JPS63262532A - Predicative operation type electronic clinical thermometer - Google Patents

Abstract

PURPOSE:To enhance prediction accuracy, by limiting the displacement quantity to the previous value so as to allow the same to be in a predetermined range prior to substituting a parameter value obtained by calculation for a function formula. CONSTITUTION:B-calculation circuits 50, 58, C-calculation circuits 54, 60 and A-calculation circuits 52, 56 as parameter calculation circuits respectively calculates respective parameter values according to predetermined formulae at each time when they input the parameter calculation signals 27-29 from a calculation state indicating circuit 11. There parameter values are limited by B-limit circuits 51, 59 and C-limit circuits 55, 61 as parameter limit circuits. A B.A output circuit 53, a C.A output circuit 57 and a B.C output circuit 62 as parameter output circuits output the parameter values from of the parameter calculation circuits or parameter light circuits.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a temperature detection unit that detects the temperature of a measurement site, a calculation unit that predicts and calculates an equilibrium temperature based on the detected temperature, and a display unit that displays the temperature. The present invention relates to a predictive calculation type electronic thermometer including the following.

[Conventional technology]

Conventional predictive calculation type electronic thermometers are disclosed in Japanese Patent Application Laid-open No. 58-2, for example.
As in Publication No. 25324, (1) Detect the current temperature and elapsed time,
(2) Calculate the predicted value by substituting the temperature data and time data into a functional formula containing arbitrary parameters, (3) Repeating the above steps (1) and (2) at regular intervals, and (4) ) The obtained predicted value is constantly monitored, and if the temporal change amount of the predicted value is within a predetermined range, the result is displayed and the process ends; if it exceeds the range, (5) other of the above parameters are The values are selectively read from the storage area, the functional formula is determined again, and the process returns to step (1).

Calculation, monitoring, and modification of predicted values (1) ~
By making the process (5) sequential by performing a negative feedback operation of changing the selection of parameters, the accuracy of fitting the predicted value was improved.

However, since the parameters that determine the functional expression of the calculation are selectively changed, a certain number of changes are required before the optimum parameter values are selected. In addition, the predicted value must be monitored for a certain period of time in order to decide whether to change the parameter, and it takes a certain amount of time to change the parameter once.

Furthermore, in this method, there is only one parameter,
It does not fully account for variations in body temperature rise curves due to differences in people and measurement conditions.

In order to solve the above-mentioned problems and realize shortening of prediction time and improvement of prediction accuracy, the present applicant has proposed a predictive calculation type electronic thermometer in the earlier application, Japanese Patent Application No. 61-263572. There is. The configuration and operation of the predictive calculation type electronic thermometer of the prior application will be explained below with reference to FIGS. 4 to 7.

FIG. 5 is a circuit block diagram showing the basic configuration of the predictive calculation type electronic thermometer of the prior application. The temperature detection section 1 is connected to the calculation section 2 at a portion that detects the temperature of the measurement site and outputs it intermittently as an electrical signal. The calculation unit 2 includes a parameter calculation unit 4, calculates a predicted value of the equilibrium temperature from the temperature data input from the temperature detection unit 1, and displays the predicted value as an electric signal on the display unit 6.
Output to. The parameter calculation unit 4 is a part that calculates parameters of a functional formula that provides a predicted value. The display section 6 is a section that converts the predicted value inputted as an electrical signal from the calculation section 2 into a corresponding numerical value and displays it.

FIG. 4 is a circuit block diagram showing a specific configuration of the predictive calculation type electronic thermometer of the prior application. The temperature detection unit 1 is composed of a sensor 5 such as a thermistor, for example, and a temperature detection circuit 6. The temperature detection circuit 6 detects a temperature signal output from the sensor 5 in response to the ambient temperature, and detects the corresponding temperature. It is intermittently outputted to the arithmetic unit 2 as data signals 21 and 22.

The calculation section 2 includes a measurement control circuit 7, a time measurement circuit 8, a temperature memory circuit 9, a parameter calculation section 4, and a predicted value calculation circuit 10.
It is made up of. The measurement control circuit 7 inputs the temperature data signal 21 from the temperature detection circuit 6 and constantly monitors the temperature of the measurement site in real time. , and the prediction calculation start signal 24 is output to the time measurement circuit 8 by satisfying the conditions 10 seconds after detecting that the temporal change in temperature is 01°C/SCC or more, and at the same time, the temperature In synchronization with the output timing of the intermittent temperature data signal 21 from the detection circuit 6, the calculation instruction signal 26 is output to the parameter calculation unit 4 until an instruction to end the calculation is given.

The time measurement circuit 8 starts counting time at the same time that the electronic thermometer is powered on, the circuit is activated, and the temperature detection circuit 6 outputs a temperature data signal. When input, the current time data is reset and new time counting starts. This time data is output to the parameter calculation section 4 as an elapsed time data signal 25.

The temperature memory circuit 9 is a circuit that temporarily stores the input temperature data according to a certain rule when the temperature data signal 22 from the temperature detection circuit 6 is input. It outputs to the parameter calculation section 4 according to the following. Upon receiving the signal from the measurement control circuit 7, the parameter calculation section 4 calculates a plurality of parameters of a functional formula that gives a predicted value of the equilibrium temperature using the signals from the time measurement circuit 8 and the temperature memory circuit 9, which will be described in detail later. The parameters are calculated by a predetermined operation, and a BA signal 64, a CA signal 65, and a BC signal 66 (hereinafter collectively referred to as parameter signals) are output to the predicted value calculation circuit 10.

The predicted value calculation circuit 10 receives each of the parameter signals 64, 65, and 66 from the parameter calculation section 4 and outputs the results to the display section 3 after performing a predetermined operation.

Here, the process of calculating the predicted value will be explained using the flowchart in FIG. 6 and the characteristic diagram in FIG. 7, focusing on the specific configuration and operation of the parameter calculation unit 4.

