CN102323977B - Electrocardio data storage method based on electrocardio characteristic points - Google Patents
Electrocardio data storage method based on electrocardio characteristic points Download PDFInfo
- Publication number
- CN102323977B CN102323977B CN201110178695.9A CN201110178695A CN102323977B CN 102323977 B CN102323977 B CN 102323977B CN 201110178695 A CN201110178695 A CN 201110178695A CN 102323977 B CN102323977 B CN 102323977B
- Authority
- CN
- China
- Prior art keywords
- point
- ripple
- value
- dictionary
- ecg
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention relates to an electrocardio data compression storage method based on electrocardio characteristic points. Electrocardio waveforms are predicted by adopting a prediction function and errors of the electrocardio waveforms are modified and stored by adopting a dictionary. The electrocardio data compression storage method comprises the steps of: firstly, constructing the prediction function according to electrocardio physiological properties and constructing a dictionary base according to precision requirements; secondly, figuring out a coefficient in the prediction function according to specific characteristic point information of each hearth beat, predicting a voltage value of each sampling point through the function, storing an error between the prediction value and the actual value in a dictionary manner, i.e., searching an optimally-matched value from the dictionary for presenting the error, and using a serial number of the error in the dictionary as a storage value of the prediction error; and finally, coding and compressing the storage value, and storing other useable information. Because the processing in the process is limited in a smaller error range, the distortion rate of signals is lower, analysis is not influenced by restored signals, and coding storage is adopted by adopting the dictionary method so that compression efficiency is greatly increased.
Description
Technical field
The present invention relates to a kind of electrocardiogram (ECG) data storage means based on ecg characteristics point, belong to medical information processing field.
Background technology
Heart disease is just in the serious threat mankind's health, cardiogram (ECG) is to analyze and the of paramount importance information source of Diagnosing Cardiac disease, and cardiogram comprises huge data volume, in order effectively to carry out the remote transmission of structure, systematic searching and electrocardiogram (ECG) data of ecg database, the compress technique of electrocardiogram (ECG) data seems more and more important.
Current compression method is broadly divided in time domain and directly compresses, adopts certain transform domain and build characteristic of division template three kinds of methods.Wherein construction feature template method takes full advantage of the feature that electrocardiosignal is a quasi-periodic signal, and electrocardiosignal is analyzed to rear modeling classification, possesses good performance, has become the emphasis of ECG data compression research in the last few years.
Summary of the invention
The object of the invention is the compression storage problem in order to solve electrocardiogram (ECG) data, propose one and predict electrocardiogram (ECG) data by ecg characteristics point, to predicting the outcome, compare with actual result, then by the also electrocardiogram (ECG) data storage means of compression storage of dictionary library correction.
According to technical scheme provided by the invention, the described electrocardiogram (ECG) data storage means based on ecg characteristics point, by cardiac electrical unique point, carry out structure forecast function ecg information is predicted, construct a correction dictionary library error of predicted value and actual value is revised and stored; Described anticipation function is as follows:
For a given cardiac cycle, its each sampled point is followed successively by [(t
1, v
1), (t
2, v
2) .. (t
n, v
n)]; t
irepresent the sampling time of this point, v
irepresent the sampled voltage of this point, n is heartbeat sampling length, i=1, and 2 ..., n, the unique point wherein comprising is called after respectively: P ripple starting point [T
0, V
0], P crest value point [T
1, V
1], P ripple terminal [T
2, V
2], QRS ripple starting point [T
3, V
3], Q point [T
4, V
4], R point [T
5, V
5], S point [T
6, V
6], QRS ripple terminal [T
7, V
7], T ripple starting point [T
8, V
8], T crest value point [T
9, V
9], T ripple terminal [T
10, V
10], heartbeat terminal [T
11, V
11];
Anticipation function P:P[t]=P
base[t]+P
seg[t];--------formula A
Wherein, sampling time t=t
1, t
2..., t
n, P[t] and be t point prediction value, P
base[t] is baseline forecast function, P
seg[t] is signature waveform anticipation function;
P
base[t]＝A+k*t t∈[T
0，T
11]....
