 |
Description  |
|
|
BACKGROUND AND SUMMARY OF THE INVENTION
The present invention relates to an electronic sphygmomanometer or to an
automatic indirect blood pressure measuring device.
An automatic indirect blood pressure measuring device has been developed,
which utilizes the Korotkoff sounds, or blood vessel sounds derived from a
microphone positioned under an arm cuff. In such a device, first a high
pressure is applied to the arm cuff through the use of an air pump and,
then, the pressure applied to the arm cuff is gradually reduced at a rate,
for example, 2 to 4 mmHg/sec. During the reduction procedure of the
applied pressure, the Korotkoff sounds appear at the systolic pressure
point, and the Korotkoff sounds disappear at the diastolic pressure point.
A typical control system for an electronic sphygmomanometer is disclosed in
U.S. Pat. No. 4,273,136, "ELECTRONIC SPHYGMOMANOMETER" issued on June 16,
1981.
In such an electronic sphygmomanometer, it is strictly required that the
Korotkoff sounds are distinguished from noises in order to ensure an
accurate detection of the blood pressure. In the conventional system, the
microphone output signal is passed through a low-pass filter or a bandpass
filter, and the filter output signal is applied to a comparator to
determine the peak value of the obtained signal. However, the frequency
characteristics of the Korotkoff sounds are variable depending upon the
person to be measured. Especially, if a person has hardened arteries or
thick subcutaneous fat, the determination as to whether the microphone
output signal is the Korotkoff sounds or not is very difficult.
Accordingly, an object of the present invention is to provide a detection
system for an electronic sphygmomanometer, which ensures an accurate
detection of the blood pressure.
Another object of the present invention is to provide a digitally
controlled detection system for an electronic sphygmomanometer, which
ensures an accurate detection of the Korotkoff sounds.
Other objects and further scope of applicability of the present invention
will become apparent from the detailed description given hereinafter. It
should be understood, however, that the detailed description and specific
examples, which indicating preferred embodiments of the invention, are
given by way of illustration only, since various changes and modifications
within the spirit and scope of the invention will become apparent to those
skilled in the art from this detailed description.
To achieve the above objects, pursuant to an embodiment of the present
invention, the microphone output is applied to an A/D converter via a
filter. The thus obtained digital signal is introduced into a
microcomputer to check the characteristics of the digital signal. More
specifically, the digital signal is applied to a determination circuit at
a predetermined interval. The determination circuit first detects the
leading edge peak and the trailing edge peak of the output signal derived
from the A/D converter. Then, the determination is conducted as to whether
the value difference between the leading edge peak and the trailing edge
peak is greater than a preselected value, and as to whether the time
interval between the peaks is within a preselected period of time.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will be better understood from the detailed
description given hereinbelow and the accompanying drawings which are
given by way of illustration only, and thus are not limitative of the
present invention and wherein:
FIG. 1 is a schematic block diagram of an embodiment of an electronic
sphygmomanometer according to the present invention;
FIGS. 2(A), 3(A), 4(A), 5(A) and 6(A) are graphs showing output signals of
a pickup sensor included in the electronic sphygmomanometer of FIG. 1 when
a generally healthy person is being measured;
FIGS. 2(B), 3(B), 4(B), 5(B) and 6(B) are graphs showing output signals of
a filter included in the electronic sphygmomanometer of FIG. 1 when the
pickup sensor outputs of FIGS. 2(A), 3(A), 4(A), 5(A) and 6(A) are applied
to the filter, respectively;
FIGS. 7(A), 8(A), 9(A), 10(A) and 11(A) are graphs showing output signals
of a pickup sensor included in the electronic sphygmomanometer of FIG. 1
when a person having arteriosclerosis is being measured;
FIGS. 7(B), 8(B), 9(B), 10(B) and 11(B) are graphs showing output signals
of a filter included in the electronic sphygmomanometer of FIG. 1 when the
pickup sensor outputs of FIGS. 7(A), 8(A), 9(A), 10(A) and 11(A) are
applied to the filter, respectively;
FIGS. 12(A), 13(A), 14(A), 15(A) and 16(A) are graphs showing output
signals of a pickup sensor included in the electronic sphygmomanometer of
FIG. 1 when a person having thick subcutaneous fat is being measured;
FIGS. 12(B), 13(B), 14(B), 15(B) and 16(B) are graphs showing output
signals of a filter included in the electronic sphygmomanometer of FIG. 1
when the pickup sensor outputs of FIGS. 12(A), 13(A), 14(A), 15(A) and
16(A) are applied to the filter, respectively;
FIGS. 17, 17(A), 17(B), 17(C), 17(D) and 17(E) are flow charts for
explaining an operational mode of the electronic sphygmomanometer of FIG.
