds for the coughing and yawning).

In case the amplitude on p-diastole side has increased (i.e., in case the peaks of amplitude on p-diastole side have exceeded a predetermined level or have become smaller than the lower-limit level LL, it is determined that the motion artifact has occurred (motion artifact flag T1 is set as described later).

In case the waveform of pulse waves has varied from the sinusoidal wave, it is determined that the motion artifact has occurred. Whether the waveform has varied from the sinusoidal waves can be determined based on a correlation to the waveform of the ordinary (e.g., preceding) pulse wave. For example, it can be regarded that the waveform has varied if a coefficient of the correlation is not larger than 0.7.

In case the amplitude of the respiration waveform has varied and the waveform of pulse waves has varied from the sinusoidal waves, it can be determined that the motion artifact has occurred. Here, if the waveform has varied from the sinusoidal waves can be determined based on a correlation to the waveform of the ordinary pulse waves (for example, it can be regarded that the waveform has varied if a coefficient of the correlation is not larger than 0.6).

In case there is a change in the amplitude AW2 of the base level BL divided by the amplitude BW2 of pulse waves (FIG. 7), the change is a nature of a single-shot and the time of change is as short as 0 to 4 seconds or as long as 12 seconds or more, it is determined that the motion artifact has occurred (motion artifact flag T3 is set as described later).

A change from the sinusoidal waves or a change in the amplitude can be determined from the analysis of variation in the peak-to-peak pitch or from the analysis of chaos in addition to utilizing the correlation (the same holds hereinafter).

(2) Method of Detecting Cough

Referring to FIG. 5A, if the cough has occurred, the amount of terminal blood temporarily increases due to instantaneous contraction of muscles accompanying the coughing. The pulse wave on p-systole side, first, increases greatly from the state of ordinary pulse wave signals.

The coronary veins are compressed by the abdominal muscles that have tensed due to coughing, the blood returning to the heart instantaneously decreases. The heart blows out the blood in decreased amounts. Accordingly, the pulse wave on p-systole side immediately after the rise falls greatly in excess of the lower-limit level. The time of coughing is so short that there is no change in the amplitude of the pulse wave signals, and the base level readily restores. In the case of the coughing, usually, no change is seen in the frequency and there is almost no change in the shape of the sinusoidal waves, either.

In case the cough occurs, further, a double waveform (double triangular wave) of a characteristic acute angle appears on the respiration waveform obtained from the pulse waves as shown in FIG. 5B. A large ratio (AW1/BW1) of the longitudinal width (amplitude) AW1 and the transverse width (period) BW1 of the double triangular wave represents an intense coughing in which the intrathoracic pressure sharply changes within a short period of time.

What makes the double triangular wave may be determined relying, for example, upon “if the determining line is exceeded that is separated away from the ordinary line of the respiration waveform by more than a predetermined percentage”.

Therefore, if the pulse wave on p-systole side, first, greatly increases (in excess of the predetermined upper-limit level) and, immediately thereafter, the pulse wave on p-diastole side exceeds the lower-limit level, it can be determined that the cough has occurred. Further, if the double triangular wave appears on the respiration waveform, it can be determined that the cough has occurred.

Referring to FIGS. 6A to 6F, further, the pulse wave signal and the respiration waveform differ depending upon the intensity of coughing. From the states thereof, therefore, the intensity of coughing can be checked if the coughing is intense, for example, the pulse wave signals vary greatly in the up-and-down direction as shown in FIG. 6A, the respiration waveform falls greatly as shown in FIG. 6B, and the ratio AW1/BW1 becomes large. If the coughing is of the medium degree, further, the pulse wave signals vary up and down to a slightly large extent as shown in FIG. 6C, the respiration waveform falls to a medium degree as shown in FIG. 6D, and the ratio AW1/BW1 is of a medium degree. Further, if the coughing is weak, the pulse wave signals vary up and down to a small degree, and the respiration waveform falls mildly.

Thus, the coughing can be accurately detected by the combination of one or two or more kinds of the following determining methods.

If the amplitude on p-systole side increases (i.e., if the peak of amplitude increases on p-systole side) and the amplitude on p-diastole side increases (i.e., if the peak of amplitude is lowered on p-diastole side), it is determined that the coughing has occurred (cough flag S1 is set as described later).

If the amplitude on p-systole side increases (i.e., if the peaks increase on p-systole side only) without causing the waveform of pulse waves to be varied from the sinusoidal waves, it can be regarded that the cough has occurred.

If the amplitude of the respiration waveform varies (increases by 30% above the normal value) and a double waveform of an acute angle appears on the respiration waveform, it is determined that the coughing has occurred (cough flag S2 is set as described later).