The parameter calculating section 4 in FIG.
-C calculation circuit 14. These B-A calculation circuit 12, C@A calculation circuit 16, B-C calculation circuit 14
When the B-A calculation signal 27, the C-A calculation signal 28, and the B-C calculation signal 29 (hereinafter collectively referred to as parameter calculation signals) are input from the calculation state instruction circuit 11, the predicted values are calculated. Three parameters A, B, and C of the given function formula
This is a circuit that calculates . For example, the B-A calculation circuit 12
Parameters B and A will be calculated.

Now, the calculation state instruction circuit 11 receives the elapsed time data signal 25 and the temperature data signal 21 from the time measurement circuit 8 and the temperature memory circuit 9 every time the calculation instruction signal 26 is input from the measurement control circuit 7 .

Each time, it is output as a parameter calculation signal 27, 28, 29 containing elapsed time data and temperature data to the three parameter calculation circuits 12, 13, and 14 according to a predetermined rule described later. This parameter calculation signal 27
．． 28.29 is, for example, B for the B-A calculation circuit 12.
-A calculation signal 27. Each time the parameter calculation circuits 12, 16, and 14 receive parameter calculation signals 27, 28, and 29 from the calculation state indication circuit 11, they calculate specified parameters from the elapsed time data and temperature data, and send the calculated parameters to the predicted value calculation circuit 10. Parameter signal 6
Output as 4.65.66.

The predicted value calculation circuit 10 calculates three parameters A of the functional formula,
Of B and C, only the parameters corresponding to the input parameter signals are rewritten to the latest data, and a predicted value is calculated from the determined functional formula and output to the display unit 3.

Before going into the detailed explanation of the operation, it is necessary to explain the functional formula that gives the predicted value and the method for calculating the parameters of the functional formula. A visual graph of the changes in temperature observed over time up to the point in time shows a wide variety of changes depending on the person being measured, the measurement site, and the physical characteristics of the sensor (mainly heat capacity). shows.

This body temperature rise curve, which appears to have no regularity, is collected, faithfully reproduced, and statistically analyzed, and the representative body temperature obtained by taking the average value of all the data. The rising curve can be approximated by the following functional formula (U fs).

Further, this functional expression U(S) corresponds to what was referred to above as a functional expression that gives one predicted value.

U(S)=A−C/(S+B) ・・・・・・
■S is time and the unit is [seconds], and A%B and C are constants that determine this functional expression U (s), and are parameters.
This is what is called. By repeating the calculation of this parameter in accordance with the prescribed rules described below, it is finally determined as the parameter that gives the most suitable functional formula for formulating the body temperature rise curve, and as a result, an accurate predicted value of the equilibrium temperature is obtained. will be derived.

By the way, the functional formula ■ has a form similar to 1/χ, but this functional formula is not uniquely determined as the formula for deriving the predicted value (for example, it is not shown in the literature representing the conventional example). Even if the equation is a multidimensional function that includes an exponential function, an exponential function, or even a quadratic or higher order term, the equilibrium temperature can be predicted with a certain degree of accuracy by selecting the constants that make up the equation, but it is not always possible to predict the equilibrium temperature with a certain degree of accuracy. It cannot correspond to all body temperature rise curves, and the functional formula itself requires complex calculations.

On the other hand, the function formula ■ allows you to freely change the curve drawn by changing and setting three types of parameters. Functional expressions can be associated with each curve. By the way, when parameter A is changed in the functional formula U (s) of It moves in equilibrium. Also, when parameter C is changed, the curve expands and contracts vertically and horizontally.

In order to realize what has been described above, the parameters of the function expression are calculated as follows. Calculation status indication circuit 11
Each time the measurement control circuit 7 receives a signal, it inputs the elapsed time data S and temperature data To, T2, T4 from the time measurement circuit 8 and temperature memory circuit 9, respectively. Here, TO1T2 and T4 are temperature data temporarily stored in the temperature memory circuit 9, and are detected at the current elapsed time S, the time S-2 of the temperature detection before the previous time, and the time S-4 of the temperature detection the time before the previous time, respectively. represents the temperature data corresponding to the temperature. Assuming that the sampling time for temperature detection is 2 seconds, the above elapsed times are S-2 and S-4 with respect to the current value S. First, calculation state calculation circuit 1
1, when the calculation instruction signal 23 is input from the measurement control circuit 7 for the first time after the power is turned on, parameters A, B, and C are reset by the following equation and calculated as initial values.

A=T2+AI ・・・・・・■B
=A1/+2 ・・・・・・■C=A
12/K... ■A1 is a numerical value statistically determined using the above-mentioned typical body temperature rise curve, and is a predetermined value. Further, K is calculated from the current temperature data TO and the temperature data T4 detected two times before (four seconds ago) using the following equation (2), and is called a "temperature gradient".

K=(T4-To)/4...
(2) The above equations (2), (2), and 0 are given by transforming the function equation (2) using three data: elapsed time, temperature, and A1. Parameters A, B, and C calculated using the above formulas ■, ■, and 0 are values that give a functional formula corresponding to a typical body temperature rise curve, and are is not very effective.Therefore, the values of each of these parameters are treated as gradual values.In other words, in this state, the calculation state indication circuit 11, the B-A calculation circuit 12,
No parameter calculation signal is output to each parameter calculation circuit of the C-A calculation circuit 16 and the B-C calculation circuit 14, and therefore no predicted value is calculated. After 2 seconds have elapsed, when the calculation instruction signal 23 is output again from the measurement control circuit 7, the calculation state instruction circuit 11 receives the calculation instruction signal 26, and the time measurement circuit 8
The elapsed time data S and temperature data To, T2, and T4 at that time are input from the temperature memory circuit 9 and the temperature memory circuit 9, respectively.

Each value of S, TO, T2, and T4 is different from the previous value,
Needless to say, each value is the latest at present. The calculation state indicating circuit 11 sends out the above-mentioned elapsed time data S and temperature data To, T2, T4 as a B-A calculation signal 27, and upon receiving this, the B-A calculation circuit 12 first calculates the parameter B. Calculated from the following formula.