1;
FIG. 18 is an enlarged graph of the filter output signal shown in FIG.
3(B);
FIGS. 19(A) and 19(B) are enlarged graphs of the filter output signal shown
in FIG. 16(B);
FIG. 20 is a block diagram of another embodiment of an electronic
sphygmomanometer according to the present invention;
FIG. 21 is an enlarged graph of an output signal of a filter included in
the electronic sphygmomanometer of FIG. 20;
FIG. 22 is a schematic block diagram of still another embodiment of an
electronic sphygmomanometer according to the present invention;
FIG. 23 is a schematic block diagram of yet another embodiment of an
electronic sphygmomanometer according to the present invention; and
FIG. 24 is a schematic block diagram of a further embodiment of an
electronic sphygmomanometer according to the present invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 schematically shows an embodiment of an electronic sphygmomanometer
of the present invention, which includes a pickup sensor 10 for detecting
the sound or vibration derived from the artery. An output signal of the
pickup sensor 10 is applied to a filter 12 which is preferably a lowpass
filter below 40 Hz or a bandpass filter of 10 Hz to 40 Hz. An output
signal of the filter 12 is introduced into an analog-to-digital converter
14 which develops a digital signal of eight (8) bit construction in
response to the level of the signal applied thereto. The thus developed
digital signal is introduced into a determination system 16 which performs
the determination as to whether the introduced signal is derived from the
Korotkoff sounds. If an affirmative answer is obtained, the determination
system 16 develops a determination output at an output terminal 160.
The determination system 16 includes a central processor unit 18, a read
only memory 20 and a random access memory 22 incorporated in a single chip
microcomputer. The read only memory 20 stores program orders for
conducting the programmed operation, and constants .alpha., .beta. and
.gamma. for controlling the operation of the electronic sphygmomanometer
of the present invention. The random access memory 22 includes register
sections Ar, Br and Cr, and a flag section Fr. The determination system 16
further includes a timer 24 (T.sub.1 and T.sub.2).
The determination system 16 introduces the digital signal developed from
the analog-to-digital converter 14 in a predetermined interval. The
determination system 16 functions to detect peaks of the digital signal in
the level increasing process and in the level decreasing process. The
determination system 16 performs the determination as to whether the level
difference between the peaks is greater than a predetermined value and as
to whether a time interval between the peaks is within a preselected
period of time, thereby recognizing the Korotkoff sounds.
FIGS. 2(A), 3(A), 4(A), 5(A) and 6(A) show the output signals of the pickup
sensor 10 of a healthy person when the pressure applied to the arm cuff
varies. FIGS. 7(A), 8(A), 9(A), 10(A) and 11(A) show the pickup sensor
output signals when a person having hardened arteries is being measured.
FIGS. 12(A), 13(A), 14(A), 15(A) and 16(A) show the pickup sensor output
signals when a person having thick subcutaneous fat is being measured.
Furthermore, FIGS. 2(A), 7(A) and 12(A) show the pickup sensor output
signals when the cuff pressure is above the systolic pressure. FIGS. 3(A),
8(A) and 13(A) show the pickup sensor output signals when the cuff
pressure is at the systolic pressure point. FIGS. 4(A), 9(A) and 14(A)
show the pickup sensor output signals when the cuff pressure is between
the systolic pressure and the diastolic pressure. FIGS. 5(A), 10(A) and
15(A) show the pickup sensor output signals when the cuff pressure is at
the diastolic pressure point. FIGS. 6(A), 11(A) and 16(A) show the pickup
sensor output signals when the cuff pressure is lower than the diastolic
pressure.
The thus obtained pickup sensor output signal is applied to the filter 12.
In a preferred form, the filter 12 is a combined construction of a
highpass filter having the cut-off frequency of 20 Hz and having the
cut-off characteristic of 6 dB/oct., and a lowpass filter having the
cut-off frequency of 40 Hz and the cut-off characteristic of 12 dB/oct.
FIG. 2(B) shows an output signal of the filter 12 when the pickup sensor
output signal of FIG. 2(A) is applied to the filter 12. FIG. 3(B) shows an
output signal of the filter 12 when the signal of FIG. 3(A) is applied to
the filter 12. In a same manner, FIG. 16(B) shows an output signal of the
filter 12 when the signal of FIG. 16(A) is applied to the filter 12.