0
Described correction dictionary library building method is
D[x]=D
0+ k*exp (x*x/2)-1-------formula B
Wherein x=[0,1,2 ..., n-1], n is the dictionary length that needs structure, D
0for the minimum predicated error allowing, the maximum error of if desired revising is I
max:
k＝(I
max+1-D
0)/(exp((n-1)*(n-1)/2))；
Now can construct dictionary D:[D
0, D
1..., D
n-1].
The described electrocardiogram (ECG) data storage means based on ecg characteristics point, calculates the coefficient in anticipation function according to the unique point of each heartbeat.By each sampled point of a heartbeat is carried out to progressively iteration with anticipation function, obtain predicted data value.
For each heartbeat, by anticipation function, calculate successively the error between predicted value and actual sample value, from dictionary library, find the value of mating the most with error, using its position as storing value, and these values are stored together with anticipation function coefficient, unique point, the compression of heartbeat numbering.
As do not found the value matching in dictionary, now do not adopt dictionary position to represent, and adopt actual difference to store, and give mark, to guarantee the complete recovery to this abnormal data.
Be specially:
Coefficient A, k in formula 0 are by unique point [T
0, V
0], [T
2, V
2], [T
3, V
3], [T
10, V
10], [T
11, V
11] by least square line matching, obtain;
Formula 1,2 is the prediction of P ripple, wherein
B
1＝V
0-P
base[T
0]
A
1＝V
1-V
0，A
2＝V
1-V
2
k
1＝0.5*π/(T
1-T
0)，k
2＝0.5*π/(T
2-T
1)；
When not there is not P ripple, make T
2=T
1=T
0, V
2=V
1=V
0, k
1=k
2=0;
Formula 4 is Q ripple anticipation function, when not there is not Q ripple, and T
3=T
4, V
3=V
4; Its coefficient
B
3＝V
3-P
base[T
3]，k
3＝(V
4-V
3)/(T
4-T
3)；
When not there is not Q ripple, T
3=T
4, V
3=V
4, k
3=0;
Formula 5,6 is main ripple anticipation function in QRS wave group, wherein
B
4＝V
4-P
base[T
4]
A
4＝V
5-V
4，A
5＝V
5-V
6
k
4＝0.784/(T
5-T
4)，k
5＝0.784/(T
6-T
5)；
Formula 7 is S ripple anticipation function, when not there is not S ripple, and T
6=T
7, V
6=V
7; Its coefficient
B
6＝V
6-P
base[T
6]，k
6＝(V
7-V
6)/(T
7-T
6)；
When not there is not S ripple, T
6=T
7, V
6=V
7, k
6=0;
Formula 8 is that QRS terminal is to T ripple starting point anticipation function, its coefficient
B
7＝V
7-P
base[T
7]，k
7＝(V
8-V
7)/(T
8-T
7)；
Formula 9 is T ripple anticipation function, and this function is by three unique point [T of T ripple
8, V
8], [T
9, V
9], [T
10, V
10], wherein [T
9, V
9] for its extreme point carrys out matching, form, can try to achieve coefficient A
8, A
9, A
10, A
11if while there is not T ripple, make T
8=T
9=T
10=T
11, V
8=V
9=V
10=V
11.
Advantage of the present invention is: according to each heartbeat unique point, build template and store electrocardiogram (ECG) data and can eliminate the impact of difference between different heartbeats, cardiac electrical unique point can well reflect electrocardio basic trend, again in conjunction with cardiac electrical physilogical characteristics can make predicted value and actual value difference very little, thereby making by the revised value of dictionary library is a sparse signal that comprises a large amount of 0 values, and the probability that less value occurs is higher, adopts so existing coding techniques can very effectively carry out the compression of data.On the fidelity of data, the first complete of paramount importance data-unique point of electrocardiosignal of having preserved, secondly adopts certain error to limit modified value and makes the ecg wave form distortion rate of recovery very low, not impact analysis judgement.
