Summary of the invention
The present invention proposes can according to preset time value in a kind of fetal movement counting process, discontinuity detects the detection method of best fetal movement counting position, in more fetal movement sensor data acquisition, the sensing station selecting part best gathers source as the fetal movement enumeration data of next time period, to reduce fetal movement number system workload and to improve count accuracy.
Technical scheme of the present invention is:
Based on a detection method for the best fetal movement counting position of acceleration transducer, it is characterized in that it adopts 8 3-axis acceleration sensors to carry out fetal movement detection to the abdominal part of anemia of pregnant woman, the detection method of described best fetal movement counting position comprises the following steps:
The first step: be the abdominal part that two horizontally-arranged layouts are close to anemia of pregnant woman by 8 3-axis acceleration sensors, setting fetal movement interval T detection time, starting loop sensor detects fetal movement, after time T, obtain m time and detect data, and record detects time point corresponding to data each time;
Second step: detect data for the 1st time to the m time, choose 4 position P1-P4 that 8 3-axis acceleration sensors amplitude in each axle of X-axis, Y-axis and Z axis is maximum respectively;
3rd step: for an one-time detection data of aforementioned arbitrary 3-axis acceleration sensing, 4 position descendings by size in X-axis, Y-axis and Z axis, row operation of going forward side by side is as follows:
4 positions maximum for amplitude in each axle of X, Y and Z are P1-P4, and peak swing i.e. the first amplitude corresponding to each position is a1-a4; The fractional threshold parameter n1 of the second amplitude and the first amplitude is set, the fractional threshold parameter n2 of the 3rd amplitude and the second amplitude, the fractional threshold parameter n3 of the 4th amplitude and the 3rd amplitude, n1>n2>n3, S1, S2 and S3 represent the score value giving sensor respectively, S1>S2>S3;
(1) appoint one-time detection data, for arbitrary axle, select the a1 that amplitude is maximum, the s1 giving corresponding position P1 corresponding divides;
(2), for position P2 corresponding to a2, the ratio η of the second amplitude and the first amplitude is calculated
1,
if η
1meet:
η
1>=n
1, then s1 gives P2;
η
1<n
1then s2 gives P2
(3), for position P3 corresponding to a3, the ratio η of the 3rd amplitude and the second amplitude is calculated
2,
if η
2meet:
η
2>=n
2, then s2 gives P3;
η
2<n
2then s3 gives P3
(4), for position P4 corresponding to a4, the ratio η of the 4th amplitude and the 3rd amplitude is calculated
3,
if η
3meet:
η
3>=n
3, then s3 gives P4;
η
3<n
3then s4 gives P4
(5) score of all the other 4 positions counts 0
Detect data according to (1)-(5) to m time of X, Y of 8 3-axis acceleration sensors and Z axis all to process, obtain the score of the X-axis each time of each 3-axis acceleration sensor in m detection data, Y-axis, 4 amplitude maximum positions that Z axis is corresponding;
4th step: m time is detected for this and detects data, calculate respectively 8 3-axis acceleration sensors in the score of each axle of position X-axis, Y-axis and Z axis and, calculate the score summation of 8 3-axis acceleration sensors, four positions that score is higher are 4 best fetal movement counting position of fetal movement counting in next time period T.
In 8 3-axis acceleration sensors of the present invention, lower row is followed successively by position 1, position 2, position 3 and position 4 from left to right, upper row is followed successively by position 5, position 6, position 7 and position 8 from left to right, desired location 1 and position 3 are one group, namely organize A, position 2 and position 4 are one group, namely organize B, position 5 and position 7 are one group, namely organize C; Position 6 and position 8 are one group, namely organize D; After the 4th step, compare A group and B group, the score of C group and D group, the position corresponding to two groups that score is higher is 4 best fetal movement counting position of fetal movement counting in next time period T.
In 8 3-axis acceleration sensors of the present invention, Z-direction is paid the utmost attention to during selection, higher priority is set, X-axis and Y-axis arrange lower priority, namely when asking 8 position Synthesis scores, setting Z axis coefficient is higher than the coefficient of X-axis and Y-axis, and the coefficient of X, Y and Z axis, all between 0 to 1, calculates the integrate score of 8 positions.
Setting Z axis coefficient of the present invention is 0.5, and X-axis and Y-axis coefficient are 0.25, thus calculate the integrate score of 8 positions.
8 3-axis acceleration sensors of the present invention are embedded in anemia of pregnant woman's intelligence clothing or bellyband.
In the first step of the present invention: detect data for m time, the interval time of data is detected twice before and after calculating, if be less than or equal to Δ t interval time, one-time detection data and frontly once belong to fetal movement data then, using the time of the time of front one-time detection data as these fetal movement data, front and back are detected amplitude that in data, amplitude the is large amplitude as these fetal movement data for twice; Detect data for m time and carry out aforementioned processing, obtain M fetal movement enumeration data, for subsequent treatment, wherein, 3min < Δ t < 6min.