As already discussed above, the filter output signal is applied to the
determination system 16 via the analog-to-digital converter 14. The
determination system 16 determines that the input signal includes the
Korotkoff sound components when the signal of FIG. 3(B), 4(B), 5(B), 8(B),
9(B), 10(B), 13(B), 14(B) or 15(B) is developed from the filter 12, and
then the determination system 16 develops the determination output through
the output terminal 160.
The register Ar included in the determination system 16 stores the digital
value of the last sampled signal level developed from the filter 12. The
register Br stores the digital value of the newly sampled signal level
developed from the filter 12. The register Cr stores the digital value of
the peak level of the sampled signal. Furthermore, the flag register Fr
memorizes the inclination or slope of the signal developed from the filter
12. When the filter output signal is in the increasing process, a data "0"
is memorized in the flag register Fr. When the filter output signal is in
the decreasing process, a data "1" is memorized in the flag register Fr.
The timer T.sub.1 measures 1 msec. which will be used to determine the
sampling time. The timer T.sub.2 functions to measure the time interval
between two peaks which will appear in the output signal derived from the
filter 12.
An operational mode of the determination system 16 will be described with
reference to FIGS. 17, 17(A), 17(B), 17(C), 17(D) and 17(E). The following
operation is controlled by the micro-orders stored in the read only memory
20.
FIG. 17(A) shows an operation for setting an initial value before
conducting the recognition operation of the Korotkoff sounds.
At the step n1, the counting operation of the timer T.sub.2 is stopped, and
the timer T.sub.1 is reset to start a new counting operation. The digital
value developed from the analog-to-digital converter 14 is introduced into
the register Ar at the step n2. The sampling operation is conducted in
accordance with the counting operation of the timer T.sub.1. That is, the
sampling operation is conducted with the time interval of 1 msec. More
specifically, when 1 msec. passed, the program is advanced from the step
n3 to the step n4 to re-start the counting operation of the timer T.sub.1.
Then, the digital value developed from the analog-to-digital converter 14
is introduced into the register Br at the step n5. At the following step
n6 a determination is conducted as to whether the digital value stored in
the register Ar is greater than or equal to the digital value stored in
the register Br. If an affirmative answer is obtained at the step n6, the
program is advanced to the step n7 to set "1" in the flag register Fr. If
an affirmative answer is not obtained, the program is advanced to the step
n8 to set "0" in the flag register Fr. That is, when the level of the
output signal of the filter 12 is in the increasing process, the data "0"
is memorized in the flag register Fr. When the output signal of the filter
12 is in the decreasing process, the data "1" is memorized in the flag
register Fr. At the following step n9, the digital value stored in the
register Br is transferred to the register Ar as the last sampled data.
Furthermore, the digital value stored in the register Br is transferred to
the register Cr as the peak value. When another 1 msec. has passed, the
program is advanced from the step n10 to the following steps.
FIGS. 17(B) and 17(C) (steps n11 through n26) show an operation for
detecting a first negative (trailing edge) peak in the filter output
signal, and for checking whether the introduced signal relates to the
Korotkoff sounds.
At the steps n11 and n12, the timers T.sub.1 and T.sub.2 are reset and
controlled to start the new counting operations. The new digital data is
introduced into the register Br at the step n13. A determination is
conducted at the step n14 as to whether the digital value stored in the
register Ar is greater than or equal to the digital value newly introduced
into the register Br. When an affirmative answer is obtained, that is when
the level of the filter output signal is in the decreasing process, the
program is advanced to the step n15 to check the condition of the flag
register Fr which stores the information as to whether the last sampling
data showed an increasing or decreasing signal. When the last sampling
data showed an increasing signal, the operation is advanced to the step
n16 to transfer the digital value stored in the register Br to the
register Cr as the positive (leading edge) peak value. Then, the data "1"
is set in the flag register Fr, and the counting operation of the timer
T.sub.2 is restarted. When 1 msec. is counted by the timer T.sub.1, the
program is advanced from the step n17 to the step n18 to transfer the
digital value stored in the register Br to the register Ar.
Then, the operation is returned to the step n12 to repeat the
above-mentioned operation.
If a negative answer is obtained at the level of the step n14, that is when
the newly introduced filter output signal is in the increasing process,
the program is advanced from the step n14 to the step n19 at which the
memory state of the flag register Fr is checked. When the data "0" is
stored in the flag register Fr, that is when the last sampling data showed
an increasing signal, the operation is advanced from the step n19 to the
step n17. Contrarily when the data "1" is detected at the step n19, that
is when the last sampling data showed a decreasing signal, the
determination system 16 determines that the negative (trailing edge) peak
is detected. Then, the operation is advanced to the step n20 in FIG. 17(C)
in order to determine as to whether the detected negative (trailing edge)
peak is derived from the Korotkoff sounds.