Accompanying drawing explanation
Fig. 1 is implementing procedure figure of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described.
In the storage of electrocardiogram (ECG) data, the template construction method having adopted can improve compression efficiency effectively.Along with the development of Computer signal detection technique, feature point detection accuracy rate is very high.Consider the characteristic of electrocardiogram (ECG) data, also roughly the same in its form under identical physiological condition, if adopt unique point to carry out template prediction to each heartbeat, can eliminate the difference between different heartbeats, simultaneously again can be by the error limitation of prediction in less scope.By revise storage to predicting the outcome, utilize dictionary to carry out compression coding, can recover preferably data, reach the object of efficient storage.
The technical solution used in the present invention comprises the steps:
(1) according to cardiac electrical physilogical characteristics, construct the anticipation function that can adapt to various decentraction electrical features, and a dictionary library for predicted value is revised;
Anticipation function building method is as follows: for a given cardiac cycle, its each sampled point is followed successively by: [(t
1, v
1), (t
2, v
2) .. (t
n, v
n)] (t
irepresent the sampling time of this point, v
irepresent the sampled voltage of this point, n is heartbeat sampling length, i=1, and 2 ..., n), the unique point wherein comprising is called after respectively: P ripple starting point [T
0, V
0] (being also pacemaker), P crest value point [T
1, V
1], P ripple terminal [T
2, V
2], QRS ripple starting point [T
3, V
3], Q point [T
4, V
4], R point [T
5, V
5], S point [T
6, V
6], QRS ripple terminal [T
7, V
7], T ripple starting point [T
8, V
8], T crest value point [T
9, V
9], T ripple terminal [T
10, V
10], heartbeat terminal [T
11, V
11].
Anticipation function P:P[t]=P
base[t]+P
seg[t];--------formula A
Wherein, sampling time t=t
1, t
2..., t
n, P[t] and be t point prediction value, P
base[t] is baseline forecast function, P
seg[t] is signature waveform anticipation function;
P
base[t]＝A+k*t t∈[T
0，T
11]....
0
1. coefficient A, the k in formula 0 is by unique point [T
0, V
0], [T
2, V
2], [T
3, V
3], [T
10, V
10], [T
11, V
11] by least square line matching, obtain.
2. formula 1,2 is the prediction of P ripple, wherein
B
1＝V
0-P
base[T
0]
A
1＝V
1-V
0，A
2＝V
1-V
2
k
1＝0.5*π/(T
1-T
0)，k
2＝0.5*π/(T
2-T
1)
When not there is not P ripple, make T
2=T
1=T
0, V
2=V
1=V
0, k
1=k
2=0;
3. formula 4 is Q ripple anticipation function, its coefficient:
B
3＝V
3-P
base[T
3]，k
3＝(V
4-V
3)/(T
4-T
3)
When not there is not Q ripple, T
3=T
4, V
3=V
4, k
3=0;
4. formula 5,6 is main ripple anticipation function in QRS wave group, wherein:
B
4＝V
4-P
base[T
4]
A
4＝V
5-V
4，A
5＝V
5-V
6
k
4＝0.784/(T
5-T
4)，k
5＝0.784/(T
6-T
5)
5. formula 7 is S ripple anticipation function, its coefficient:
B
6＝V
6-P
base[T
6]，k
6＝(V
7-V
6)/(T
7-T
6)
When not there is not S ripple, T
6=T
7, V
6=V
7, k
6=0;
6. formula 8 be QRS terminal to T ripple starting point anticipation function, its coefficient:
B
7＝V
7-P
base[T
7]，k
7＝(V
8-V
7)/(T
8-T
7)
7. formula 9 is T ripple anticipation function, and this function is by three unique point [T of T ripple
8, V
8], [T
9, V
9], [T
10, V
10], wherein [T
9, V
9] for its extreme point carrys out matching, form, can try to achieve coefficient A
8, A
9, A
10, A
11if while there is not T ripple, can make T
8=T
9=T
10=T
11, V
8=V
9=V
10=V
11;
Revising dictionary library building method is:
D[x]=D
0+ k*exp (x*x/2)-1-------formula B
Wherein x=[0,1,2 ..., n-1], n is the dictionary length that needs structure, D
0for the minimum predicated error allowing, the maximum error of if desired revising is I
max:
k＝(I
max+1-D
0)/(exp((n-1)*(n-1)/2))
Now can construct dictionary D:[D
0, D
1..., D
n-1].
