检测模式在波 [英] Detecting patterns in waves

问题描述

我想读一个心电图一个形象和检测主波的每一个在它（P波，QRS波群和T波）。现在我可以读取图像，并得到这样一个向量（4.2; 4.4; 4.9; 4.7; ......）在心电图的价值再presentative，什么是问题的一半。我需要一个算法，可以通过这个载体行走，并检测每本波的开始和结束。

I'm trying to read a image from a electrocardiography and detect each one of the main waves in it (P wave, QRS complex and T wave). Now I can read the image and get a vector like (4.2; 4.4; 4.9; 4.7; ...) representative of the values in the electrocardiography, what is half of the problem. I need a algorithm that can walk through this vector and detect when each of this waves start and end.

下面是它的图形之一的例子：

Here is a example of one of its graphs:

会很容易，如果他们总是有相同的大小，但它不喜欢它的工作原理，如果我知道有多少波心电图会有，但它可以改变了。有没有人有一些想法？

Would be easy if they always had the same size, but it's not like it works, or if I knew how many waves the ecg would have, but it can vary too. Does anyone have some ideas?

谢谢！

更新

什么我想要实现示例：

由于波

我可以提取矢量

[0; 0; 20; 20; 20; 19; 18; 17; 17; 17; 17; 17; 16; 16; 16; 16; 16; 16; 16; 17; 17; 18; 19; 20; 21; 22; 23; 23; 23; 25; 25; 23; 22; 20; 19; 17; 16; 16; 14; 13; 14; 13; 13; 12; 12; 12; 12; 12; 11; 11; 10; 12; 16; 22; 31; 38; 45; 51; 47; 41; 33; 26; 21; 17; 17; 16; 16; 15; 16; 17; 17; 18; 18; 17; 18; 18; 18; 18; 18; 18; 18; 17; 17; 18; 19; 18; 18; 19; 19; 19; 19; 20; 20; 19; 20; 22; 24; 24; 25; 26; 27; 28; 29; 30; 31; 31; 31; 32; 32; 32; 31; 29; 28; 26; 24; 22; 20; 20; 19; 18; 18; 17; 17; 16; 16; 15; 15; 16; 15; 15; 15; 15; 15; 15; 15; 15; 15; 14; 15; 16; 16; 16; 16; 16; 16; 16; 16; 16; 15; 16; 15; 15; 15; 16; 16; 16; 16; 16; 16; 16; 16; 15; 16; 16; 16; 16; 16; 15; 15; 15; 15; 15; 16; 16; 17; 18; 18; 19; 19; 19; 20; 21; 22; 22; 22; 22; 21; 20; 18; 17; 17; 15; 15; 14; 14; 13; 13; 14; 13; 13; 13; 12; 12; 12; 12; 13; 18; 23; 30; 38; 47; 51; 44; 39; 31; 24; 18; 16; 15; 15; 15; 15; 15; 15; 16; 16; 16; 17; 16; 16; 17; 17; 16; 17; 17; 17; 17; 18; 18; 18; 18; 19; 19; 20; 20; 20; 20; 21; 22; 22; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 32; 33; 33; 33; 32; 30; 28; 26; 24; 23; 23; 22; 20; 19; 19; 18; 17; 17; 18; 17; 18; 18; 17; 18; 17; 18; 18; 17; 17; 17; 17; 16; 17; 17; 17; 18; 18; 17; 17; 18; 18; 18; 19; 18; 18; 17; 18; 18; 17; 17; 17; 17; 17; 18; 17; 17; 18; 17; 17; 17; 17; 17; 17; 17; 18; 17; 17; 18; 18; 18; 20; 20; 21; 21; 22; 23; 24; 23; 23; 21; 21; 20; 18; 18; 17; 16; 14; 13; 13; 13; 13; 13; 13; 13; 13; 13; 12; 12; 12; 16; 19; 28; 36; 47; 51; 46; 40; 32; 24; 20; 18; 16; 16; 16; 16; 15; 16; 16; 16; 17; 17; 17; 18; 17; 17; 18; 18; 18; 18; 19; 18; 18; 19; 20; 20; 20; 20; 20; 21; 21; 22; 22; 23; 25; 26; 27; 29; 29; 30; 31; 32; 33; 33; 33; 34; 35; 35; 35; 0; 0; 0; 0]