At the step n20 in FIG. 17(C), the flag register Fr is set to "0", and the
counting operation of the timer T.sub.2 is stopped. A determination is
conducted at the following step n21 as to whether the difference between
the digital value stored in the register Cr (which represents the last
peak value) and the digital value stored in the register Ar (which
represents the peak value of the newly detected negative (trailing edge)
peak) is greater than or equal to a predetermined value ".alpha.". If an
affirmative answer is not obtained, the system determines that the
introduced signal does not relate to the Korotkoff sounds, and the
operation is advanced to the step n22. When 1 msec. has been counted by
the timer T.sub.1, the operation is advanced from the step n22 to the step
n23 at which the digital value stored in the register Ar is transferred to
the register Cr, and the digital value stored in the register Br is
transferred to the register Ar. Then, the operation is returned to the
step n11 in FIG. 17(B).
When an affirmative answer is obtained at the step n21, the operation is
advanced to the step n24 at which the time interval between the two peaks
is checked. When more than 200 msec. have been passed from the last peak
to the now detected peak, the system determines that the introduced signal
does not relate to the Korotkoff sounds and, then, the operation is
advanced to the above-mentioned step n22. When the now detected peak
appeared within 200 msec. after the appearance of the last peak, the
system determines that the introduced signal may relate to the Korotkoff
sounds. The operation is advanced to the following step n25 at which the
digital value stored in the register Ar is transferred to the register Cr
as the peak value, and the digital value stored in the register Br is
transferred to the register Ar as the last sampling value. Furthermore,
the data "1" is set in the flag register Fr. When 1 msec. has passed, the
operation is advanced from the step n26 to the step n27.
FIGS. 17(C) and 17(D) (steps n27 through n39) show an operation for
detecting a positive (leading edge) peak in the filter output signal, and
for checking whether the introduced signal relates to the Korotkoff
sounds.
At the steps n27 and n28, the timers T.sub.1 and T.sub.2 are reset and
controlled to start the new counting operations. The new digital data is
introduced into the register Br at the step n29. A determination is
conducted at the step n30 as to whether the digital value stored in the
register Ar (last sampling data) is greater than or equal to the digital
value introduced into the register Br. When an affirmative answer is not
obtained, that is when the level of the filter output signal is increasing
process, the operation is advanced to the step n31 at which the digital
value stored in the register Br is transferred to the register Ar. When 1
msec. has passed, the operation is returned from the step n32 to the step
n28.
When an affirmative answer is obtained at the step n30, the operation is
advanced to the step n33 in FIG. 17(D). That is, when the level of the
filter output signal is decreasing, the system determines that the
positive (leading edge) peak is detected. Therefore, the timer T.sub.2 is
stopped at the step n33 and, then, the operation is advanced to the
following step n34. A determination is conducted at the sten n34 as to
whether the difference between the now detected peak value stored in the
register Ar and the last detected negative (trailing edge) peak value
stored in the register Cr is greater than or equal to a predetermined
value ".beta.". If an affirmative answer is not obtained at the step n34,
the system determines that the introduced signal does not relate to the
Korotkoff sounds, and the operation is advanced to the step n35. When 1
msec. has passed, the program is advanced from the step n35 to the step
n36 at which the digital value stored in the register Ar is transferred to
the register Cr, and the digital value stored in the register Br is
transferred to the register Ar. Then, the operation is returned to the
step n11 in FIG. 17(B).
Contrarily when an affirmative answer is obtained at the step n34, the
operation is advanced to the step n37 in order to check the time interval
between the appearance of the last detected negative (trailing edge) peak
and the appearance of the now detected positive (leading edge) peak. If
the now detected peak appeared after 100 msec. from the appearance of the
last peak, the system determines that the introduced signal does not
relate to the Korotkoff sounds, and the operation is advanced to the
above-mentioned step n35. When the now detected peak is within 100 msec.
from the last detected peak, the system determines that the introduced
signal may relate to the Korotkoff sounds, and the operation is advanced
to the following step n38. The digital value stored in the register Ar is
transferred to the register Cr as the peak value. The digital value stored
in the register Br is transferred to the register Ar. Furthermore, a data
"0" is set in the flag register Fr, and the program is advanced from the
step n39 to the step n40 in FIG. 17(E) when 1 msec. has passed.
FIG. 17(E) shows an operation for detecting a sound negative (trailing
edge) peak in the filter output signal, for checking whether the
introduced signal relates to the Korotkoff sounds, and for developing the
determination output from the output terminal 160 when the introduced
signal relates to the Korotkoff sounds.