If (amplitude AW2 of the base level/amplitude BW2 of the pulse waves) of pulse waves varies little (decreased by 30% below the normal) and the time of change is shorter than a predetermined level (e.g., shorter than 4 seconds), it is determined that the cough has occurred.

Here, it is important that in detecting the coughing, the motion artifact is determined, too. Even in case it is determined that the cough has occurred by the above determining method, it is determined that the cough has occurred only when it is so determined that there is no motion artifact. This is to isolate the motion artifact from the cough to render an accurate determination.

(3) Method of Detecting Yawn (Deep Respiration)

Referring to FIG. 7A, if the yawn (or deep respiration) has occurred, the amount of terminal blood temporarily increases to some extent due to mild contraction of muscles accompanying the yawn. Therefore, the systolic pulse wave, first, increases to some extent from the ordinary state of pulse wave signals.

Thereafter, the coronary veins are compressed by the abdominal muscles that have tensed due to coughing, the blood returning to the heart gradually decreases. Therefore, the heart blows out the blood in decreased amounts for several seconds. Accordingly, the base level of the pulse wave signals gradually decreases. At this moment, the base level gradually decreases as designated at AW2 in proportion to the intensity of yawning, and the amplitude BW2 of the pulse waves decreases.

The reflux of blood to the heart is squeezed for several seconds due to the yawning, and about 10 seconds are needed before the pulse wave is recovered. In the case of the yawning, the sinusoidal waves of pulse wave signals do not change much, and a double triangle of an obtuse angle having a wide bottom side is seen on the respiration waveform.

It can be determined that the yawn has occurred based on a decrease in the base level of pulse wave signals, on a decrease in the respiration waveform (e.g., based on a decreased state lasting for 4 to 12 seconds) or on a decrease in the amplitude of pulse waves.

Referring to FIGS. 8A to 8F, further, the pulse wave signal and the respiration waveform differ depending upon the intensity of yawning. From the states thereof, therefore, the intensity of yawning can be checked if the yawning is intense, for example, the pulse wave signals and the respiration waveform fall greatly as shown in FIGS. 8A and 8B. Besides, the width of fall is great (4 to 12 seconds) and the amplitude of pulse wave signals becomes considerably small after the fall as shown in FIG. 8A. If the yawning is of the intermediate degree, further, the pulse wave signals and the respiration signals fall to a medium degree as shown in FIGS. 8C and 8D. Besides, the width of fall is of a medium degree and the amplitude of the pulse wave signals does not change much as shown in FIG. 8C. Further, if the yawning is weak, the pulse wave signals and respiration signals fall little or the width of fall is small as shown in FIGS. 8E and 8F, and the amplitude of the pulse wave signals does not change much as shown in FIG. 8E.

Therefore, the yawning can be accurately detected by the combination of one or two or more kinds of the determining methods.

If the waveform on p-systole increases (i.e., if the peak of pulse waves during systole increases) but the peak of pulse waves on the side of the next diastole does not become smaller than that of the ordinary case, it is determined that the yawn has occurred (yawn flag A1 is set as described later).

If the amplitude on p-systole side increases and, thereafter, if the whole amplitude BW2 of pulse waves decreases without causing the waveform of pulse waves to vary from the sinusoidal waves, it is determined that the yarn has occurred.

If the amplitude of the respiration waveform varies and a double curve of an obtuse angle appears on the respiration waveform, it is determined that the yawn has occurred (yawn flag A2 is set as described later).

If the ratio AW2/BW2 (amplitude AW2 of the base level divided by amplitude BW2 of the pulse waves) of pulse waves varies greatly (larger by 30% or more above the normal) and the time of change is longer than a predetermined value (e.g., longer than 8 seconds), it is determined that the yawn has occurred.

It is important that in detecting the yawn, the motion artifact is determined, too. Even in case it is determined that the yawn has occurred by the above determining method, it is determined that the yawn has occurred only when it is so determined that there is no motion artifact. This is to separate the motion artifact from the yawning to render an accurate determination.

Next, processing executed by the control unit 3 based on the above principles will be described with reference to flowcharts of FIGS. 9 to 12.

These processing employs some of the above determining methods to finally determine the occurrence of the motion artifact, coughing and yawning.

(I) Referring to FIG. 9, it is checked at step (S) 100 if there is a change in the amplitude of pulse wave signals produced by the pulse wave sensor 1. Specifically, it is checked if the amplitude of pulse wave signals is varying up and down by not less than 30% beyond the up-and-down amplitude of the ordinary pulse wave signals, or if the amplitude on p-systole side (or the amplitude on p-diastole side) is varying by not less than 30% beyond the amplitude of the ordinary pulse wave signals on p-systole side (or beyond the amplitude on p-diastole side). If the determination is affirmative, the routine proceeds to step 110. If the determination is negative, the processing once ends.