S1=1, s2=0.5, s3=0.2, s4=0 of the present invention; N1=0.8, n2=0.5, n3=0.4.
In the first step of the present invention, arranging four groups of initial optimum detection positions, is position 1, position 3, position 5 and position 7 respectively.
In the first step of the present invention, interval T is 2-8 hour.
Beneficial effect of the present invention:
The present invention proposes the detection method of best fetal movement counting position in a kind of fetal movement counting process, in more fetal movement sensor data acquisition, select the most effective sensor position of part as the collection source of fetal movement enumeration data, catch Fetal Movement Signal to greatest extent, optimize Fetal Movement Signal image data source, alleviate workload of operation and the date processing difficulty of fetal movement number system software, decrease the data space of system simultaneously, show that fetal movement calculates real-time and accurately, and certain reduction system soft and hardware cost of manufacture; The present invention according to the individual variation of anemia of pregnant woman's fetal movement change, can also be arranged different test interval section T and selects, and strengthens or reduce the frequency of optimum position adjustment, to improve the accuracy that optimum position is detected further.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is further illustrated.
As shown in Figure 1, in order to obtain more comprehensively Fetal Movement Signal accurately, the present invention introduces 83 axle acceleration sensors, be embedded in anemia of pregnant woman intelligence clothing or bellyband in, be close to the abdominal part of anemia of pregnant woman, respectively these 8 positions are defined as position 1,2 ... 8.Optimum position trace routine is started, as the detection position of fetal movement counting in next T time at interval of after time T.
When each optimum position is detected, sensor can detect that several times fetal movement changes, and counts m time.M fetal movement change-detection data, the interval time of data is detected twice before and after calculating, if be less than or equal to Δ t interval time, one-time detection data and frontly once belong to fetal movement data then, using the time of the time of front one-time detection data as these fetal movement data, front and back are detected amplitude that in data, amplitude the is large amplitude as these fetal movement data for twice; Detect data for m time and carry out aforementioned processing, obtain M fetal movement enumeration data, for subsequent treatment, wherein, 3min < Δ t < 6min.
As the present embodiment, start optimum position trace routine apart from last fetal movement counting after T hour.Because of experiment condition restriction, in embodiment, only collect the first time fetal movement counting behind anemia of pregnant woman's T time interval, here suppose that T interval time is shorter and be in the fetus quiet period, therefore fetal movement counting only there occurs once in T time, detect data basis using this fetal movement delta data as the optimum position of next T time section.
Concrete steps are as follows:
The first step: the detection data gathering 83 axle acceleration sensors in the M time fetal movement process after preset time interval T, there occurs fetal movement counting altogether, and this fetal movement sensor for countering detect that fetal movement changes m time.Digital coding+X, Y, Z axis is used to represent some data received that are coupling of certain position.Such as 3x represents that the data that the X-axis of position 3 receives, 4Y represent the data that the Y-axis of position 4 receives.The data obtained as shown in Table 2-4.
Second step: this fetal movement there occurs 14 fetal movement changes and sensor obtains 14 detections data, i.e. m1=14.The time point of the change of fetal movement each time correspondence also can go on record.Choose 4 positions that in the X-axis in the 1st time to the m time fetal movement each time change corresponding to 83 axle sensor positions, Y-axis, Z axis, amplitude is maximum respectively.As shown in Table 2-4, the 1st fetal movement changes, and 4 positions that amplitude is maximum in X-axis are 1,2,3,6; 4 positions that in Y-axis, amplitude is maximum are 2,3,5,7; 4 positions that on Z axis, amplitude is maximum are 2,3,4,5.
3rd step: by each time continuously in fetal movement change procedure respectively by 4 maximum to X-axis, Y-axis, Z axis amplitude position descendings by size, and carry out score computing.Computing is as follows:
Position p1, p2, p3, p4 are arranged in these 4 positions, and peak swing corresponding to each position is a1, a2, a3, a4.Threshold parameter n1, n2, n3 are set, obtain the score of these 4 positions according to these threshold parameters.
1) select the a1 that amplitude is maximum, the s1 giving position p1 corresponding divides
2) position p2 corresponding to a2 is investigated, if η (η=a2/a1) meets
η >=n1 then s1 gives p2
η <n1 then s2 gives p2
3) position p3 corresponding to a3 is investigated, if η (η=a3/a2) meets
η >=n2 then s2 gives p3
η <n2 then s3 gives p3
4) position p4 corresponding to a4 is investigated, if η (η=a4/a3) meets
η >=n3 then s3 gives p4
η <n3 then s4 gives p4
5) score of all the other 4 positions counts 0
The score detecting fetal movement X-axis each time in data, Y-axis, 4 amplitude maximum positions that Z axis is corresponding for m time can be calculated thus.