In a similar way as discussed above, the new digital value is introduced
into the register Br at the step n42. A determination is carried out at
the step n43 as to whether the new introduced data is in the increasing
process or the decreasing process. If the level of the filter output
signal is decreasing, the operation is returned from the step n43 to the
step n41 through the steps n44 and n45. When the level of the filter
output signal is increasing, the counting operation of the timer T.sub.2
is stopped at the step n46 and the counted contents are prepared for the
determination at the step n48. Another determination is conducted at the
step n47 as to whether the value difference between the digital values
stored in the register Cr and the register Ar is greater than or equal to
a predetermined value ".gamma.". When an affirmative answer is obtained at
the step n47, still another determination is carried out at the step n48
whether the present peak is detected within 100 msec. from the detection
of the lask peak. When an affirmative answer is not obtained at the step
n48, the system determines that the introduced signal does not relate to
the Korotkoff sounds, and the operation is returned to the step n11 in
FIG. 17(B) through the steps n50 and n51. If an affirmative answer is
obtained at the step n48, the system recognizes that the introduced signal
relates to the Korotkoff sounds, and the operation is advanced through the
step n52 to the step n53 for developing the determination output at the
output terminal 160.
The above-mentioned constants .alpha., .beta. and .gamma. are determined in
the following manner. The constant .alpha. determines the minimum value of
the difference P.sub.1 (see FIG. 18) between the first negative (trailing
edge) peak p.sub.1 and the preceding positive (leading edge) peak p.sub.0.
The minimum value is determined taking into account the difference P.sub.1
in FIG. 15(B) (filter output signal around the diastolic pressure of a
person having a thick subcutaneous fat). The constant .beta. determines
the minimum value of the difference P.sub.2 between the first negative
(trailing edge) peak p.sub.1 and the next positive (leading edge) peak
p.sub.2. The minimum value is selected taking into account the difference
P.sub.2 in FIG. 3(B) (filter output signal at the systolic pressure of a
healthy person or the difference P.sub.2 in FIG. 13(B) (filter output
signal at the systolic pressure of a person having a thick subcutaneous
fat). Alternatively, the minimum value is selected in accordance with the
difference P.sub.2 in FIG. 15(B) (filter output signal around the
diastolic pressure of a person having a thick subcutaneous fat). In a
preferred form, the constant .beta. is selected at a value slightly lower
than the smallest value of the abovementioned three difference P.sub.2.
The constant .gamma. determines the minimum value of the difference
P.sub.3 between the positive (leading edge) peak p.sub.2 and the second
negative (trailing edge) peak p.sub.3. The constant .gamma. is selected at
a value slightly smaller than the lower value as between the difference
p.sub.3 in FIG. 3(B) (filter output signal at the systolic pressure of a
healthy person) and the difference P.sub.3 in FIG. 13(B) (filter output
signal at the systolic pressure of a person having a thick subcutaneous
fat).
When a filter output signal as shown in FIG. 2(B), 7(B) or 12(B) is
introduced into the determination system 16, the determination system 16
recognizes that the introduced signal does not relate to the Korotkoff
sounds because the determination at the step n34 provides a negative
answer. This is because the filter output signal has a small difference
P.sub.2.
When a filter output signal as shown in FIG. 16(B) is introduced into the
determination system 16, the determination system 16 determines that the
introduced signal does not relate to the Korotkoff sounds because the step
n34 in FIG. 17(D) or the step n47 in FIG. 17(E) provides a negative
answer. This is because the introduced signal does not have the difference
corresponding to the above-mentioned difference P.sub.1. Accordingly, as
shown in FIGS. 19(A) and 19(B), the differences P.sub.1, P.sub.2 and
P.sub.3 shown in FIG. 19(A) are considered as the differences P.sub.1,
P.sub.2 and P.sub.3 shown in FIG. 19(B), respectively.
When a filter output signal as shown in FIG. 6(B) or 11(B) is introduced
into the determination circuit 16, the determination circuit 16 determines
that the introduced signal does not relate to the Korotkoff sounds at the
determination of the step n21 shown in FIG. 17(C). This is because, it
will be clear from FIGS. 6(B) and 11(B), the difference P.sub.1 is very
small.
The following is an operation when the filter output signal shown in FIG.
3(B) is introduced into the determination circuit through the
analog-to-digital converter 14.
The difference P.sub.1 in FIG. 3(B) is greater than the constant .alpha..
In the charts of FIGS. 2(A) through 16(A) and 2(B) through 16(B), one
abscissa scale represents 100 msec. In the signal of FIG. 3(B), the time
difference T(I) provided between the positive (leading edge) peak p.sub.0
and the first negative (trailing edge) peak p.sub.1 is less than 200 msec.