B-f “-8+2 at 7 ・・・・・・■
C is the value found last time, and is the initial value here. From this result, parameter A is calculated using the following formula. .

A=TO+C/(S+B)...
- The values of parameter B and parameter A calculated by the equations 2 and 0 are output as a B-A signal 64.

The predicted value calculation circuit 10 rewrites the parameters corresponding to the input signal to the latest values, determines the function formula (2) that provides the predicted value, and calculates the predicted value U using the following formula.

U=A-C/ (300+B)...
・■This is the value obtained when the values of parameters A, B, and C are substituted in the function formula (■), and the elapsed time S is set to 300 seconds. The reason why we set the value of elapsed time to 300 seconds or 5 minutes is because the equilibrium temperature during the day is generally considered to be approximately 5 minutes after the start of temperature measurement, and in the axilla (armpit) or rectum. Each has a corresponding value. By the way, it is about 10 minutes in the armpit, and the elapsed time S is 600 seconds. ■
The predicted value U calculated by the formula is outputted to the display unit B as a predicted value signal 63, and the corresponding numerical value is displayed as a result. Next, when receiving the signal input from the measurement control circuit 7 again, the calculation state indication circuit 11 in turn receives the signal from the C-A calculation circuit 1.
6, a CA calculation signal 28 is output. The C-A calculating circuit 16 calculates the corresponding parameters 〇 and A using the following formula.

C=Kx (S-2+B)2...
■Based on this result, the value for the parameter is given by A2TO+C/(S+B)...■. However, ■, the value of parameter B in equation 0 is the value determined during the previous calculation of parameter B-A. Parameters C and A determined by ■ and formula 0
The value of is output as the C@A signal 35, and in the predicted value calculation circuit 10, the values of C and A are rewritten as the corresponding parameters and the function formula ■ is determined, and then the predicted value U is calculated by the formula ■, The predicted value signal 36 is outputted to the display unit 6. At this time, the displayed predicted value is rewritten to the latest value. After another 2 seconds have elapsed, when the calculation instruction signal 26 is output again from the measurement control circuit 7, the calculation state instruction circuit 11 repeats the same operation as the previous time, that is, 4 seconds ago. First, when the calculation instruction signal 26 is input to the calculation state instruction circuit 11, it outputs the BA calculation signal 27 including elapsed time data and temperature data to the BA calculation circuit 12.

After calculating the parameters B and A from equations 1 and 0, they are outputted to the predicted value calculation circuit 10 as the corresponding BA signal 64. In the predicted value calculation circuit 10, the value of the corresponding parameter is rewritten to determine the function formula (■), and the predicted value U is determined by the formula (■).
is calculated and output to the display unit 6 as a predicted value signal 36.

In this way, the B-A calculation circuit 12, the C-A calculation circuit 16, and the KB-A calculation circuit 1 are operated until a predetermined condition is satisfied.
Parameters B and A and parameters 〇 and A are repeatedly calculated by operating the parameter 〇 and 〇 and A alternately.
The B-A calculation circuit 12 and the C@A calculation circuit 16 are operated (this may be determined as the number of repetitions. After this condition is met, when a signal from the measurement control circuit 7 is input, the calculation state is changed. The instruction circuit 11 outputs the B-C calculation signal 29 including elapsed time data and temperature data to the B-C calculation circuit 14. When the B-C calculation circuit 14 receives the B-C calculation signal 29, the following The values of parameters B and C are calculated from the equations and output to the predicted value calculation circuit 10.

B=(A-T2)/Ni-(S-2)...[Phase]C2(A-T2)2/K...OHere, parameter A is the previous parameter B This is the value calculated at the time of @A or parameter-C-A calculation. The predicted value calculation circuit 10 calculates B- corresponding to the values of parameters B and C.
After inputting the C signal 66 and determining the function formula ■ in the same way as before,
(2) Calculate the predicted value U using the formula and output it to the display section B. At this time, the final predicted value is displayed by simultaneously sounding a buzzer or the like.

I will briefly explain the above. First, under the predetermined conditions (30℃
(After 10 seconds have passed after detecting 0.1°C/see or higher) and the calculation starts, parameters A and B
, and C are reset to their initial values. Parameters B and A are then calculated. Next, parameters C and A are calculated, and this operation is repeated thereafter, but each time the predicted values are rewritten and displayed continuously. After a certain period of time has passed, the values of parameters B and C are calculated,
A buzzer sounds and the final predicted value is displayed.

A series of operations from the start of temperature measurement to the final display of the predicted value according to this method of calculating the predicted value is shown in the characteristic diagram shown in FIG. In FIG. 7, the numerical values 0.2.4... representing the elapsed time S shown on the horizontal axis represent the times detected by the temperature detection unit 1 and output as temperature data, and at the same time, the calculation state The timing at which the instruction circuit 11 outputs signals to each parameter calculation circuit 12, 16, and 14 is also synchronized. This elapsed time S is not the time from the start of temperature measurement, but is calculated by the calculation unit 2.

The time when the calculation starts is 0 seconds.

T(.)-T(?)-... is the elapsed time 3, but o,
The actual temperature at 2... is continuously graphed as T (1). T(,...) is the equilibrium temperature, which is the temperature after 300 seconds as a daytime temperature measurement.

U (.], U (2)..., the elapsed time S is 0, 2
. . . The predicted value calculated at the time of . Also, the dotted line represents the functional graph that gives the predicted values of U(.) to U(,.).
This is the curve of Alc""' TjSe, and from the same concept as the equilibrium temperature, the value of the function formula ■ when the elapsed time S is 300 seconds is calculated from the formula 0 as the predicted value U(.)~ It can be seen that TJ (,.).

First, when the power is turned on and temperature measurement begins, the temperature is T (,
) as it rises while drawing a curve. Then, when the elapsed time S is O, the predetermined conditions (30°C or higher, 0.1'C/&
ec or more is detected, and the calculation instruction signal 26 is output from the measurement control circuit 7 when the condition (after 10 seconds has elapsed) is satisfied.
Parameters A, B, and C are set above the initial value KIJ.