(2) according to the unique point of each heartbeat, calculate the coefficient in anticipation function;
(3) by each sampled point of a heartbeat is carried out to progressively iteration with anticipation function, obtain predicted data value;
(4) calculate the difference of predicted data and real data, look for a value of mating most with this absolute difference revising in dictionary, and write down the position of this value in dictionary, with this position and symbol, be used as forecast value revision storing value;
(5) repeating step (3), (4) be until complete the prediction of a heartbeat, and compress storage to revising storing value, preserves heartbeat numbering, unique point, anticipation function and coefficient thereof simultaneously;
(6) repeating step (2)～(5) complete the storage of whole electrocardiogram (ECG) data.
In above-mentioned steps (4) as do not find the value matching in dictionary, to run into the situations such as interference that appearance is larger or pulse pacing signal, now do not adopt dictionary position to represent, and adopt actual difference to store, and give mark, to guarantee the complete recovery to this abnormal data.
As shown in Figure 1, ecg information storage means of the present invention specific embodiment is as follows.
1. step 100: first structure forecast function and dictionary library;
(1). selecting formula A is anticipation function;
(2). use formula B to construct dictionary library, define according to actual needs the length of dictionary, maximum modified error, minimum permissible error also constructs dictionary D:[D
0, D
1..., D
n-1];
2. step 110: gather one section of continuous electrocardiosignal and carry out identification and the extraction of unique point, according to each QRS wave group, define heartbeat, take the P ripple starting point of first appearance before QRS wave group as pacemaker, after QRS wave group, before first P ripple starting point, be some heartbeat terminal (if being P ripple position depending on 0.2s before QRS starting point without P ripple), on this basis electrocardiosignal be divided into continuous heartbeat and be numbered;
3. step 120: select a heartbeat to start to compress processing;
4. step 130: calculate the coefficient in anticipation function according to the type of the heartbeat of selecting and its unique point;
5. step 140: to each sampled point (t in this heartbeat
i, v
i) carry out successively prediction and calculation;
(1). step 150: calculate predicated error S according to anticipation function
i=v
i-P[t
i];
(2). according to S
icalculate storing value I
i:
| S
i| < D
0: remember I
i=0.
| S
i| >=D
0: from dictionary D, look for one with | the value D that Si| mates most
x, x is D
xposition number in dictionary, note I
i=x; If do not find the value of coupling in dictionary, represent to run into the situations such as high frequency interference or pacemaker impulse, now use I
i=S
irepresent, to guarantee the complete recovery of this abnormal signal.
Step 160: by I
iadd symbol and classification mark, as the storing value of i sampled point;
(3). process next sampled point i+1 until complete the prediction and calculation of this heartbeat;
6. step 170: preserve the numbering of this heartbeat, unique point, anticipation function and correction storing value, and carry out compression coding to revising storing value;
7. repeat above-mentioned steps 120～step 170, complete prediction and the storage of next heartbeat, until all signal is finished dealing with.