Accordingly, the step n21 and step n24 in FIG. 17(C) provide affirmative
answers. That is, the determination operation is advanced to the step n27
through steps n25 and n26. Furthermore, the signal shown in FIG. 3(B) has
the difference P.sub.2 greater than the constant .beta.. Moreover, the
time interval T(II) between the first negative (trailing edge) peak
p.sub.1 and the positive (leading edge) peak p.sub.2 is shorter than 100
msec. Accordingly, the steps n34 and n37 provide affirmative answers to
advance the determination to the step n40 through the steps n38 and n39 of
FIG. 17(D). The filter output signal shown in FIG. 3(B) has the difference
P.sub.3 greater than the constant .gamma.. Furthermore, the time interval
T(III) between the positive (leading edge) peak p.sub.2 and the second
negative (trailing edge) peak p.sub.3 is shorter than 100 msec.
Accordingly, the determination system 16 obtains the affirmative answers
at the steps n47 and n48 shown in FIG. 17(E). Thus, the determination
system 16 recognizes that the filter output signal relates to the
Korotkoff sounds, and develops the determination output at the output
terminal 160.
In a same manner, when the filter output signals shown in FIGS. 4(B), 5(B),
8(B), 9(B), 10(B), 13(B) 14(B) and 15(B) are applied to the determination
system 16 via the analog-to-digital converter 14, the determination system
16 develops the determination output at the output terminal 160.
FIG. 20 shows another embodiment of an electronic sphygmomanometer of the
present invention. Like elements corresponding to those of FIG. 1 are
indicated by like numerals.
A Korotkoff sounds recognition system 30 includes shift registers 32 (Ar1,
Ar2, Ar3 and Ar4) to which the eight bit digital signal is introduced from
the analog-to-digital converter 14. Each of the shift registers 32 (Ar1,
Ar2, Ar3 and Ar4) is of eight bit construction. A clock pulse .0. (1 kHz)
is applied to a terminal 320 to control the shift operation and the data
read operation of the shift registers 32. That is, the digital value
developed from the analog-to-digital converter 14 is introduced into the
shift registers 32 with the time interval of 1 msec. The three items of
eight bit data stored in the shift registers Ar1, Ar2 and Ar3 are
introduced into a comparator 34 for checking the characteristics of the
filter output signal. When Ar2-Ar1.gtoreq.0 and Ar3-Ar2<0, or when
Ar2-Ar1<0 and Ar3-Ar2>0, a pulse signal is developed from a terminal 340,
and the eight bit data stored in the register Ar2 is developed from
another output terminal 342.
The eight bit data item developed from the output terminal 342 of the
comparator 34 is introduced into shift registers 36 (Br1, Br2, Br3 and
Br4) each of which is of eight bit construction. The pulse signal
developed from the terminal 340 of the comparator 34 is applied to an
input terminal 360 of the shift registers 36 in order to control the shift
operation in the shift registers 36 and to introduce the eight bit data
developed from the output terminal 342 of the comparator 34 into the shift
register Br1. A subtractor 38 is connected to the shift registers Br1 and
Br2 to subtract the data stored in the shift register Br2 from the data
stored in the shift register Br1. The calculation result (Br1-Br2) is
developed from a first output terminal 380, and a second output terminal
382 develops a borrow signal for obtaining a parameter required in the
following stage. When (Br1-Br2).gtoreq.0, the barrow signal bears the
logic "H". When (Br1-Br2)<0, the borrow signal bears the logic "L". The
calculation result developed from the subtractor 38 is applied to shift
register 40(Cr1, Cr2 and Cr3) for storing purposes. Each of the shift
register (Cr1, Cr2 and Cr3) is of nine bit construction. When the pulse
signal developed from the terminal 340 of the comparator 34 is applied to
an input terminal 400 of the shift registers 40, the shift registers 40
perform the shift operation in order to introduce the calculation result
(Br1-Br2) developed from the first output terminal 380 of the subtractor
38 into the shift register Cr1-A. The borrow signal developed from the
second output terminal 382 of the subtractor 38 is introduced into the
shift register Cr1-B.
A twelve bit counter 42 is provided for calculating the parameter. The
above-mentioned clock pulse .0. (1 KHz) is applied to a first input
terminal 420 of the counter 42. When the pulse signal developed from the
terminal 340 of the comparator 34 is applied to a second input terminal
422 of the counter 42, the counter 42 develops a control signal at a first
output terminal 424 and develops the counted contents at a second output
terminal 426. Then, the count contents are reset to zero.