At this time, the predicted value is not calculated, but for the sake of clarity, it is calculated here and is shown as U (.). After 2 seconds, the parameters B and A are calculated by inputting the signal again, and the functional formula that draws the curve of Ul is determined.
As a result, the predicted value U(1) is calculated. Another 4 seconds later.

Parameters C and A are calculated in the same way, and a functional formula that draws a curve of UaA is determined, thereby obtaining U (1) as a predicted value. This process continues until a certain amount of time has elapsed. After 10 seconds, if this condition is satisfied, the values of parameters B and C are calculated,
1JIl, is determined, and U(so) is determined as the final predicted value.

FIG. 6 is a flowchart showing the specific operation of the device shown in the circuit block diagram of FIG. 4. Step 10
0 to 108 mainly represent operations of temperature detection and error detection. Error detection is the part that detects whether the temperature is being measured correctly.It constantly monitors the rise in temperature, and if it is determined to be incorrect, it notifies you and stops execution. This should prevent inaccurate predicted values from being derived when the temperature changes significantly in a short period of time, such as when the sensor is removed from the measurement site or when the temperature is measured using an incorrect measurement method. has been done.

For example, if the temperature gradient K calculated from the above formula (■) is within the range of 0 (K≦0.08[:°C/sec]), it is assumed that the temperature has been measured correctly, and if it is outside the above range, an error is detected. I judge that.

The branches below the judgment step 109 are classified as modes M, as indicated by numerical values from 1 to 5, depending on the state in which the predicted value is calculated. Proceed to the branch state indicated by the corresponding numerical value.

The mode 1 branch is a part that conditionally monitors whether or not parameters of a function expression should be computed, and provides initial values for each parameter after the start of computation. Modes 2 to 40 are the parameter calculation part.

The mode 5 branch is a part that displays a message to that effect and stops the operation due to the above-mentioned error or the like. Steps 126-127
is the part that calculates predicted values from the calculated parameters and performs display and buzzer processing.

First, in step 100, the power is turned on and all the circuits shown in FIG. 2 are brought into operation. At this time, mode M is set to 1 and other data are also initialized. In step 101, Fl, F2. Initialize each of the 7 lags of F3, that is, OK, elapsed time measurement step 102,
Subsequently, the temperature detection step 106 is entered. Step 104 is a part for temporarily storing temperature data, and TO stores current temperature data, that is, the value detected in the immediately previous temperature detection step 103.

Thereafter, the latest temperature data is always stored for each KTOK that passes through this step 104, and the previous value is changed from TO→T2. T2→
It is shifted and stored as T4.

Step 105 calculates the rate of change in temperature with respect to elapsed time, that is, the above-mentioned temperature gradient I'l. According to the judgment step 106, since the mode M is 1 here, the judgment step 10
9 to proceed to step 110.

In the judgment step 110, it is assumed that the predetermined condition is not satisfied, and the process proceeds to the display processing step 127, but since the display flag F1 is 0, the process returns to the step 101 again without displaying anything.

Thereafter, this process is repeated until it is detected in decision step 110 that a predetermined condition is met. here,
The predetermined conditions may be those described above, but for the purpose of improving the accuracy of predicted values and shortening the calculation time, it is better to use the following conditions.

Namely.

(1) The elapsed time S is 4 seconds or more.

(11) Temperature data TO is 34°C or higher.

011) The temperature gradient K is between 0.02 and 0.06°C.

Condition (1) is that temperature data TO1T is set in step 104.
2. The time required for data to be stored in all of T4 was set based on the fact that 4 seconds is required after the power is turned on.

When all conditions (1) to 011) are satisfied and the measurement control circuit 7 outputs a signal to the calculation state indication circuit 11, step 1 is completed.
Step 11 reads the value of the constant A1, and in step 112 the initial values of parameters A%B, C of the function formula ■ are set to ■, ■, 0.
Calculated using the formula. In step 116, mode M is set to 2, and preparations are made to output a signal to the B and A calculation circuit 12.

Further, the measurement control circuit 7 outputs a signal to the time measurement circuit 8, and in step 114, the elapsed time S is reset to S20 and counting starts anew. Step 101 again. 109
is repeated and a signal is successively output from the measurement control circuit 7, the calculation state indication circuit 11 sends a signal to the B-A calculation circuit 12, and in the calculation step 115 and the calculation step 116, the parameter is calculated from equations 1 and 0. Calculate the values of B and A. In step 117, mode M is set to 3 to prepare for calculation of parameters C and A, and then step 126 is executed in predicted value calculation circuit 10, and predicted value U is calculated from equation (2). The predicted value U is set by setting the display flag F1 to 1 in step 124.
By doing so, it is displayed on the display section B in the display processing step 127. The judgment step 125 is a part for detecting whether the conditions for terminating the parameter calculation are satisfied, and here, the conditions are 1. The elapsed time S is 10 seconds or more.

When this condition is satisfied, a buzzer is set to sound in judgment step 126, and the final predicted value is displayed and notified by the buzzer.

Also, as will be described later, by setting 81 to 8 seconds in the determination step 120, (v) the elapsed time S is 8 seconds or more.

The condition is set. This is done by repeating the calculation of parameter B%A and parameters C and A twice alternately, and finally by the above-mentioned conditions of judgment step 125.
, C is calculated once to complete the process of calculating each parameter A, B, and C.

Now, according to calculation steps 115 and 116, nokula meter B,
After A is calculated and the series of processes up to step 124 is completed, it is determined in a judgment step 125 that the above-mentioned conditions are not satisfied, so a display is performed on the display section B in a display processing step 127, and the process returns to step 101 again.

Thereafter, the process proceeds to the judgment step 109 using the same operation, and the calculation state indicating circuit 11 receives the signal from the measurement control circuit 7.
- Step 118 by outputting a signal to the A calculation circuit 16;
Subsequently, step 119 is executed. The calculations of parameters C and A performed here are given by formulas (1) and (2). Since the current elapsed time S is 4 seconds, the judgment step 120 judges that the condition (■) is not satisfied, and proceeds to the next step 122, setting the mode M to 2 and setting the parameter B again.
.