Claims (5)
1. the electrocardiogram (ECG) data storage means based on ecg characteristics point, is characterized in that, carrys out structure forecast function ecg information is predicted by cardiac electrical unique point, constructs a correction dictionary library error of predicted value and actual value is revised and stored; Described anticipation function is as follows:
For a given cardiac cycle, its each sampled point is followed successively by [(t
1, v
1), (t
2, v
2) .. (t
n, v
n)]; t
irepresent the sampling time of this point, v
irepresent the sampled voltage of this point, n is heartbeat sampling length, i=1, and 2 ..., n, the unique point wherein comprising is called after respectively: P ripple starting point [T
0, V
0], P crest value point [T
1, V
1], P ripple terminal [T
2, V
2], QRS ripple starting point [T
3, V
3], Q point [T
4, V
4], R point [T
5, V
5], S point [T
6, V
6], QRS ripple terminal [T
7, V
7], T ripple starting point [T
8, V
8], T crest value point [T
9, V
9], T ripple terminal [T
10, V
10], heartbeat terminal [T
11, V
11];
Anticipation function P:P[t]=P
base[t]+P
seg[t];--------formula A
Wherein, sampling time t=t
1, t
2..., t
n, P[t] and be t point prediction value, P
base[t] is baseline forecast function, P
seg[t] is signature waveform anticipation function;
P
base[t]=A+k*t t ∈ [T
0, T
11] formula 0
Described correction dictionary library building method is
D[x]=D
0+ k*exp (x*x/2)-1-------formula B
Wherein x=[0,1,2 ..., n-1], n is the dictionary length that needs structure, D
0for the minimum predicated error allowing, the maximum error of if desired revising is I
max:
k=(I
max+1-D
0)/(exp((n-1)*(n-1)/2))；
Now can construct dictionary D:[D
0, D
1..., D
n-1];
Coefficient A, k in formula 0 are by unique point [T
0, V
0], [T
2, V
2], [T
3, V
3], [T
10, V
10], [T
11, V
11] by least square line matching, obtain;
Formula 1,2 is the prediction of P ripple, wherein
B
1=V
0-P
base[T
0]
A
1=V
1-V
0,A
2=V
1-V
2
k
1=0.5*π/(T
1-T
0),k
2=0.5*π/(T
2-T
1)；
When not there is not P ripple, make T
2=T
1=T
0, V
2=V
1=V
0, k
1=k
2=0;
Formula 4 is Q ripple anticipation function, when not there is not Q ripple, and T
3=T
4, V
3=V
4; Its coefficient
B
3=V
3-P
base[T
3],k
3=(V
4-V
3)/(T
4-T
3)；
When not there is not Q ripple, T
3=T
4, V
3=V
4, k
3=0;
Formula 5,6 is main ripple anticipation function in QRS wave group, wherein
B
4=V
4-P
base[T
4]
A
4=V
5-V
4,A
5=V
5-V
6
k
4=0.784/(T
5-T
4),k
5=0.784/(T
6-T
5)；
Formula 7 is S ripple anticipation function, when not there is not S ripple, and T
6=T
7, V
6=V
7; Its coefficient
B
6=V
6-P
base[T
6],k
6=(V
7-V
6)/(T
7-T
6)；
When not there is not S ripple, T
6=T
7, V
6=V
7, k
6=0;
Formula 8 is that QRS terminal is to T ripple starting point anticipation function, its coefficient
B
7=V
7-P
base[T
7],k
7=(V
8-V
7)/(T
8-T
7)；
Formula 9 is T ripple anticipation function, and this function is by three unique point [T of T ripple
8, V
8], [T
9, V
9], [T
10, V
10], wherein [T
9, V
9] for its extreme point carrys out matching, form, can try to achieve coefficient A
8, A
9, A
10, A
11if while there is not T ripple, make T
8=T
9=T
10=T
11, V
8=V
9=V
10=V
11.
2. the electrocardiogram (ECG) data storage means based on ecg characteristics point as claimed in claim 1, is characterized in that calculating the coefficient in anticipation function according to the unique point of each heartbeat.
3. the electrocardiogram (ECG) data storage means based on ecg characteristics point as claimed in claim 2, is characterized in that carrying out progressively iteration by the each sampled point anticipation function to a heartbeat, obtains predicted data value.