The counted contents developed from the second output terminal 426 of the
counter 42 are introduced into shift registers 44 (Dr1, Dr2 and Dr3) each
of which are of twelve bit construction. When the control signal developed
from the first output terminal 424 of the counter 42 is applied to a first
input terminal 440 of the shift registers 44, the shift registers 44
perform the shift operation and memorize the counted contents developed
from the second output terminal 426 of the counter 42 in the shift
register Dr1 via a second input terminal 442.
The shift registers 40 and 44 function as address circuits for a read only
memory 46. The read only memory 46 develops a signal through an output
terminal 460 in response to the contents stored in the shift registers 40
and 44. The signal developed from the output terminal 460 of the read only
memory 46 takes the logic "H" when the address selected by the shift
registers 40 and 44 stores a data "1" which represents that the data
relates to the Korotkoff sounds. When the selected address stores the data
"0", the signal developed from the output terminal 460 bears the logic
"L". The read only memory 46 has the capacity of 2.sup.(36+48) bits. The
read only memory 46 is constructed to store "1" at preselected addresses.
More specifically, since the Korotkoff sounds can be represented by
exemplified combinations of the contents stored in the shift registers 40
and 44, the data "1" is set at addresses which correspond to the
exemplified combinations.
In a preferred form, the data "1" is set at the addresses corresponding to
the following combinations of the contents of the shift registers 40 and
44.
TABLE I
______________________________________
SHIFT REGISTER CONTENTS (RANGE)
______________________________________
Cr1 100.sub.H -1BF.sub.H
Cr2 040.sub.H -0FF.sub.H
Cr3 100.sub.H -1EF.sub.H
Dr1 001.sub.H -08O.sub.H
Dr2 001.sub.H -08O.sub.H
Dr3 001.sub.H -0CO.sub.H
______________________________________
(represented by Hexadecimal Notation)
That is, the data "1" is stored at the addresses which correspond to the
positions where the contents of the shift register Cr1 are between
100.sub.H and 1BF.sub.H, and the contents of the shift register Cr2 are
between 040.sub.H and 0FF.sub.H, and the contents of the shift register
Cr3 are between 100.sub.H and 1FF.sub.H. These three conditions must be
satisfied. Furthermore, the contents stored in the shift registers 44 must
satisfy the following three conditions. The contents stored in the shift
register Dr1 are between 001.sub.H and 080.sub.H, and the contents stored
in the shift register Dr2 are between 001.sub.H and 080.sub.H, and the
contents stored in the shift register Dr3 are between 001.sub.H and
0C0.sub.H. Addresses other than the positions which satisfy the
above-mentioned six conditions store the data "0".
An operation of the system shown in FIG. 20 will be described with
reference to FIG. 21. In FIG. 21 one abscissa scale indicates 2 msec. and
one ordinate scale indicates 10.sub.H. FIG. 21 shows a filter output
signal developed from the filter 12, which is applied to the Korotkoff
sounds recognition system 30 via the analog-to-digital converter 14.
When the digital value representing the level of the point A.sub.5 is
introduced into the shift register Ar1 via the analog-to-digital converter
14, the shift register Ar2 stores the digital value representing the level
of the point A.sub.4 and the shift register Ar3 stores the digital value
representing the level of the point A.sub.3. Because (Ar2-Ar1).gtoreq.0
and (Ar3-Ar2)<0, the comparator 34 develops the pulse signal from the
terminal 340. Accordingly, the shift register Br1 stores the digital value
representing the level of the point A.sub.4, which has been stored in the
shift register Ar2. At this moment, the shift register Br2 stores the
digital value representing the level of the point A.sub.1. Thus, in
response to the output signal from the subtractor 38, the shift register
Cr1 stores the digital value representing the difference H.sub.1 between
the positive (leading edge) peak A.sub.4 and the negative (trailing edge)
peak A.sub.1. Furthermore, the shift register Dr1 stores the time
difference T.sub.1 ' (=T.sub.1) between the two peaks A.sub.1 and A.sub.4.
Similarly, when the digital value representing the level of the point
A.sub.14 is introduced into the shift register Ar1, the shift registers 40
and 44 store the following values.
shift register Cr1--the difference H.sub.4
shift register Cr2--the difference H.sub.3
shift register Cr3--the difference H.sub.2
shift register Dr1--time difference T.sub.4 '(=T.sub.4)
shift register Dr2--time difference T.sub.3 '(=T.sub.3)
shift register Dr3--time difference T.sub.2 '(=T.sub.2)
When the filter output signal shown in FIG. 2(B) is introduced into the
Korotkoff sounds recognition system 30 through the analog-to-digital
converter 14, the contents stored in the shift register Cr2 will not be
higher than 040.sub.H because the filter output signal does not include
the peak which has the level higher than 0.8 scale. In FIG. 2(B) one
ordinate scale represents the level of 50.sub.H. Accordingly, the signal
developed from the output terminal 460 of the read only memory 46 bears
the logic "L".