The process then proceeds to calculation steps 115 and 116.

After that, the parameter calculation process up to now is performed, and when the elapsed time S reaches 8 seconds, the values of parameters C and A are calculated in steps 118 and 119, and the judgment step 120 is performed because the condition (■) is satisfied. , the process proceeds to step 121, where mode M is set to 4.

In the elapsed time measurement step 102, the elapsed time S becomes 10 seconds, and the process proceeds from the judgment step 109 to step 128. The values of parameters B and C are calculated from the [phase] and 0 formulas, and
Proceed to 29.123.124.

In judgment step 125, since condition GV) is satisfied, the buzzer flag F2 is set to 1 in judgment step 126, and at the same time as the predicted value is displayed in display processing 127, the buzzer sounds and the entire process ends.

By the way, in the judgment step 125 and the judgment step 120, the conditions 1ψ and (v) are defined as described above, but the parameter calculation method is changed by setting them as follows, for example. First, in the determination step 125, &D elapsed time S is 14 seconds or more.

In the determination step 120, ~11) The elapsed time S is 12 seconds or more.

By setting such a condition Mp, (■11), the calculation of parameters B and A performed in mode 2 and the calculation of parameter C%A performed in mode 6 are repeated three times alternately, and finally Parameters B and C performed in mode 4
calculation is performed. Further, the conditions for the judgment step 125 are as follows: (Viii) The elapsed time S is 18 seconds or more.

and the judgment step 1200 conditions are (IX) The elapsed time S is 4 seconds or more.

By doing so, calculations performed in mode 2, mode 6, and mode 4 are performed sequentially, and these are repeated three times as one cycle. Mode M in step 129
is set to 2 assuming that this state is realized.

In the two examples shown above, the accuracy of matching the predicted value to the equilibrium temperature is improved, but on the other hand, it takes some time until all parameter calculations are completed and the final predicted value is displayed.

[Problem that the invention seeks to solve]

In FIG. 4 showing the configuration of the predictive calculation type electronic thermometer of the earlier application, each time the temperature detection unit 1 detects temperature data, the parameter calculation unit 4 converts the elapsed time data and the temperature data to each of the above-mentioned items. After determining the value of the parameter by substituting it into the parameter calculation formula, the predicted value was obtained from the functional formula. By adopting the configuration shown in FIG. 4, since there are a plurality of parameters and each parameter is calculated in a predetermined order, it is possible to simplify the functional expression for giving a predicted value and improve the prediction accuracy. Also, each time temperature data is detected,
Since the parameters are calculated, predicted values can be derived in a short time.

However, the value of the parameter that directly affects the derived predicted value is determined by directly substituting the detected temperature data into the parameter calculation formula, albeit in the form of a temperature gradient.
If the temperature data fluctuates abnormally, inaccurate predicted values may be derived.

The following may be considered as abnormal fluctuations in the temperature data. First of all, if the temperature data has a large abnormal fluctuation range, such as when the sensor is removed from the measurement site or when the temperature is measured using an incorrect measurement method, then the temperature signal detected by the sensor is Thirdly, there are cases where the fluctuation range of temperature data is extremely small, such as pit errors when converting temperature data into digital data, or slight fluctuations in body temperature due to changes in the physical condition of the person being measured. Fluctuations in temperature data at intermediate levels include electrical noise due to static electricity and disturbances, movement of the sensor within the measurement area, and momentary disconnection.

First, if there is a large abnormal change in temperature data due to the removal of the sensor, etc., it is detected as an error, as shown in the explanation of the predictive calculation type electronic thermometer of the earlier application, and the process is executed before starting the process of calculating the predicted value. stops, so an inaccurate predicted value will not be derived.

Next, if the fluctuation range of temperature data due to minute fluctuations in body temperature is extremely small (as can be seen from formula ■ is given as K=(T4-TO)/4...■, and if we assume that temperature data T4 from 4 seconds ago is
If there is a fluctuation in T, and it is detected as T4 = T4 + ΔT, then the fluctuation range Δ in the temperature gradient is as follows.

Δ=+ΔT/4, and the influence of the fluctuation range ΔT of the temperature data on the temperature gradient K is reduced to one-fourth. Since the variation width ΔT is extremely small, it is considered that it has almost no effect on the temperature gradient K, so it does not affect the values of each parameter calculated from the temperature gradient K. (As a result, an inaccurate predicted value will not be derived.

Finally, intermediate level fluctuations in temperature data due to electrical noise etc. are not detected as errors, and have a considerable influence on the temperature gradient KK, so the predicted value derived is as follows: This may result in inaccurate values. When explaining the predictive calculation type electronic thermometer of the earlier application, the detection standard for an error is set in the range of 0 (K! 0.08 C°C/sec), and if the temperature gradient K is outside the above range, it is an error. However, if we were to narrow this range and increase the proportion of abnormal fluctuations in temperature data that are judged as errors, the number of incorrect predicted values would certainly be reduced, but on the other hand, Processing occurs frequently, and body temperature measurements are difficult to display, making it difficult for users to use.

On the other hand, if the range is widened, the rate of errors being treated as errors decreases, but the prediction accuracy decreases and the user's sense of confidence in the measured body temperature decreases.

As described above, the predictive calculation type electronic thermometer of the prior application did not take into account abnormal fluctuations in the intermediate level of temperature data, and there was a problem.

As described above, the purpose of the present invention is to provide high performance that can sufficiently deal with practical problems such as electrical noise and movement of the sensor within the measurement area, and that can provide relatively high prediction accuracy. The present invention provides a predictive calculation type electronic thermometer.

[Means for solving problems]

According to the present invention, this object is achieved by the following configuration.