As, parameters n1=0.8, n2=0.5, n3=0.4; S1=1, s2=0.5, s3=0.2, s4=0.4 positions that in 6th continuous fetal movement change, the amplitude of Z axis is maximum are 2,3,7,8 respectively, and the amplitude of its correspondence is 0.025g, 0.08g, 0.01g, 0.03g.Be adjusted to 3,8,2,7 according to descending, to be set to the position of answering be the amplitude of its correspondence of p1, p2, p3, p4 is a1, a2, a3, a4.
1) p1 response is the strongest, therefore p1=s1=1
2) p2 is investigated, because η=a2/a1=0.03/0.08=0.375<n1 (0.8), so p2=s2=0.5
3) p3 is investigated, because η=a3/a2=0.025/0.03=0.833>n2 (0.5), so p3=s2=0.5
4) p4 is investigated, because η=a4/a3=0.01/0.025=0.4>=n3 (0.4), so p4=s3=0.2
Therefore the grade form 1 of Z-direction eight positions in the 6th continuous fetal movement change can be obtained:
Position |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
Scores |
0 |
0.5 |
1 |
0 |
0 |
0 |
0.2 |
0.5 |
The scoring of Z-direction eight positions in table 1 the 6th continuous fetal movement change
Process the score of 14 continuous fetal movement change X, Y, Z axis data that this fetal movement collects successively, obtain 8 3-axis acceleration sensor positions in the score in X, Y, Z tri-directions, count table 2, table 3, table 4.
The left side of table represents the sequence number of continuous fetal movement change frequency, such as FetMov1 represents the fetal movement delta data that this fetal movement sensor for countering detects, the distance initial time time in moment that fetal movement occurs is shown in Time list, as, the time that 1st fetal movement change distance program triggers is 5 seconds, and the time that the 14th fetal movement change distance program triggers is 261 seconds.SelectedTimes represent to detect in data for m time choose be one of 4 amplitudes the strongest number of times.Scores represents the score accumulation of each position.
In this fetal movement process of table 2, fetal movement changes X-axis peak swing position score continuously each time
In this fetal movement process of table 3, fetal movement changes Y-axis peak swing position score continuously each time
In this fetal movement process of table 4, fetal movement changes Z axis peak swing position score continuously each time
4th step: this fetal movement process accumulative m continuous fetal movement changes, the score summation of 8 3-axis acceleration sensor position X-axis, Y-axis, Z axis.As shown in table 4.
Position |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
X |
7.6 |
5.8 |
5 |
5.3 |
1.7 |
1.4 |
2 |
1 |
Y |
4.1 |
7 |
8.4 |
0 |
4.3 |
1.6 |
3.1 |
1.7 |
Z |
4.4 |
10.2 |
10.2 |
2.7 |
1.6 |
0.2 |
0.8 |
1 |
The score of each position in X, Y, Z axis in this fetal movement process of table 4
5th step: consider the accuracy of test and comprehensive, when selecting in position, desired location 1 and position 3 are one group, namely organize A; Position 2 and position 4 are one group, namely organize B; In two groups, select one group to catch the action of fetus below her abdominal; In like manner, position 5 and position 7 are one group, namely organize C; Position 6 and position 8 are one group, namely organize D.One group is selected to catch the action of fetus above her abdominal.Pay the utmost attention to Z-direction when group and the selection principle organized, arrange higher priority, X-axis and Y-axis arrange lower priority.Therefore when asking 8 position Synthesis scores, can suppose that Z axis coefficient is 0.5, X-axis and Y-axis coefficient are 0.25.Thus calculate the integrate score of 8 positions.
As, the final score of position 1 is:
p1=1Z*0.5+1X*0.25+1Y*0.25=4.4*0.5+4.1*0.25+7.6*0.25=5.125;
In like manner, obtain finally obtaining of other 7 positions, as shown in table 5.
Position |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
Scores |
5.125 |
8.3 |
8.45 |
2.675 |
5.125 |
0.85 |
1.675 |
1.175 |
The final score of each position in this fetal movement process of table 5
6th step: compare A group and B group, the score of C group and D group, position corresponding in higher two groups of score is 4 optimum detection positions of next T time section fetal movement counting.
As: A group score: 5.125+8.45=13.575; In like manner obtain B, C, D group score and be respectively 10.975,6.8,2.025.The response intensity of A group and C group is greater than its corresponding group.
Therefore 4 optimum detection positions of next T time section fetal movement counting are 1,3,5,7.
The part that the present invention does not relate to prior art that maybe can adopt all same as the prior art is realized.