When the filter output signal shown in FIG. 3(B) is introduced into the
Korotkoff sounds recognition system 30 through the analog-to-digital
converter 14, at a time when the digital value representing the level of
the point d is stored in the shift register Ar2, the shift register Cr1
stores the difference between the peaks d and c, namely about 1AF.sub.H.
The shift register Cr2 stores the difference between the peaks c and b,
namely about 040.sub.H. The shift register Cr3 stores the difference
between the peaks b and a, namely about 1EF.sub.H. Furthermore, the shift
register Dr1 stores the time difference between the peaks d and c, namely
about 50.sub.(10). The shift register Dr2 stores the time difference
between the peaks c and b, namely about 60.sub.(10). The shift register
Dr3 stores the time difference between the peaks b and a, namely about
40.sub.(10). These values stored in the shift registers 40 and 44 are
included within the range shown in the TABLE I and, therefore, the signal
developed from the output terminal 460 of the read only memory 46 bears
the logic "H".
When the filter output signal shown in FIG. 4(B) is introduced into the
Korotkoff sounds recognition system 30 through the analog-to-digital
converter 14, the shift registers 40 and 44 store the following values at
a time when the digital value representing the level of the point d is
introduced into the shift register Ar2.
shift register Cr1--about 100.sub.H
shift register Cr2--about 0A0.sub.H
shift register Cr3--about 1AF.sub.H
shift register Dr1--about 30.sub.(10)
shift register Dr2--about 30.sub.(10)
shift register Dr3--about 70.sub.(10)
Since these values are included within the range shown in the TABLE I, the
signal developed from the output terminal 460 of the read only memory 46
bears the logic "H".
When the filter output signal shown in FIG. 5(B) is introduced into the
Korotkoff sounds recognition system 30 through the analog-to-digital
converter 14, the shift registers 40 and 44 store the following values at
a time when the digital value of the point d is introduced into the shift
register Ar2.
shift register Cr1--about 13F.sub.H
shift register Cr2--about 060.sub.H
shift register Cr3--about 1E8.sub.H
shift register Dr1--about 20.sub.(10)
shift register Dr2--about 15.sub.(10)
shift register Dr3--about 20.sub.(10)
Since these values are included within the range shown in the Table I, the
Korotkoff sounds recognition system 30 recognizes that the introduced
signal relates to the Korotkoff sounds and the signal developed from the
output terminal 460 of the read only memory 46 bears the logic "H".
When the filter output signal shown in FIG. 6(B) is introduced into the
Korotkoff sounds recognition system 30 through the analog-to-digital
converter 14, the shift registers 40 and 44 store the following values at
a time when the digital value of the point d is stored in the shift
register Ar2.
shift register Cr1--about 18F.sub.H
shift register Cr2--about 040.sub.H
shift register Cr3--about 1F8.sub.H
In this case, the contents stored in the shift register Cr3 are not
included within the range 100.sub.H through 1EF.sub.H shown in the TABLE
I. Accordingly, the signal developed from the output terminal 460 of the
read only memory 46 bears the logic "L".
In a same manner, when the filter output signals shown in FIGS. 8(B), 9(B),
10(B), 13(B), 14(B) and 15(B) are introduced into the Korotkoff sounds
recognition system 30, the signal developed from the output terminal 460
of the read only memory 46 bears the logic "H". When the filter output
signals shown in FIGS. 7(B), 11(B), 12(B) and 16(B) are introduced into
the Korotkoff sounds recognition system 30, the signal developed from the
output terminal 460 of the read only memory 46 bears the logic "L".
In the embodiment of FIG. 1, the determination system 16 is implemented
with a one chip microcomputer which includes the central processor unit
18, the read only memory 20, the random access memory 22 and the timer 24.
FIG. 22 schematically shows still another embodiment of an electronic
sphygmomanometer of the present invention, wherein an analog-to-digital
converter 50 and a digital filter 52 are included in the determination
system 16 made of a single chip LSI.
FIG. 23 schematically shows yet another embodiment of an electronic
sphygmomanometer of the present invention. Like elements corresponding to
those of FIG. 1 are indicated by like numerals. In this embodiment, an
amplifier 54 is disposed between the pickup sensor 10 and the filter 12 to
amplify the output signal developed from the pickup sensor 10.
FIG. 24 schematically shows a further embodiment o | | |