That is, it has a temperature detection section that detects the temperature of the measurement site, a calculation section that predicts and calculates the equilibrium temperature based on the detected temperature, and a display section that displays the results calculated by the calculation section, and the calculation section includes a temperature memory circuit that temporarily stores temperature data detected by the temperature detection section, a time measurement circuit that measures elapsed time from the start of temperature measurement, and temperature data stored in the temperature memory circuit; a parameter calculation unit that calculates parameters of a functional equation that determines the equilibrium temperature using elapsed time data from the time measurement circuit; and a predicted value calculation circuit that calculates the equilibrium temperature based on the data of the functional equation calculated by the parameter calculation unit. In the predictive calculation type electronic thermometer, the parameter calculation circuit includes a parameter calculation circuit that temporarily calculates the value of the parameter from the temperature data and the elapsed time data, and a parameter calculation circuit that stores the previously calculated value of the parameter. (The previous value memory circuit, the reference setting circuit that calculates the value of the previous value memory circuit and sets the reference value, the limit setting circuit that sets the limit value by similar calculation, and the parameter setting circuit from the limit setting circuit) a comparison circuit that compares a limit value with a parameter calculation value from the parameter calculation circuit and outputs the result; and a parameter calculation value from the parameter calculation circuit or the reference setting circuit in response to a signal from the comparison circuit. and a selection circuit that selects and outputs a reference value from.

〔Example〕

Hereinafter, embodiments of the predictive calculation type electronic thermometer according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a circuit block diagram specifically showing a parameter calculation section of a predictive calculation type electronic thermometer according to the present invention.

Each of the B calculation circuit 50.58, the C calculation circuit 54.60, and the A calculation circuit 52.56 serves as a parameter calculation circuit.
Parameter calculation signal 27 from calculation state instruction circuit 11.
This is a circuit that calculates the value of each parameter each time 28.29 is input according to the calculation formula shown in the explanation of the predictive calculation type electronic thermometer of the prior application. The parameter calculation circuit limits the value of this parameter, and the B limit circuit 5i 59
and C limit circuit 55.61.

The limiting method is to limit the calculated value of the parameter so that it does not exceed a predetermined reference range, and the details will be described later. B-A output circuit 5 which is a parameter output circuit
6. The C-A output circuit 57 and the B-C output circuit 62 are circuits that output parameter values from the parameter calculation circuit or parameter restriction circuit.

In addition, FIG. 1 shows a part corresponding to the parameter calculation unit 4 of FIG. Although not written, there is a temperature detection section 1, a measurement control circuit B, a time measurement circuit 8, and a temperature memory circuit 9. The specific configuration and operation of the predicted value calculation circuit 10 and the display section B are the same as those of the prior application shown in FIG.

That is, the configuration of the present invention is based on the three parameter calculation signals that constitute the parameter calculation unit 4 of the predictive electronic thermometer of the prior application shown in FIG. For the C calculation circuit 14VC, a B limit circuit 51.59 and a C limit circuit 55.59 are parameter limit circuits.
61 is added.

Here, we will explain a series of processes from parameter calculation and restriction to predicted value calculation, focusing on the specific configuration and operation of the parameter limiting circuit, using the flowchart in Figure 2 and the characteristic diagram in Figure 3. .

In FIG. 1, the B limit circuit 51 selects the initial setting value of the parameter B immediately after the prediction calculation starts, and selects the initial setting value of the parameter B after that. A selector 71 selects the calculated value and outputs it to the previous value memory circuit 70, and sets a reference value by adding and subtracting the stored value of the previous value memory circuit 70 and a set value based on a predetermined amount of change in the parameter. A reference setting circuit 72 for setting a limit value, a limit setting circuit 76 for setting a limit value, and data from the B calculation circuit 50 are compared with the limit value of the limit setting circuit 76 to determine the magnitude relationship between the two;
A comparison circuit 74 outputs the result, and a signal from the comparison circuit 74 selects the data of the B calculation circuit 50 or the reference value from the reference setting circuit 72 and outputs it to the B-A output circuit 53 and the calculation circuit. It is composed of a circuit 9o.

Further, the reference setting circuit 72 adds preset values +3 and -3 as predetermined constants to the contents of the previous value memory circuit 70 (here, adding -3 naturally means subtracting +3). It is composed of an arithmetic circuit 75, an upper reference value memory circuit 76, and a lower reference value memory circuit 77, each of which stores the results.Similarly, the limit setting circuit 76 has +5 and -5 set as preset values. It consists of an arithmetic circuit 78, an upper limit memory circuit 79 and a lower limit memory circuit 80 that store the results of addition and subtraction by the arithmetic circuit 78. Furthermore, a comparison circuit 74 compares the data of the B calculation circuit 50 with the data of the B calculation circuit 50. It consists of an upper limit comparison circuit 81 that compares the data in the upper limit memory circuit 79 and a lower limit comparison circuit 82 that compares the data in the lower limit memory circuit 80.

First of all, at present, the various conditions (1), (11), (i++
It is assumed that + has been filled in and is about to move on to the process of calculating parameters B and . First, the calculation state instruction circuit 11 instructs the B-A calculation circuit 12 to include a B reset data signal D1, which is the initial setting value of parameter B, a control signal S1, a temperature data signal D2, and so on.
Parameter calculation signal 27 that combines elapsed time data signal D3
Output.

The B calculation circuit 50 operates every time the control signal S1 is input, calculates the value of the parameter B from the temperature data signal D2 and the elapsed time data D3, and calculates the value of the parameter B from the comparison circuit 74,
A B calculation data signal D4 is output to the selection circuit 90.

The selector 71 is a circuit that selectively inputs data from the data input terminals 11 and 2 depending on the presence or absence of an input signal from the control input terminal φ, and outputs it from the output terminal Q.
Here, the data input terminal 1 is selected, and the selection circuit 90
At the same time, a signal from the previous value memory circuit 70 is inputted and outputted to the previous value memory circuit 70. Here, the signal from the selection circuit 90 is a B calculation data signal D16 which is a calculated value of parameter B, which will be described later.

When the control signal S1 is input, the previous value memory circuit 70 inputs the previous value selection data signal D5 from the selector 71 and stores it as the previous value BO, and also stores the previously stored contents in the reference setting circuit 72. and is outputted to the limit setting circuit 76 as a B memory data signal D6.