4. the electrocardiogram (ECG) data storage means based on ecg characteristics point as claimed in claim 3, it is characterized in that by anticipation function, calculating successively the error between predicted value and actual sample value for each heartbeat, from dictionary library, find the value of mating the most with error, using its position as storing value, and these values are stored together with anticipation function coefficient, unique point, the compression of heartbeat numbering.
5. the electrocardiogram (ECG) data storage means based on ecg characteristics point as claimed in claim 4, it is characterized in that, as do not found the value matching in dictionary, now not adopting dictionary position to represent, and adopting actual difference to store, and give mark, to guarantee the complete recovery to this abnormal data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110178695.9A CN102323977B (en) | 2011-06-29 | 2011-06-29 | Electrocardio data storage method based on electrocardio characteristic points |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110178695.9A CN102323977B (en) | 2011-06-29 | 2011-06-29 | Electrocardio data storage method based on electrocardio characteristic points |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102323977A CN102323977A (en) | 2012-01-18 |
| CN102323977B true CN102323977B (en) | 2014-04-30 |
Family
ID=45451719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110178695.9A Active CN102323977B (en) | 2011-06-29 | 2011-06-29 | Electrocardio data storage method based on electrocardio characteristic points |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102323977B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9301703B2 (en) * | 2012-02-08 | 2016-04-05 | Kyushu Institute Of Technology | Biological information processing device, biological information processing system, biological information compression method, and biological information compression processing program |
| EP3056138B1 (en) | 2015-02-11 | 2020-12-16 | Samsung Electronics Co., Ltd. | Electrocardiogram (ecg)-based authentication apparatus and method thereof, and training apparatus and method thereof for ecg-based authentication |
| CN105989266B (en) * | 2015-02-11 | 2020-04-03 | 北京三星通信技术研究有限公司 | Authentication method, device and system based on electrocardiosignals |
| WO2017059569A1 (en) * | 2015-10-08 | 2017-04-13 | 深圳迈瑞生物医疗电子股份有限公司 | Pacing signal processing method, system and electrocardiogram monitor |
| KR101930337B1 (en) * | 2015-12-07 | 2018-12-19 | 삼성전자 주식회사 | Electronic apparatus and the control metho d thereof |
| CN110420022B (en) * | 2019-07-29 | 2020-12-11 | 浙江大学 | P wave detection method based on dual-density wavelet transform |
| CN110784288A (en) * | 2019-11-04 | 2020-02-11 | 浙江大学 | Real-time lossless compression method and system for electrocardiosignals |
| CN113143284B (en) * | 2021-04-13 | 2022-10-21 | 浙江大学 | Electrocardiosignal compression method based on wavelet transformation and dual-mode prediction |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1251449A (en) * | 1998-10-18 | 2000-04-26 | 华强 | Combined use with reference of two category dictionary compress algorithm in data compaction |
| CN1612252A (en) * | 2003-10-31 | 2005-05-04 | 浙江中控技术股份有限公司 | Real-time data on-line compression and decompression method |
| CN1786939A (en) * | 2005-11-10 | 2006-06-14 | 浙江中控技术有限公司 | Real-time data compression method |
| CN101669819A (en) * | 2009-09-25 | 2010-03-17 | 西安电子科技大学 | Electrocardiogram signal lossless compression method based on PT conversion and linear prediction combination |
-
2011
- 2011-06-29 CN CN201110178695.9A patent/CN102323977B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1251449A (en) * | 1998-10-18 | 2000-04-26 | 华强 | Combined use with reference of two category dictionary compress algorithm in data compaction |
| CN1612252A (en) * | 2003-10-31 | 2005-05-04 | 浙江中控技术股份有限公司 | Real-time data on-line compression and decompression method |