That is, if the amplitude exceeds by more than 30% beyond the amplitude of the ordinary pulse wave signals, it can be presumed that motion artifact, coughing, yawning or the like, which is different from the ordinary calm state, has occurred.

At step 110, it is checked if a large change in the amplitude detected at step 100 is on p-systole side. If the determination is affirmative, the routine proceeds to step 120, if the determination is negative, the routine proceeds to step 130.

When a first large change in the amplitude (peak protruding downward) is on p-diastole side, it is highly probable that the motion artifact has occurred as shown in FIG. 3A. Therefore, at step 130, a motion artifact flag T1 is set to represent the above fact, and the processing once ends.

If the first large change in the amplitude is on p-systole side, the probability of motion artifact is low. It is, therefore, probable that the cough or the yawn has occurred. It is checked at step 120 if the amplitude in the next diastole is varying by more than 30%. If the determination is affirmative, the routine proceeds to step 140. If the determination is negative, the routine proceeds to step 150.

If the amplitude in the next diastole is varying by more than 30%, as shown in FIG. 5A, it can be presumed that the cough has occurred highly probably. That is, the pulse wave signals are, first, greatly varying on p-systole side and, immediately thereafter, are greatly varying on p-diastole side. Thus, it is presumed that the probability of cough is high. A cough flag S1 is set at step 140 to represent the above fact, and the processing once ends.

If neither the motion artifact nor the cough is occurring, based on the elimination method, it is presumed that the above variation in the pulse wave signals at step 100 above is caused by the yawning. Therefore, a yawn flag A1 is set at step 150 to represent the above fact, and the processing once ends.

If there is a large change in the pulse wave signals, the flag of any one of the motion artifact, coughing or yawning can be set based on the state where the pulse wave signals are varying.

(II) Referring to a flowchart of FIG. 10, a respiration waveform (respiration curve) is generated or formed at step 200 based on the pulse wave signals produced from the pulse wave sensor 1.

At subsequent step 210, it is checked if there is a double triangular wave on the respiration curve. Specifically, it is checked if there is a W-shaped waveform (waveform having a protrusion on the upper side) under the predetermined determining level of the respiration curve as shown in FIG. 5B. If the determination is affirmative, the routine proceeds to step 220. If the determination is negative, the routine proceeds to step 230.

If there is no double triangular wave, it is presumed that there is the motion artifact and the motion artifact flag T2 is set at step 230 to once end the processing.

At step 220, it is checked if the double triangular wave is a double triangular wave of an obtuse angle of more than a predetermined time width BW1 (e.g., 4 seconds at the predetermined determining level PDL). Namely, it is checked here if the double triangular wave has an obtuse angle by checking the predetermined width BW1. If the determination is affirmative, the routine proceeds to step 250. If the determination is negative, the routine proceeds to step 240.

Here, the addition of a condition “a change in the amplitude of the pulse wave curve is greater than the normal value by more than 30%” further improves the accuracy of determination.

If the double triangular wave has an acute angle as shown in FIG. 5B, it is highly probable that the cough has occurred. Therefore, a cough flag S2 is set at step 240 to represent the above fact, and the processing once ends.

If the double triangular wave is not of an acute angle, it is highly probable that the yawn has occurred instead of the cough (without motion artifact). Therefore, a yawn flag A2 is set at step 250 to represent the above fact, and the processing once ends.

If there is a large change in the respiration curve, a flag of any one of the motion artifact, coughing or yawning is set based on the state of change in the respiration curve.

(III) Referring to a flowchart of FIG. 11, a base level (line) BL is formed based on the pulse wave signals produced from the pulse wave sensor 1.

At subsequent step 310, it is checked if the base level of the pulse waves is fluctuating. Specifically, as shown in FIG. 7A, it is checked if ration AW2/BW2 (amplitude AW2 of the base level (herein, amplitude from an average value of the base level in normal operation) divided by amplitude BW2 of each pulse wave signal) is greater than a predetermined determining value (e.g., if the change is greater than the normal value by more than 30%). If the determination is affirmative, the routine proceeds to step 320. If the determination is negative, the routine proceeds to step 330.

If there is no fluctuation in the base level BL, a cough flag S3 is set at step 330 presuming that the cough has occurred, and the processing once ends.