When the reference setting circuit 72 receives the B memory data signal D6, an arithmetic operation circuit 75 first performs addition and subtraction operations.

That is, the preset values 3 and -3 are added to the B memory data. The result is set as the reference calculation data signal D7, and the lower reference value memory circuit 76 receives data with the addition of 3, and the lower reference value memory circuit 77 receives data with the addition of -3, that is, subtracts 3. Output data. When the lower reference value memory circuit 76 and the lower reference value memory circuit 77 input the reference calculation data signal D7, they store them as the lower reference value Bh and lower reference value Bl, respectively, and at the same time, the selection circuit 9
8. Upper reference value data signal for 0. It is output as the lower reference value data signal D9. Similar operations are performed in the limit setting circuit 76, but addition and subtraction operations in the arithmetic circuit 78 are performed using the preset value '+5. - after being done as 5,
The upper limit value B□ and lower limit value B5, which are output as the limit calculation data signal D10 and stored in the upper value memory circuit 79 and the lower limit value memory circuit 80, are sent to the comparison circuit 74 as the upper limit value data signal D11 and the lower limit value data signal. It is output as D12.

The comparison circuit 74 compares the data of the limit setting circuit 76 and the data of the B calculation circuit.

First, the upper limit comparison circuit 81 outputs the upper limit value data signal D11.
By inputting the B calculated value data signal D4, the calculated value of B is compared with the upper limit value B□, and if B≧B□, the upper limit value signal S2 is output, and if B<B, the upper limit value signal S2 is output. No output. Further, the lower limit value comparison circuit 82 inputs the lower limit value data signal D12 and the B calculated value data signal D4, and
If L, the lower limit signal S3 is output, and if B≧B1, it is not output.

The selection circuit 90 uses the signal from the comparison circuit 74 to
Select possible values for parameter B.

First, when the upper limit signal S2 is input, the AND gate 8
7 passes the lower reference value data signal D8 and outputs it as the lower reference value selection data signal D14. At this time, the lower limit value signal S3
is not output due to the previous comparison condition, so the AND gate 88 is closed. Further, by configuring the inverter 86 and the AND gate 86 as shown in FIG. 1, the AND gate 85 also remains closed. In other words, or gate 8
Since only the lower reference value selection data signal DI'4 is inputted to 9, the B calculation data signal D1.6 outputted to the B-A output circuit 56 becomes the lower reference value Bh.

Next, when the lower limit value signal S3 is input, the AND gate 88 passes the lower reference value data signal D9 and outputs it as the lower reference value selection data signal D15, but the upper limit value signal S
Based on the same idea as when 2 is input, the AND gate 87 is closed, and the inverter 84 and the AND gate 86 are closed.
Considering the configuration, the AND gate 85 is also in a closed state, so the B calculation data signal D16 from the OR gate 89 becomes the lower reference value BA. Furthermore, if neither the upper limit value signal S2 nor the lower limit value signal S3 is input,
．． 88 are both closed, but the inverter 86
．． 84 causes the AND gate 86 to output a signal, so
The AND gate 85 passes the B calculated value data signal D4 and
Output as calculation selection data signal.

This means that the B calculation data signal D16 output from the OR gate 89 becomes the calculated value of B. The B calculation data signal D16 from the selection circuit 90 is input to the A calculation circuit 52,
The value of parameter A is calculated from the temperature data and the elapsed time data, and is output to the B-A output circuit 56 as the A calculation data input terminal together with the B calculation data input terminal. At the same time, the B calculation data D16 is selectively inputted from the data input terminal 11 of the selector 71 and outputted to the previous value memory circuit 70, where it is stored as previous value data BO. The above selection operation in the selection circuit 90 is easy to understand (Table K) To summarize, the following * indicates that the signal is not output when the signal is at L level.

Although not directly related to the present application, the calculation state indication circuit 11
Let me explain what happens when the first signal is output. As shown in the conventional example of FIG. 4, the calculation state instruction circuit 11 is connected to the measurement control circuit 7, and when the calculation instruction signal 26 is input for the first time, it outputs the parameter calculation signal 27 shown in FIG. At this time, the B reset data signal DI, which is the initial setting value of the parameter B, the temperature data signal D2. The elapsed time data signal D3 is output, but the control signal is not output. Therefore, B calculation circuit 5
0 does not operate and no signal is input from the control input terminal φ of the selector 71, so the data input terminal 2 is selected and as a result, the previous value selection data signal D5 is 131J.
The contents of the set data signal D1 will be output. That is, the previous value memory circuit 70 stores the initial setting value of parameter B. This initial operation in the B control circuit 51 is similarly performed in the C control circuit, and the initial value of parameter 0 is stored.

Now, the specific operation of the parameter calculation and restriction method configured as above will be explained with reference to the flowchart of FIG. 2 and the characteristic diagram of FIG. 3. In the characteristic diagram of FIG. 3, the horizontal axis, vertical axis, and the same characters and symbols as shown in the characteristic diagram of FIG. 7 are the same as in FIG. In other words, T (S) is a continuous graph of the actual temperature when the elapsed time S is 0.2.4..., and the value when S is 300 seconds is T (soo), In other words, it is shown as the equilibrium temperature.

In addition, U-I is the T (S) when the elapsed time S is 6 seconds.
This is a graph of a functional formula that gives a predicted value calculated for the temperature rise of , and the predicted value is shown as U(ej, ).

Here, t(8) is a model that continuously shows the temperature rise when temperature data fluctuates due to electrical noise, movement of the sensor within the measurement site, instantaneous removal, etc. during temperature measurement. For the sake of clarity, let us assume that the equilibrium temperature is given by T (300), which is the same as in the case of T(s) above.

Conditions (1), (I++, and (iIi)) for starting the calculation shown in the explanation of the predictive calculation type electronic thermometer of the earlier application have all been satisfied, and the process of calculating each parameter A, B, and C has now started. Assume that the elapsed time S is 6 seconds and the operation of calculating the values of parameters B and A is about to start.