| CN1786939A (en) * | 2005-11-10 | 2006-06-14 | 浙江中控技术有限公司 | Real-time data compression method |
| CN101669819A (en) * | 2009-09-25 | 2010-03-17 | 西安电子科技大学 | Electrocardiogram signal lossless compression method based on PT conversion and linear prediction combination |
Non-Patent Citations (8)
| Title |
|---|
| EEG无损压缩技术的研究;彭松;《上海生物医学工程》;20001231;第21卷(第2期);全文 * |
| 彭松.EEG无损压缩技术的研究.《上海生物医学工程》.2000,第21卷(第2期), |
| 心电图(ECG)数据压缩方法综述;高永丽 等;《楚雄师范学院学报》;20061231;全文 * |
| 心电远程监护系统的数据库系统设计与数据压缩算法研究;阴玺;《中国优秀硕士学位论文全文数据库 信息科技辑》;20071115;全文 * |
| 游晓明,等.数据压缩算法分析与改进.《小型微型计算机系统》.1999,第20卷(第8期), * |
| 王文成，等.图像无损压缩的预测编码及量化误差处理.《光电子·激光》.2004,第15卷(第5期), * |
| 阴玺.心电远程监护系统的数据库系统设计与数据压缩算法研究.《中国优秀硕士学位论文全文数据库 信息科技辑》.2007, |
| 高永丽 等.心电图(ECG)数据压缩方法综述.《楚雄师范学院学报》.2006, |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102323977A (en) | 2012-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102323977B (en) | Electrocardio data storage method based on electrocardio characteristic points | |
| CN109171712B (en) | Atrial fibrillation identification method, atrial fibrillation identification device, atrial fibrillation identification equipment and computer readable storage medium | |
| CN108714026B (en) | Fine-grained electrocardiosignal classification method based on deep convolutional neural network and online decision fusion | |
| Karpagachelvi et al. | ECG feature extraction techniques-a survey approach | |
| CN100457023C (en) | Pulse signal processing method based on lifting ripple | |
| CN102688032A (en) | Electrocardiogram signal sparse decompression and compression system based on concise common dictionary base | |
| CN111407243B (en) | Pulse signal pressure identification method based on deep learning | |
| CN102247129B (en) | Method for identifying untypical wave crests and wave troughs of pulse wave | |
| Zifan et al. | Automated ECG segmentation using piecewise derivative dynamic time warping | |
| CN110313894A (en) | Arrhythmia cordis sorting algorithm based on convolutional neural networks | |
| CN115153581A (en) | Electrocardiosignal characteristic waveform detection method and device and storage medium | |
| CN114582503A (en) | Blood pressure monitoring method based on multiple linear regression and correlation weighting selection algorithm | |
| Feng et al. | Dcmn: Double core memory network for patient outcome prediction with multimodal data | |
| Kundu et al. | Electrocardiogram synthesis using Gaussian and fourier models | |
| Wang et al. | Automatic diagnosis of ECG disease based on intelligent simulation modeling | |
| CN115281676B (en) | Fatigue detection method based on GRU neural network and ECG signal | |
| Li et al. | A novel abnormal ECG beats detection method | |
| CN111481193A (en) | Fall risk assessment and early warning method and system | |
| Chia et al. | Computationally generated cardiac biomarkers: Heart rate patterns to predict death following coronary attacks | |
| de Medeiros et al. | Heart arrhythmia classification using the PPM algorithm | |
| Wu et al. | The sparse decomposition and compression of ECG and EEG based on matching pursuits | |
| Huang et al. | ECG signal compression for low-power sensor nodes using sparse frequency spectrum features | |
| Fan et al. | Prediction of acute hypotensive episodes using random forest based on genetic programming | |
| CN111803062A (en) | Atrial fibrillation event detection method based on deep learning | |
| Fraser et al. | Time-delay lifts for physiological signal exploration: An application to ECG analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 214072 west side of three floor four, 100 Chui Cui Road (creative garden), Liyuan Development Zone, Binhu District, Wuxi, Jiangsu, China Patentee after: Jiangsu lucky Technology Co., Ltd. Address before: 214072 3 floor, 4 building, 100 Chui Cui Road (creative garden), Binhu District, Wuxi, Jiangsu. Patentee before: Wuxi Jitian Communication Technology Co., Ltd. |
|
| CP03 | Change of name, title or address |