It is checked at step 320 if the base level is fluctuating periodically (at a period of, for example, 6 to 15 seconds). If the determination is affirmative, the routine proceeds to step 340. If the determination is negative, the routine proceeds to step 330.

If the period of fluctuation of the base level is long, it is so determined that the fluctuation is arising from the motion of blood vessels. A blood vessel motion flag K1 is set at step 340 to represent the above fact, and the processing once ends.

At step 330 it is checked if the fluctuation of the base level is of a nature of a single-shot and is within a predetermined period (e.g., 4 to 12 seconds). If the determination is affirmative, the routine proceeds to step 360. If the determination is negative, the routine proceeds to step 350.

If the fluctuation of the base level is of the nature of a single-shot and is within the predetermined period (e.g., 4 to 12 seconds) as shown in FIG. 7B, it can be presumed that the yawn has occurred. Therefore, a yawn flag A3 is set at step 360, and the processing once ends.

If the fluctuation is not a single-shot or more than the predetermined period, a motion artifact flag T3 is set at step 350 presuming that the motion artifact may have occurred, and the processing once ends.

The fluctuation of the base level is thus analyzed, and the flag of any one of the motion artifact, coughing or yawning is set.

(IV) Referring to a flowchart of FIG. 12, vital function, such as respiration and pulsation is comprehensively determined based on the flags.

For example, when three kinds of motion artifact flags T1 to T3 are set, it can be reliably determined that the motion artifact has occurred. Further, when one or two flags are set among the three kinds of motion artifact flags T1 to T3, it may be so determined that the motion artifact has occurred. The reliability of motion artifact increases with an increase in the kinds of the flags that are set. Among the three kinds of motion artifact flags T1 to T3, there is a flag which highly represents the motion artifact (e.g., motion artifact flag T1). If this flag is set, it may be so determined that the motion artifact has occurred. Alternatively, a large threshold value may be set to the counter for the motion artifact flag that represents a high probability. The motion artifact may be determined based on a total value of the motion artifact flags counted by the counter.

What is described above holds even for the three kinds of cough flags S1 to S3 and yarn flags A1 to A3. If the cough flag S1 is set, it is most probable that the cough has occurred. If the yawn flag A3 is set, it is most probable that the yawn has occurred.

Even if it is determined that the cough or the yawn has occurred based on the cough flags S1 to S3 or the yawn flags A1 to A3, it may be often determined that the motion artifact has occurred based on the motion artifact flags T1 to T3. In such a case, it is not determined that the cough or the yawn has occurred but, instead, it is determined that the motion artifact has occurred, reducing erroneous determination.

At next step 410, actuators are controlled, such as the air conditioner, navigation system, seats, seatbelts, etc. based on the above determined results, and the processing once ends.

The following processing can be employed for controlling the actuator.

For example, disorders of respiration such as asthma and emphysema of the lungs, disorders of inspiration such as rhinitis and nasal congestion, and cold can be presumed from the frequency and intensity of coughing, number of beats, number of breaths, a curve of respiration and fluctuation in the respiration (respiration signals). The air conditioner, air cleaner, navigation system, audio equipment, seats and seat belts are adjusted depending upon the symptoms that are presumed.

Specifically, the following control operations may be executed, for example.

If the cough is detected while an alarm of ear pollen has been issued, the conditioned air in the compartment is switched to the air-recirculation mode (internal circulation mode), and an auxiliary filter such as an air-purifier is operated. If the coughing occurs more frequently than usual, it is so presumed that the person is having a slight cold, the compartment temperature and the humidity are set to be slightly higher. Further, if the yawn is frequently detected, information is produced from the navigation system to stimulate the driver's brain to prevent him from becoming sleepy. If the yawn still occurs too frequently, the tension of the seat belt is varied to stimulate the body to keep the person awake.

In addition to the above results of presumption, other pulse wave data (pulse waveform, fluctuation in the pulse wave amplitude, pulse rate, variation in the interval between pulses, etc.) may be added or any other body data signals (blood pressure, electrocardiograph, electromyograph, camera image) may be added to presume the condition of the body at that moment to correct the control of air-conditioner, air-purifier, navigation system, audio equipment, seats and seatbelts.

The present invention is not limited to the above embodiments only but can be put into practice in a variety of other ways.

Claims

1. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andmotion artifact determining means which determines an occurrence of a motion artifact when an amplitude during a diastolic phase of the pulse (on p-diastole side for short) corresponding to a diastole of a heart has exceeded a predetermined limit level that corresponds to a diastolic blood pressure of the body.

2. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andrespiration waveform calculation means for finding a respiration waveform that represents a respiration condition from the pulse waves detected by the sensing means; andcough determining means which determines an occurrence of cough when a double triangular wave is detected, wherein the double triangular wave has an acute angle in which two peaks of respiration waveforms are consecutively exceed a predetermined level.

3. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andcough determining means which determines an occurrence of cough when an amplitude on p-systole side corresponding to a systolic phase of a heart has exceeded a predetermined level and when the amplitude on p-diastole side corresponding to a succeeding diastolic phase of the heart has exceeded a predetermined level.

4. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andcough determining means which determines an occurrence of cough when an amplitude on p-systole side corresponding to a systolic phase of a heart has increased by more than a predetermined level without causing a waveform of the pulse waves to be varied from the waveform of ordinary pulse waves.

5. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andcough determining means which determines an occurrence of cough when a change in a ratio AW2/BW2 of an amplitude AW2 of a base level of the pulse waves divided by an amplitude BW2 of the pulse waves is within a predetermined value and when a time of change is within a predetermined period.

6. An apparatus for detecting vital functions: motion artifact determining means which determines an occurrence of motion artifact based on the pulse waves; andcough determining means which determines an occurrence of cough when no motion artifact is detected by the motion artifact determining means while determining an occurrence of cough based on at least one of the cough determining means according to any one of claims 2 to 5.

7. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andyawn determining means which determines an occurrence of yawn when a base level of the pulse waves is lowered over a predetermined period corresponding to yawning.

8. The apparatus according to claim 7, wherein the yawn determining means determines the occurrence of yawn when an amplitude of the pulse waves has become smaller than a predetermined level within a period in which the base level of the pulse waves is lower than a predetermined level.

9. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andyawn determining means which determines an occurrence of yawn when an amplitude on p-systole side corresponding to a systolic phase of a heart exceeds a predetermined level but when the amplitude of a succeeding pulse waves signals on p-diastole side corresponding to a diastolic phase of the heart does not become smaller than a predetermined level.

10. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body; andyawn determining means which determines an occurrence of yawn when a whole amplitude of pulse waves of a sum of amplitudes on p-systole side and of amplitudes on p-diastole side becomes smaller than a predetermined level after the amplitudes on p-systole side corresponding to the systolic phase of the heart has exceeded a predetermined level without causing the waveform of the pulse waves to be varied from the waveform of ordinary pulse waves.

11. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body;respiration waveform calculation means which finds a respiration waveform that represents a respiration condition from the pulse waves; andyawn determining means which determines an occurrence of yawn when the amplitude of the respiration waveform is not smaller than a predetermined level and when a double triangular wave of an obtuse angle is detected in which two peaks of the respiration waveform consecutively exceed a predetermined level.

12. An apparatus for detecting vital functions: sensing means which is attachable to the body and produces pulse wave signals corresponding to pulse waves of the body;yawn determining means which determines an occurrence of yawn when a change in a ratio AW2/BW2 of an amplitude AW2 of a base level of the pulse waves divided by an amplitude BW2 of the pulse waves is not smaller than a predetermined level and when a time of change is within a predetermined period corresponding to a yawning.

13. An apparatus for detecting vital functions: motion artifact checking means which determines an occurrence of motion artifact; andyawn determining means which determines an occurrence of yawn when no motion artifact is detected by the motion artifact determining means while determining the occurrence of yawn based on at least one of the yawn determining means according to any one of claims 7 to 12.

14. The apparatus according to claim 13, wherein the motion artifact determining means determines an occurrence of motion artifact when an amplitude on p-diastole side corresponding to a diastolic phase of a heart has exceeded a predetermined limit level that corresponds to a diastolic blood pressure of the body.

15. An apparatus according to claim 13 further comprising: an actuator; andcontrol means which controls the actuator based on results of determination when the occurrence of cough or yawn is determined.

16. A pulse wave sensor for detecting vital functions: an optical device for projecting light onto a skin of the body to measure pulse waves of the body; anda buffer member arranged between the optical device and the skin so that the optical device of the skin side does not come in contact with the skin or pushes the skin with a pressure which is less than a predetermined level.

17. The pulse wave sensor according to claim 16, wherein the buffer member is a sponge having a rugged pattern on the skin side.

18. A pulse wave sensor for detecting vital functions: an optical device for projecting light onto a skin of the body to measure pulse waves of the body; andelastic means arranged on the optical device on a side opposite to the skin for mounting the optical device on a surface of the body, so that the optical device pushes the skin with a pressure which is less than a predetermined level.

19. The pulse wave sensor according to claim 18, wherein the elastic means includes a band for detachably fixing the optical device to the body.

20. The pulse wave sensor according to claim 19, wherein the elastic means further includes an elastic member arranged between the band and the optical device.