The flowchart in FIG. 2 shows in detail the parameter B calculation step 115 in the flowchart shown in FIG. First, in calculation step 200, the value of parameter B is calculated from the 0 formula shown in the predictive calculation type electronic thermometer of the prior application. Next, the calculation circuit 7 uses the previous value BO stored in the previous value memory circuit 70 in calculation steps 201 and 202.
5, the addition and subtraction operations are performed and stored as upper reference value Bh and lower reference value Bl in upper reference value memory circuit 76 and lower reference value memory circuit 77, respectively. Continuation (Calculation step 206. In 204, the previous value memory circuit 70, the calculation circuit 78, the upper limit value memory circuit 79, and the lower limit value memory circuit 80 are similarly passed through a series of steps to calculate the upper limit value B□ and the lower limit value BL, A storage operation is performed.In the next judgment step 205', the calculated value of B in the upper limit comparison circuit 81 is compared with the upper limit value B□, and in judgment step 204, a comparison operation in the lower limit comparison circuit 82 is performed. The comparison results from the judgment steps 205 and 206 are as follows:
As described above for the selection circuit 90, a predetermined selection is made, and the upper reference value Bh is outputted in step 207, the lower reference value Bl is outputted in step 208, or the calculated value B is outputted in step 208. It will be output as is.

Incidentally, if the calculated value of B exceeds the upper limit value B□, since B>B□, the process passes through step 207 and the upper reference value Bh is output. In the next step 209, the value of parameter B output from the selection circuit 90 as the B calculation data signal D1.6 is stored in the previous value memory circuit 70 through the selector 71, and the previous value Bo for the next parameter calculation process is stored. It can be seen that Step 209 and subsequent steps continue to step 116 for calculating parameter A in FIG.
It turns out.

The functional expression for the values of parameters B and A obtained here becomes the graph of uba shown in FIG. 3, and u(6) is obtained as the predicted value at that point.

Here, the steps from step 201 to step 209 represent the present application, and if omitted, that is, step 115 would be replaced by step 2.
If only 00 is displayed, this is exactly the same as the predictive calculation type electronic thermometer of the previous application. In this case, the function equation to be found is represented by the graph shown in uOba, and the predicted value is 0.
0(a).

Comparing the above u(6) 0(6), if temporary fluctuations occur in the temperature rise process and U odor, the predicted value uo(1)
It can be seen that while deviates from the equilibrium temperature T(, . . .) by a large amount, the difference in u(I+) is small.

In the above explanation, it is also conceivable that the arithmetic circuit 78 in FIG. 1 has two limit values and two reference values.

In this case, this can be realized by removing the reference setting circuit 72 and connecting the output of the limit setting circuit 76 to the selection circuit 90. Further, if the calculated value of parameter B in the B calculation circuit 50 exceeds the upper limit or the lower limit, it is possible to cancel the current calculation by outputting the previous value as is. However, as a result of collecting data through clinical trials and the like, it has been found that statistically speaking, the configuration shown in FIG. 1 is the most effective.

[Effects of the Invention] As is clear from the above explanation, according to the present invention, the value of the parameter obtained by calculation is determined so that the amount of displacement from the previous value is within a certain range before being substituted into the function equation. Since the temperature data is limited to a certain temperature, it may not be accurate even if it is not accurate due to electrical noise, movement of the sensor within the measurement area, momentary removal of temperature data, etc. Since the influence on the predicted value can be kept to a minimum without deviating from the value, relatively high prediction accuracy can be obtained, and it is highly effective in providing a high-performance predictive calculation type electronic thermometer. there were.

[Brief explanation of drawings]

FIG. 1 is a circuit block diagram showing the specific configuration of the main parts of the predictive calculation type electronic thermometer according to the present invention, FIG. 2 is a flowchart showing the operation of the device shown in FIG. 1, and FIG. A graph t(8) representing a temperature rise model for explanation, a graph showing the corresponding functional formula and predicted values, and FIG. 4 is a circuit block line showing the specific configuration of a conventional predictive calculation type electronic thermometer. 5 is a circuit block diagram showing the basic configuration of a conventional predictive calculation type electronic thermometer, and FIG.
The figure is a flowchart showing the operation of the device shown in FIG.
Figure 7 is a graph T (S
), a graph for explaining how the corresponding predicted value changes over time and the functional formula at that time. 4...Parameter calculation unit, 50.58...B calculation circuit, 51.59...B limit circuit, 70...Previous value memory circuit, 72 ...Reference setting circuit, 76...Limit setting circuit, 74...Comparison circuit, 90...Selection circuit. Figure 2 Figure 3

Claims (1)

[Claims] It has a temperature detection section that detects the temperature of the measurement site, a calculation section that predicts and calculates the equilibrium temperature based on the detected temperature, and a display section that displays the result calculated by the calculation section, and the calculation section , the temperature data detected by the temperature detection section,
Equilibrium is established using a temperature memory circuit that temporarily stores temperature, a time measurement circuit that measures the elapsed time from the start of temperature measurement, temperature data stored in the temperature memory circuit, and elapsed time data from the time measurement circuit. A predictive calculation type electronic thermometer comprising a parameter calculation section that calculates parameters of a functional equation that determines temperature, and a predicted value calculation circuit that calculates an equilibrium temperature using data of the functional equation calculated by the parameter calculation section. The calculation circuit includes a parameter calculation circuit that temporarily calculates the value of the parameter from the temperature data and elapsed time data, a previous value memory circuit that stores the previously calculated value of the parameter, and the previous value memory circuit. A reference setting circuit that calculates a stored value and a set value based on a predetermined amount of change in the parameter to set a reference value, a limit setting circuit that sets a limit value, and a limit value of the parameter from the limit setting circuit. and a comparison circuit that compares the calculated value of the parameter from the parameter calculation circuit and outputs the result; A predictive calculation type electronic thermometer comprising a selection circuit that selects and outputs a reference value.