[0; 0; 20; 20; 20; 19; 18; 17; 17; 17; 17; 17; 16; 16; 16; 16; 16; 16; 16; 17; 17; 18; 19; 20; 21; 22; 23; 23; 23; 25; 25; 23; 22; 20; 19; 17; 16; 16; 14; 13; 14; 13; 13; 12; 12; 12; 12; 12; 11; 11; 10; 12; 16; 22; 31; 38; 45; 51; 47; 41; 33; 26; 21; 17; 17; 16; 16; 15; 16; 17; 17; 18; 18; 17; 18; 18; 18; 18; 18; 18; 18; 17; 17; 18; 19; 18; 18; 19; 19; 19; 19; 20; 20; 19; 20; 22; 24; 24; 25; 26; 27; 28; 29; 30; 31; 31; 31; 32; 32; 32; 31; 29; 28; 26; 24; 22; 20; 20; 19; 18; 18; 17; 17; 16; 16; 15; 15; 16; 15; 15; 15; 15; 15; 15; 15; 15; 15; 14; 15; 16; 16; 16; 16; 16; 16; 16; 16; 16; 15; 16; 15; 15; 15; 16; 16; 16; 16; 16; 16; 16; 16; 15; 16; 16; 16; 16; 16; 15; 15; 15; 15; 15; 16; 16; 17; 18; 18; 19; 19; 19; 20; 21; 22; 22; 22; 22; 21; 20; 18; 17; 17; 15; 15; 14; 14; 13; 13; 14; 13; 13; 13; 12; 12; 12; 12; 13; 18; 23; 30; 38; 47; 51; 44; 39; 31; 24; 18; 16; 15; 15; 15; 15; 15; 15; 16; 16; 16; 17; 16; 16; 17; 17; 16; 17; 17; 17; 17; 18; 18; 18; 18; 19; 19; 20; 20; 20; 20; 21; 22; 22; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 32; 33; 33; 33; 32; 30; 28; 26; 24; 23; 23; 22; 20; 19; 19; 18; 17; 17; 18; 17; 18; 18; 17; 18; 17; 18; 18; 17; 17; 17; 17; 16; 17; 17; 17; 18; 18; 17; 17; 18; 18; 18; 19; 18; 18; 17; 18; 18; 17; 17; 17; 17; 17; 18; 17; 17; 18; 17; 17; 17; 17; 17; 17; 17; 18; 17; 17; 18; 18; 18; 20; 20; 21; 21; 22; 23; 24; 23; 23; 21; 21; 20; 18; 18; 17; 16; 14; 13; 13; 13; 13; 13; 13; 13; 13; 13; 12; 12; 12; 16; 19; 28; 36; 47; 51; 46; 40; 32; 24; 20; 18; 16; 16; 16; 16; 15; 16; 16; 16; 17; 17; 17; 18; 17; 17; 18; 18; 18; 18; 19; 18; 18; 19; 20; 20; 20; 20; 20; 21; 21; 22; 22; 23; 25; 26; 27; 29; 29; 30; 31; 32; 33; 33; 33; 34; 35; 35; 35; 0; 0; 0; 0;]

我想检测，例如：

P波[19 - 37]

P wave in [19 - 37]

在QRS波群[51 - 64]

QRS complex in [51 - 64]

等等...

推荐答案

这的的第一件事，我的会做的是看看什么是已经在那里。事实上，这个具体的问题已经被大量研究。下面是一些非常简单的方法的简要概述：<一href="http://www.physik.uni-freiburg.de/~severin/A_comparison_of_the_noise_sensitivity_of_nine_QRS_detection_Algorithms.pdf">link.

The first thing that I would do is see what is already out there. Indeed, this specific problem has already been heavily researched. Here is a brief overview of some really simple methods: link.

我必须回应另一个答案，也是如此。我做研究的信号处理和音乐信息检索。从表面上看，这个问题不会出现类似发病的检测，但问题方面是不一样的。这种类型的生物信号处理，即，检测在P，QRS和T相，的可以利用的的特定的时域特征的每个这些波形的的知识。发病检测MIR没有，真的。 （不可靠，至少。）

I must respond to another answer, as well. I do research in signal processing and music information retrieval. On the surface, this problem does appear similar to onset detection, but the problem context is not the same. This type of biological signal processing, i.e., detection of the P, QRS, and T phases, can exploit knowledge of specific time-domain characteristics of each of these waveforms. Onset detection in MIR doesn't, really. (Not reliably, at least.)

这将很好的工作QRS检测（但不一定附注开始检测）的一种方法是动态时间规整。在时域特性保持不变，DTW可以工作得非常好。下面是使用DTW对这个问题很短的IEEE文章：<一个href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.332.6782&rep=rep1&type=pdf">link.

One approach that would work well for QRS detection (but not necessarily for note onset detection) is dynamic time warping. When time-domain characteristics remain invariant, DTW can work remarkably well. Here is a short IEEE paper that uses DTW for this problem: link.

这是一个很好的IEEE杂志文章，比较了多种方法：链接。你会看到，很多常见的信号处理模式都试过了。撇去纸张，并尝试一个你在基本层面理解。

This is a nice IEEE magazine article that compares many methods: link. You'll see that many common signal processing models have been tried. Skim the paper, and try one that you understand at a basic level.

编辑：浏览这些文章后，小波为基础的方法似乎是最直观的给我。大田将工作做好，也和存在DTW模块在那里，但小波方法似乎最好的给我。别人通过利用信号的衍生物回答。我的第一个环节检查从1990年做的，以前的方法，但我怀疑他们是不是更现代的方法是稳健的。

After browsing these articles, a wavelet-based approach seems most intuitive to me. DTW will work well, too, and there exist DTW modules out there, but the wavelet approach seems best to me. Someone else answered by exploiting derivatives of the signal. My first link examines methods from before 1990 that do that, but I suspect that they are not as robust as more modern methods.

编辑：我会尽力给出一个简单的解决方案，当我得到机会，但原因的为什么的我觉得小波在这里适用的是因为他们在参数设置各种各样的形状，不管有用中的时间或幅度缩放的。换句话说，如果有一个信号具有相同的重复瞬时形状，但在不同的时间尺度和振幅，小波分析仍然可以识别这些形状为类似于（粗略地说）。还要注意的那种，我结块滤波器组这一类。类似的事情。

I'll try to give a simple solution when I get the chance, but the reason why I think wavelets are suited here are because they are useful at parameterizing a wide variety of shapes regardless of time or amplitude scaling. In other words, if you have a signal with the same repeated temporal shape but at varying time scales and amplitudes, wavelet analysis can still recognize these shapes as being similar (roughly speaking). Also note that I am sort of lumping filter banks into this category. Similar things.

这篇关于检测模式在波的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！